Age of Onset and Nativelikeness in a Second Language: Listener Perception Versus Linguistic Scrutiny

195 Spanish/Swedish bilinguals screened; 41 scrutinised in depth

Spanish-L1 speakers of Swedish in Sweden with ages of onset from under 1 to 47 years, ALL of whom self-identified as potentially native-like

Two-stage design and the two stages give different answers, which is the point. Stage one: native-speaker judges listen to speech samples and decide whether the speaker is a native Swede. Stage two: the subset who passed is scrutinised with a battery of 10 highly complex, cognitively demanding tasks plus detailed measurements of linguistic performance, representation and processing. The sample is deliberately selected for apparent nativelikeness, so it is a study of the best cases, not a random sample - which makes the stage-two result stronger, not weaker, as a boundary marker.

Under casual listening, only a small minority of bilinguals whose L2 acquisition began after age 12 were perceived as native, while a MAJORITY of those with an age of onset below 12 were. Under linguistic scrutiny, NONE of the late learners performed within the native-speaker range - and only a few of the early learners did either. The second half of that sentence is the part usually left out: perceived nativelikeness in early starters largely dissolves on close inspection too.

High. Participants are self-selected on believing themselves potentially native-like, and age of onset is not randomised. The relevant unmeasured variable is language aptitude, which is heritable and which plausibly determines both who becomes near-native and who volunteers for a nativelikeness study.

The specific two-stage design has not been repeated at this scale. Its qualitative conclusion - that scrutiny removes apparent nativelikeness in nearly everyone - is consistent with the wider ultimate-attainment literature but rests on this one study.