Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language

46 native Korean and Chinese speakers

Korean and Chinese L1 speakers who arrived in the United States between ages 3 and 39, all with several years of US residence, tested as adults

The founding study of the modern critical-period literature for L2. A 276-item auditory grammaticality-judgement test administered once in adulthood. Small sample split into arrival-age bands, which is the source of most later methodological criticism: subgroup correlations computed on small, range-restricted bands, and a breakpoint chosen rather than estimated. Researcher-designed instrument, single modality, single occasion.

Among those who arrived between ages 3 and 15, grammaticality-judgement score declined steeply and roughly linearly with age of arrival (reported correlation about r = -0.77). Among those who arrived from about age 17 onward, score was low, highly variable, and essentially uncorrelated with arrival age. The claimed discontinuity between those two regimes is the empirical core of the critical-period hypothesis for L2 - and it is the specific pattern that later reanalyses attack.

High. n = 46, no randomisation, arrival age confounded with education, family circumstance, L1 background and language aptitude.

Repeatedly re-analysed and repeatedly weakened. Vanhove (2013) shows the analytic strategy cannot distinguish a critical period from a steady decline; Hakuta et al. (2003) find no discontinuity in a sample fifty thousand times larger. The gradient replicates; the discontinuity does not.