Almost every bilingual family hears some version of the same warnings. The grandmother who suggests sticking to English so the child won't get confused. The well-meaning friend who blames the late talker on the second language. Sometimes a pediatrician or a daycare teacher who advises dropping the heritage language for a while to "let things catch up." This advice is kind in spirit and wrong on the evidence. Children's brains are built for more than one language, and three decades of research have settled most of the questions parents are still being told to worry about. For more on early language development, visit Healthbooq.
What the Research Actually Says
Bilingual children hit the same major language milestones at the same ages as monolingual children — first words around 12 months, two-word combinations between 18 and 24 months, an explosion of vocabulary in the second and third year. The catch is in how you measure them.
If you count only the English words, a 2-year-old who has 50 English words and 40 Polish words will look like they're behind. They're not. Their total vocabulary is 90 words, which is in normal range for a 2-year-old. Bilingual children typically have somewhat smaller vocabularies in each language than a monolingual peer, because they're splitting their input time between two systems. Their combined vocabulary is comparable or larger.
The "bilingualism causes delay" idea was built on early studies that measured only one language and didn't control for socioeconomic factors. Controlled research has consistently failed to find a bilingual delay effect. The American Speech-Language-Hearing Association (ASHA) and the AAP both state plainly that exposure to two or more languages does not cause speech or language delay.
Code-Switching Is Not Confusion
When a 3-year-old says "I want más leche, please," parents often worry. They shouldn't. Code-switching — mixing words or phrases from two languages in a single utterance — is one of the most studied features of bilingual speech, and it's a sign of competence, not confusion.
Adult bilinguals do it constantly. The switches follow grammatical rules — you can't switch in random places without it sounding wrong, even to other bilingual ears. Children pick up these patterns by ear long before they could explain them. A child who code-switches has both languages active and is grabbing the more available word in the moment. Asking them to "pick one language" interferes with the natural flow of conversation and adds a monitoring task that gets in the way of communicating.
What Actually Drives Language Development
Three things matter, in roughly this order.
Quantity and quality of input. Language develops through hearing it used in conversation, directed at the child, in a back-and-forth way. Reading, songs, narrating what you're doing, talking through the day. Screen time in either language doesn't count for much under age 2 — kids learn language from people, not screens.
Consistency of exposure. Each language needs enough airtime. The "one parent, one language" approach (each parent sticks to their own language with the child) is one common pattern and works well for many families. It's not the only one. "One language at home, one outside" works. "One language with grandparents, the other with parents" works. What doesn't work is sporadic, low-quantity exposure to one of the languages — a few words a week from a relative on a video call won't build it.
Reasons to speak the language. A child who has people they love who only speak Spanish, or Mandarin, or Russian will hold onto that language. A child whose only Spanish speaker is one parent who also speaks fluent English will often drift toward responding in English by age 4 or 5.
The Minority Language Needs Protecting
In an English-dominant environment, English wins by default. Daycare, school, TV, peers, the kid down the street — it all stacks up in English's favor. The other language — the heritage language, the minority language — is the one that needs active effort.
What works:
- Regular contact with monolingual or near-monolingual speakers of that language. Grandparents are gold. So are visits to the home country, where the child has to use the language to function.
- Community: a Saturday school, a playgroup, a religious community where the language is the working language.
- Books, songs, films, and music in the language. Children's songs and rhymes are particularly useful — they pack a lot of vocabulary and grammar into a small format.
- Clear positive framing from parents. Kids absorb whether a language is treated as something valuable or something embarrassing. "Babcia speaks Polish, we speak Polish, this is our language" lands differently than apologetic explanations.
Passive Bilingualism Is Real
Many heritage-language children become "passive bilinguals" — they understand the language fluently but respond in English. This is not a failure. They have the comprehension layer fully built, which is the harder part to acquire. Active production can be reactivated later — a few months in the home country in their teens, a relationship with a monolingual speaker, a college class — much faster than starting from zero.
If you have a child who answers grandma in English when grandma speaks Polish, you have not lost. You have a half-built bridge. Keep the input rich, keep the relationship warm, and don't shame the child into silence by demanding they speak the "correct" language.
When to Get Assessment
If you are genuinely worried about your child's language development — not just hoping for more words at a given age, but seeing patterns that concern you — get an assessment. The flags don't change because of bilingualism: no babbling by 9 months, no first words by 18 months, fewer than 50 total words across all languages by 2, no two-word combinations by 2.5, hard-to-understand speech at 3 to 4.
Two important caveats. First, the assessment should look at all the child's languages — testing only the weaker one will underestimate them. Second, find a speech-language pathologist who has experience with bilingual children. Standard assessments are normed on monolingual populations; an SLP who doesn't account for that risks misdiagnosing a typically developing bilingual child as language-delayed. In the UK, the Royal College of Speech and Language Therapists (RCSLT) can help locate practitioners with the right background; in the US, ASHA maintains a directory.
Key Takeaways
Bilingualism does not cause language delay — that myth has been retired by 30 years of research. Count vocabulary across both languages, not one. Code-switching is sophisticated, not confusion. The minority language is the one that needs active protection — grandparents, books, and reasons to use it do more than rules.
