True, but I really dislike the amount of focus people put on that. With the right techniques, it's not particularly difficult for an adult to develop decently good pronunciation that allows them to be comfortably understood. And placing the bar for having learned a language to a high level of proficiency at "able to hide evidence of where they were born from native speakers" smacks of internalized xenophobia.
I also read in a second language acquisition textbook a while back (so, decent chance of being out of date, also decent chance of my memory of what I read being less-than-perfect) that the strongest predictor of how native-like an accent someone develops isn't actually the age at which they started learning the language, it's the lag between when they started learning it and when they started socially integrating into a community of native speakers. And it happens to be the case that, for all sorts of practical reasons, there's a strong correlation between this lag period and age. Which isn't to say that there's nothing to the critical period hypothesis - there is - but when we're talking about children it's difficult-to-impossible to root out selection bias effectively enough to permit even a convincing stab at partial identification of causal effects. Which creates a risky situation for the purposes of drawing firm conclusions, because our prejudices love to helpfully answer the questions that science won't.
In grad school I had a social science textbook that dubbed these kind of questions "Fundamentally Unanswerable Questions" in the first chapter, and in subsequent chapters simply described them as "FUQ'd".
I also read in a second language acquisition textbook a while back (so, decent chance of being out of date, also decent chance of my memory of what I read being less-than-perfect) that the strongest predictor of how native-like an accent someone develops isn't actually the age at which they started learning the language, it's the lag between when they started learning it and when they started socially integrating into a community of native speakers. And it happens to be the case that, for all sorts of practical reasons, there's a strong correlation between this lag period and age. Which isn't to say that there's nothing to the critical period hypothesis - there is - but when we're talking about children it's difficult-to-impossible to root out selection bias effectively enough to permit even a convincing stab at partial identification of causal effects. Which creates a risky situation for the purposes of drawing firm conclusions, because our prejudices love to helpfully answer the questions that science won't.
In grad school I had a social science textbook that dubbed these kind of questions "Fundamentally Unanswerable Questions" in the first chapter, and in subsequent chapters simply described them as "FUQ'd".