Raising a bilingual toddler: what the research says.

The two worries

“Won’t two languages confuse them?” No. Infants can tell two languages apart from very early on — bilingual newborns already discriminate their two languages, and by the toddler years children keep the two systems separate even while they borrow between them.¹ A toddler who says “I want leche” is not confused; mixing words from both languages inside one sentence is an expected part of multilingual communication, including for adults.¹

“Won’t it delay their speech?” Also no. Bilingual children babble, say first words and combine words on a similar timetable to monolingual children.^{1, 2} Nationwide Children’s Hospital puts it plainly for parents: learning two languages in childhood does not cause confusion or language delay.³

What the evidence says

Input is split, so each language gets less of it

The clearest finding in the literature is almost arithmetic. Exposure is related to growth in each language: in a large study of toddlers exposed to Spanish and English, each language’s vocabulary tracked the share of exposure that language received.² A child who hears 70% English and 30% Spanish will, for a while, know more English words — and that is not a problem to fix, it is a description of the input. If you want the home language to keep up, it needs hours, not intentions.

The input has to come from people

In the study every bilingual parent should know, nine-month-old American infants who spent a few short sessions with a live Mandarin speaker learned to hear Mandarin sound contrasts. A matched group given the identical material on video or audio only did not show the same learning on that test.⁴ This measured a particular speech-sound contrast, not every kind of language learning. Video calls can connect family members, but the WHO guideline does not specify a video-chat exception. See our screen-time guide for current guidance.^{5, 6}

More than one speaker helps

Fourteen-month-olds who heard new words from several different talkers learned to tell those words apart; the ones who heard the same words from a single talker did not.⁷ That result came from a controlled word-learning task; it does not prove a benefit from an app’s two recordings. For a heritage language this is an argument for the grandmother, the uncle, the babysitter, the friend from the same country — anyone who will say the words to the child.

You do not have to be a native speaker, or split by person

“One parent, one language” is one workable strategy, not a rule; families succeed with one language at home and another outside, or with both from both parents.¹ What matters is enough rich input in each language, and a reason for the child to use it. Use the languages you are most comfortable speaking, and include relatives or other speakers when you can.⁸

What to do, practically

• Protect the minority language. The community language will take care of itself at daycare and school. The home language is the one that needs deliberate hours: meals, bath, bedtime, and every call with family.
• Count both languages. Count words in both languages and share the list with the evaluator. That total is for the evaluator to interpret, not a pass mark.
• Let mixing happen. Respond to the meaning, in whichever language you were speaking. You can model the full sentence back without correcting.
• Give the child people, not media. Relatives, playdates, community groups, Saturday school. A recording can model pronunciation; it does not replace back-and-forth interaction.
• Keep it warm. Make room for conversation and play rather than insisting on a correct reply. The goal for a toddler is not fluency; it is enough words to hold up their end of a call to grandma.

When to ask for help

Ask about any missed milestone or communication concern; do not wait for there to be no words. Contact your child’s doctor promptly if words or other skills are lost. An assessment should consider all the child’s languages, with an interpreter if needed.^{8, 9} The first words guide explains the next steps.

Where an app fits, honestly

Not as a teacher. What a flashcard app can do for a bilingual family is hand the parent the word in the language they are less sure of — one picture, the word written the way the language writes it, and a native voice to copy — so that the parent says it to the child. Two native voices per language are a listening option; the study above does not establish their effectiveness. If you use ours, the eight languages sit on the same card, so the same picture can be leche and then milk, one tap apart.