Lead
Among parents of young children, the anxiety is familiar: we need to start English early, before the window closes. The idea that children's brains are more plastic, more absorbent in the early years, feels intuitively correct — and it has the appearance of scientific backing. Early-enrollment English conversation classes, English picture books, English-dubbed cartoons: parents encounter these choices constantly, and the framing around them usually implies urgency.
But second-language acquisition (SLA) researchers have been working on this question for more than 40 years, and the picture that has emerged looks somewhat different from the popular narrative. The emerging consensus is that when you start matters less than how much exposure you accumulate, and in what form. That conclusion deserves to be taken seriously.
Background: What Is the Critical Period Hypothesis, Exactly?
The critical period hypothesis: the theory that there is a biologically determined window in childhood during which language can be acquired with native-level fluency, after which acquisition becomes much harder in language acquisition was systematized by Lenneberg (1967). The core claim is that a biologically set window exists during which acquiring a language to native-speaker levels is possible — and once that window closes, full acquisition becomes much harder. This hypothesis has been widely cited and played a substantial role in expanding the market for early childhood English instruction.
The trouble is that researchers have never agreed on when the critical period ends. Competing proposals — "age 6–7," "early adolescence," "puberty" — have accumulated without resolution, and the actual data turn out to be more complicated than any of them.
Part 1: Where Does the Critical Period End?
DeKeyser (2000) analyzed data from 57 Hungarian immigrants to the United States and examined the relationship between age of arrival and grammatical attainment in English [1]. The study suggested a transition point at which adults lose access to implicit learning — the mechanism by which children absorb language without conscious effort — and was widely cited as supporting the critical period hypothesis. The sample of 57, however, is a significant constraint on generalization.
A far larger dataset became available in 2018. Hartshorne, Tenenbaum, and Pinker collected and analyzed data from approximately 670,000 English learners via an online survey [2]. Their finding is worth dwelling on: grammatical acquisition ability is maintained until around age 17.4, after which it declines sharply. This substantially revises the earlier "10–12 years" estimate, pushing the end of the critical period into late adolescence.
The implication for the "start before it's too late" framing is significant. While there is a tendency for earlier starters to achieve more native-like pronunciation and grammar, considerable plasticity is maintained well into the teenage years.
Part 2: Exposure Amount and Quality Matter More Than Starting Age
The finding that total amount and quality of exposure outweigh starting age has been replicated across multiple studies.
Muñoz (2006) led a controlled comparison study in Catalonia, Spain, in which learning time was held constant across groups [3]. The core finding: when exposure time is controlled, the advantage of an earlier start disappears. Earlier starters tend to show higher ultimate attainment, but this is better explained by the greater cumulative hours of contact they have accumulated — not by the age at which they began.
This matters in the context of how English is taught in Japanese schools. Even after the 2020 mandatory introduction of English instruction in fifth and sixth grade, the standard allocation is 70 class periods per year — fewer than two per week. SLA researchers refer to this kind of fragmented contact as "drip-feed" instruction, and it falls far short of the input volume that acquisition research identifies as necessary [4]. The difference in outcomes between a child who started in third grade and one who started in fifth grade, at the same weekly contact rate, is likely smaller than the difference between any child receiving two periods a week and one receiving meaningfully more.
Part 3: Designing Exposure at Home
If school instruction hours are constrained, the design of exposure at home becomes relatively more important.
The approaches best supported by research involve building up meaningful input — comprehensible input: language material that learners can mostly understand from context but that contains some new elements just beyond their current level, in the language of Krashen's input hypothesis (i+1: material just above the learner's current level) — through extensive reading, listening, and sustained engagement with English content. Picture books, audiobooks, subtitled video: the key word is sustained. The specific formulation of the input hypothesis has received criticism, but the broad direction — large quantities of enjoyable material at a manageable difficulty — is shared by many researchers.
Motivation is at least as important a variable as exposure volume. Particularly from early adolescence onward, intrinsic motivation — interest-driven engagement with the language — is what sustains continued exposure [6]. Building a sense that English is genuinely enjoyable, not a test subject, during the elementary years is in effect an investment in the learning foundation that middle and high school will demand.
One area where early-start advantages are most clearly documented is phonology: the study of the sound system of a language — including which sounds it uses and how they combine. Children who begin exposure early tend to show greater phonological flexibility — their ears and mouths adapt more easily. But this advantage in pronunciation is a separate matter from grammar, vocabulary, and functional competence, and should not be used to justify the stronger claim that early starting determines overall language outcomes.
Putting It Into Practice
Think about weekly contact hours. The threshold researchers often cite — meaningful exposure of three to five or more hours per week — is not achievable through school instruction alone. Combining home video, books, and songs to build contact volume is a realistic design goal.
Prioritize enjoyable material over accuracy. Pronunciation correction and explicit grammar instruction are less productive than finding English content the child will return to voluntarily. Repeated, self-motivated engagement is the foundation of long-term contact accumulation.
Build motivation for later learning during the elementary years. A child who experiences English as a tool for accessing interesting things — stories, games, people — rather than purely as a school subject carries something into adolescence that no class schedule can provide.
Summary
The end of the critical period is later than commonly assumed. Hartshorne et al. (2018) found that grammatical acquisition ability persists into the late teens [2]. Starting age, while not irrelevant, is a weaker predictor of ultimate outcomes than accumulated exposure and motivation quality.
When thinking about investing in English education, the more productive question is not "are we starting early enough?" but "how do we build the contact hours and motivation that acquisition actually requires?"
References
- DeKeyser RM. The robustness of critical period effects in second language acquisition. Stud Second Lang Acquis. 2000;22(4):499–533. doi:10.1017/S0272263100004022
- Hartshorne JK, Tenenbaum JB, Pinker S. A critical period for second language acquisition: evidence from 2/3 million English speakers. Cognition. 2018;177:263–277. doi:10.1016/j.cognition.2018.04.007. PMID: 29729947.
- Muñoz C, ed. Age and the Rate of Foreign Language Learning. Clevedon: Multilingual Matters; 2006.
- Butler YG. Bilingualism/multilingualism and second-language acquisition. In: Bhatia TK, Ritchie WC, eds. The Handbook of Bilingualism and Multilingualism. 2nd ed. Oxford: Blackwell; 2013:109–136.
- Singleton D. Language and the Lexicon: An Introduction. London: Arnold; 2000.
- Muñoz C. Starting age and other influential factors: insights from learner interviews. Stud Second Lang Learn Teach. 2011;1(3):465–484. doi:10.14746/ssllt.2011.1.3.9
- Nikolov M, Muñoz C. Recent research on age, second language acquisition, and early foreign language learning. Annu Rev Appl Linguist. 2011;31:1–23. doi:10.1017/S0267190511000013
- Ministry of Education, Culture, Sports, Science and Technology (MEXT). Current status and challenges in strengthening foreign language education in elementary schools. 2023. URL: https://www.mext.go.jp/a_menu/kokusai/gaikokugo/index.htm