For much of the twentieth century, parents of bilingual children in Europe and North America were advised, often firmly, to choose one language and stick to it. Speaking two languages to a young child, the prevailing wisdom held, would confuse the mind, delay speech, and ultimately weaken the command of both. The advice rested on a small number of studies, many of them poorly designed by modern standards, and on an assumption that languages compete for a limited mental capacity. Contemporary research has comprehensively overturned that picture.

The modern investigation of childhood bilingualism began in the 1960s with work by the Canadian psychologists Wallace Lambert and Elizabeth Peal, who compared French-English bilingual children in Montreal with their monolingual peers. Their careful matching of socioeconomic background, an innovation at the time, revealed that the bilingual children performed as well as or better than monolinguals on most verbal and non-verbal measures. The finding was at first greeted with scepticism, but has been repeatedly confirmed by later studies across many language pairings.

What bilingual children do differently becomes visible when they are tracked closely over their early years. Children exposed to two languages from birth typically produce their first words at roughly the same age as monolinguals, though their vocabulary in any one language may appear smaller if measured alone. When the two vocabularies are combined, however, the total number of concepts covered is similar to or larger than that of comparable monolinguals. The apparent delay is a measurement artefact, not a cognitive disadvantage. The work of the Canadian psychologist Ellen Bialystok has been particularly influential in correcting this misunderstanding.

Bilingual children do face one genuine challenge, though its practical consequences are mild. Because any given language is used less often than it would be in a monolingual home, bilinguals sometimes take slightly longer to retrieve a specific word, an effect that persists into adulthood. This phenomenon, sometimes called the tip-of-the-tongue problem, is measurable in laboratory conditions but rarely affects ordinary conversation. Against it must be set a range of advantages, many of them concerned with the management of attention and the ability to switch between tasks.

The key mechanism is probably executive control. A bilingual person, whether a child or an adult, must constantly select the language appropriate to the current conversation and suppress the other. This continuous exercise of selective attention seems to strengthen the same brain networks used for executive tasks in general, including focusing on a difficult problem, ignoring distractions, and shifting flexibly from one rule to another. Studies of children aged four to six have found that bilinguals perform reliably better than monolinguals on a standard test in which the sorting rule changes partway through a session. Similar effects have been observed in adults, though the size of the advantage is smaller.

The effects appear to persist into old age. Bialystok has reported that bilingual patients with Alzheimer's disease show clinical symptoms, on average, four to five years later than monolingual patients with similar levels of brain pathology. She does not claim that bilingualism prevents the disease; rather, she argues that it provides a cognitive reserve, a kind of buffer that allows the brain to continue functioning for a time despite underlying damage. Similar reserve effects have been attributed to education, musical training, and lifelong mental engagement.

What counts as bilingualism, however, is not a straightforward question. A child who speaks one language with each parent is in a quite different position from a child who speaks the national language at school and a heritage language at home, and again different from a child who has had a few years of second-language instruction at primary school. Most of the benefits described above are strongest in the first two cases, where both languages are used routinely. A short period of instruction, by itself, does not seem to produce lasting cognitive effects, though it can, of course, be useful for other reasons.

Anxieties about raising bilingual children often focus on mixing. A small child who combines vocabulary from two languages within a single sentence, saying something like "I want le pain", strikes some parents as a sign of confusion. In fact, such mixing is a normal feature of bilingual development and generally follows consistent grammatical rules. By the age of four or five, most bilingual children can separate their languages cleanly when addressed by a monolingual speaker. The mixing that remains is a matter of preference and context, not of underlying confusion.

The educational implications are steadily becoming clearer. Bilingual primary education, once considered risky, is now widely practised and is particularly common in regions with two official languages. Long-term studies of such programmes in Wales, Catalonia, and parts of Canada have shown that children educated through two languages typically match or exceed their monolingual peers in academic outcomes, provided both languages receive consistent support. The outstanding question is how to provide equivalent benefits to children whose second language is not supported at school, as in many immigrant communities, where heritage languages are often lost within two generations despite clear evidence that maintaining them would be beneficial.

Bilingualism, in this revised picture, is not a special condition requiring careful management but something closer to the ordinary human case. Most of the world's population is at least passively bilingual, and the sharp monolingualism of much of northern Europe and North America is a local peculiarity rather than a norm. The research of the last sixty years has, in effect, dismantled one of the more persistent myths about language and the mind.