“Children pick up languages so much faster than adults” is one of the most repeated claims about language learning, and it is only partly true. Adults typically learn faster at first. Children, on average, tend to go further in the long run, particularly in accent. The underlying question of why — and whether there is a hard biological cutoff — is one of the genuinely unresolved debates in language acquisition research, not a settled fact. Here is what the evidence actually supports, and where it stops being clear-cut.
Where the idea comes from
The idea that language has a “critical period” originates with neurologist Eric Lenneberg’s 1967 work on first-language development, which proposed that the brain’s capacity to acquire language fully is tied to a biological window closing around puberty. That original claim was about children acquiring their first language at all — it says nothing directly about adults learning a second or third language, but the term has since been extended, much more loosely and much more contentiously, to second-language acquisition.
What a large 2018 study actually found
The most extensively cited modern evidence on this question comes from a 2018 study by Joshua Hartshorne, Joshua Tenenbaum and Steven Pinker, published in the journal Cognition and based on a purpose-built grammar quiz completed by 669,498 native and non-native English speakers of varying ages and backgrounds. Using a statistical model designed to separate a person’s current age, the age they were first exposed to English, and how many years of experience they had, the researchers found that the underlying ability to learn grammar to a native-like standard appears to hold up until around age 17.4, after which it declines steadily. That is a considerably later cutoff than the popular version of the “critical period” (often assumed to close in early childhood), and the decline the study found was gradual rather than an abrupt cliff-edge at one specific age.
It is worth being precise about what this study measured: grammatical judgement on a written quiz, for English specifically, among people who had varying amounts and types of exposure to the language. It is a single (very large) study, not a settled consensus, and other researchers have pushed back on aspects of the underlying statistical approach and on whether the same pattern holds for skills beyond grammar, such as accent.
Where the disagreement actually sits
Second-language critical period research is genuinely contested, not just in the popular press but among specialists. Points of real disagreement include:
- Whether there’s a sharp cutoff or a gradual slope. Some researchers report evidence of a fairly defined decline point; others, including a widely cited statistical reanalysis published in the journal PLOS ONE, argue that the specific age patterns predicted by a strict critical-period model are not consistent across different languages and learner populations once the data is re-examined.
- Whether different language subsystems have different windows. There is broader agreement that accent (pronunciation) shows the clearest age effects — learners who start young are, on average, more likely to reach accent that native speakers judge as indistinguishable from their own — than there is for grammar or vocabulary, where adult learners can and regularly do reach very high, sometimes native-like, levels.
- Whether “critical period” is even the right framing. Some researchers prefer “sensitive period” language, meaning learning becomes progressively harder rather than switching off at a fixed point, or argue the age effects that are observed are better explained by non-biological factors — amount and type of exposure, motivation, literacy in a first language, and social integration — than by a maturational deadline in the brain itself.
Given this, a flat statement like “it’s biologically impossible to become fluent as an adult” is not supported by the research base — adult learners reach very high proficiency in second languages routinely, including in immersion and intensive-study contexts (see our guide to what to expect from immersion and study abroad programmes). What the evidence more consistently supports is a difference in typical trajectory, not a hard ceiling.
The part that’s less controversial: rate versus ultimate level
Separate from the critical-period debate, a distinction that comes up repeatedly in the wider second-language acquisition literature is the difference between how fast someone learns and how far they eventually get. Classic longitudinal studies of learners placed in the same immersion environment have found that older children and adults tend to outpace young children in the early months — they bring stronger literacy, more developed study strategies, and an existing grammatical vocabulary they can use to consciously analyse the new language. Over a longer timeframe, though, younger starters more often catch up and, on measures like accent in particular, surpass older learners. In practice this means an adult beginner should not expect a young child’s trajectory as a benchmark for early progress: outpacing a child learner in the first few months is a normal, well-documented pattern, not a sign that something is wrong.
What this means for setting realistic expectations
If you are an adult starting a new language, the honest summary of the research is: you are not working against a closed biological door, but you may need to put more deliberate, structured effort into aspects like pronunciation than a child in the same amount of immersion would, and comparing your early progress unfavourably to how “kids just pick it up” is not a fair comparison — children in genuine immersion settings are typically exposed to vastly more hours of input than most adult learners realistically have time for, which does far more work than age alone. For techniques with a stronger, less contested evidence base behind them, see our guide to language learning tips backed by evidence.
Sources
- Hartshorne, Tenenbaum & Pinker (2018), “A critical period for second language acquisition: Evidence from 2/3 million English speakers,” Cognition (via MIT DSpace)
- Vanhove, “The Critical Period Hypothesis in Second Language Acquisition: A Statistical Critique and a Reanalysis,” PLOS ONE (2013)
- “Critical periods for language acquisition: New insights with particular reference to bilingualism research,” Bilingualism: Language and Cognition (Cambridge Core)