Most advice about raising a bilingual child stops at encouragement. Speak your language, read books in it, be consistent. All true, and all frustratingly vague when what you actually want to know is whether two evenings a week is enough. It is a reasonable question, and unlike a lot of parenting questions, it has numbers attached to it.
Researchers who study bilingual development measure exposure as a proportion of everything a child hears in a day. That proportion turns out to predict vocabulary well enough that speech and language therapists use it clinically. Here is what the thresholds are, what they translate to in hours, and where the numbers stop being useful.
How much exposure does a second language actually need?
Research generally treats a child as bilingual from around 10 to 25 percent exposure to a second language, but matching a monolingual child on vocabulary takes considerably more. In a widely cited study of Montreal children acquiring French and English, Elin Thordardottir found that children needed roughly 40 to 60 percent exposure in a language to score in the monolingual range for receptive vocabulary, the words they understand, and closer to 70 percent to reach the monolingual range for expressive vocabulary, the words they produce. The study was published in the International Journal of Bilingualism in 2011.
The 10 to 25 percent floor comes from the wider literature, summarised in Bilingualism in the Early Years: What the Science Says by Krista Byers-Heinlein and Casey Lew-Williams. Their point is worth keeping: crossing that floor makes a child bilingual in the research sense, which is not the same as fluent.
What does that look like in hours per week?
A preschooler is awake roughly 12 hours a day, which is about 84 hours a week of potential language input. Converting the research thresholds into that budget makes them a lot easier to plan around.
| Share of daily input | Roughly, per week | What research associates with it |
|---|---|---|
| 10% | 8 hours | Lower bound at which studies count a child as bilingual. |
| 25% | 21 hours | Solid comprehension, vocabulary still below monolingual peers. |
| 40% | 34 hours | Bottom of the range linked to monolingual level comprehension. |
| 60% | 50 hours | Comfortably in the monolingual range for understanding. |
| 70% | 59 hours | Level associated with monolingual range spoken vocabulary. |
Two honest caveats before anyone starts a spreadsheet. These figures come from vocabulary tests with specific samples of children, mostly in Montreal, not from a universal law. And exposure is not the only variable: age of first contact, family circumstances and the number of different people a child hears all shift the picture.
Does more exposure always mean more language?
Broadly yes, but not in a straight line, and the relationship is better described as diminishing returns. A 2023 study in Cognition used data from bilingual children specifically to test how well models of early vocabulary growth handle differences in input, and found that simple more is more accounts do not fully explain what children learn.
The practical reading is that the first hours added to a weak language are worth far more than the last ones. Going from three hours a week to ten changes a great deal. Going from fifty to fifty seven changes very little.
Does the quality of the input matter as much as the quantity?
Yes, and in some respects more. Erika Hoff's review of input and language development in bilingually developing children found that the source of input matters as much as the amount: native speaker input, a variety of different speakers, and interactive conversation all support development more than the same number of hours from a single non native source.
Two findings from this literature come up repeatedly and are worth stating plainly. Language heard during shared book reading is unusually supportive of development in that language. Language heard from television is not particularly supportive at all. That is a strong argument for building the second language into things you were going to do anyway, and our list of simple activities for learning English at home is built around that principle.
The one person one language rule, incidentally, is not required. The evidence shows it is neither necessary nor sufficient for successful bilingual acquisition. Plenty of families mix languages within a conversation and raise fluent bilinguals.
Will loading up on a second language hold back the first?
No, and this is the worry most worth putting down. Bilingual children reach early language milestones at the same age as monolingual peers when you count what they know across both languages, as shown in research published in the Journal of Child Language.
What does happen is that a bilingual child's vocabulary in each single language may look smaller than a monolingual child's, simply because their words are distributed across two systems. That is a measurement artefact, not a delay, and it is the single most common reason bilingual children are wrongly flagged for assessment. We went through the evidence in detail in our piece on whether two languages delay speech.
How do I actually estimate my child's exposure?
Speech and language clinicians do this routinely, and the method is simpler than you would expect. It involves walking through a typical weekday and weekend day hour by hour and assigning each block to a language. Researchers have validated shortened versions of this interview for clinical use, described in A Simplified Approach to Quantifying a Child's Bilingual Language Experience and tested across two bilingual communities in the Journal of Speech, Language, and Hearing Research.
Do it once, honestly, and the result is usually lower than parents expect. School hours dominate the total, and a language spoken only at bedtime rarely clears 15 percent.
What if we cannot get anywhere near 40 percent?
Then aim for consistency and interaction rather than a number, and accept a realistic goal: strong comprehension now, with production catching up later if exposure increases. Many adults who grew up understanding a grandparent's language without speaking it back reactivate it quickly as teenagers or adults. Comprehension is a real asset, not a failed attempt at fluency.
Practically, the highest yield moves are the ones that add interactive hours rather than background hours. A weekly video call with family who speak only that language. Bedtime books in the minority language rather than the majority one. A childminder or activity club in the target language. Anything where your child has to answer.
Some families use a talking companion to add a few of those interactive minutes. Ted, our screen free interactive plush, speaks more than 60 languages and lets a parent set the language in the app, which is why bilingual families tend to use it for the language that gets the least airtime at home. It is worth being clear about what that is and is not: 20 minutes a day of conversational practice is a useful supplement to a family's language plan, and it is nowhere near a substitute for the hours in the table above. If you are still deciding when to start, our article on the best age to start a second language covers the timing question.
Frequently asked questions
How many hours a week does a child need to learn a second language?
Research links roughly 40 to 60 percent of a child's daily language input to monolingual level comprehension, which is about 34 to 50 hours a week for a preschooler awake 12 hours a day. Around 25 percent, or roughly 21 hours a week, supports solid understanding but usually leaves vocabulary below monolingual peers.
What percentage of exposure makes a child bilingual?
Studies typically count a child as bilingual from about 10 to 25 percent exposure to a second language. That threshold defines bilingualism for research purposes and does not imply fluency, which requires substantially more input.
Is one hour a day of a second language enough?
One hour a day is roughly 8 percent of a preschooler's waking input, which sits at the very bottom of the bilingual range. It can build meaningful comprehension over years, especially if the hour is interactive rather than passive, but it is unlikely to produce monolingual level vocabulary on its own.
Does watching cartoons in another language count?
Barely. Research on bilingual input consistently finds that television exposure is not particularly supportive of language development, while shared book reading and live conversation are. Screen input should be counted as a bonus rather than as part of your hours.
Do I have to use one parent one language?
No. The evidence shows the one person one language approach is neither necessary nor sufficient for raising a bilingual child. What matters more is the total amount of interactive input in each language and the number of different people providing it.