Has there been any study of the age at which acquiring a native accent in a second language falls off a cliff? Anecdotally, my relatives who’ve immigrated in their youth (mainly from Slavic-speaking countries) tend to speak unaccented English if they got here before their teens, and speak with a pronounced accent if they arrived any later, generally speaking with a heavy accent if they arrived after they turned 15 or so. But is there any basis for a more generalized observation about immigrants, languages, accents, and ages?
One thing that always bothered me about the TV series The Americans is the lack of a Russian accent in the two main characters, who are shown to have lived in the USSR until their late teens (I think) but who spoke with no Russian accent at all after they came to live in the US. There was supposedly deep training, enabling them to fit in as lifelong Americans, but is it really possible for someone to get adequate immersion in English to pick it up like that?
I can be shown that there are exceptional people who’ve acquired a native accent at an advanced age, but that’s not really what I’m asking about. I am asking about the age at which most people can acquire an accent that’s virtually indistinguishable from a native accent vs. the age that most people cannot. Any further insight into the reasons for this phenomenon, techniques for extending the process further into one’s teens, specific languages (?) that promote or prevent the acquisition of unaccented speech, etc. are also welcome but mostly I’m interested in the age(s) at which one can speak a second language with no discernable accent.
In cases where a person doesn’t have time or native ability to acquire a correct accent (not unaccented, there is no such thing as no accent in any language) there are coaches and teachers who specialize in that sort of thing. This is most commonly used for actors to speak the proper accent for their character in a movie. Studios don’t always bother to provide this, and some actors have more facility at this than others. When I lived in Japan, I was such a person for Japanese people who wanted to speak good American English – not professionally trained, of course, but just an example that they could listen to and imitate.
In the case of immigrants, I think a lot depends on how much time the immigrant spends immersed in the new language, listening and speaking, vs. staying comfortable listening and speaking mostly in their native language. My frame of reference specifically is Japanese people who have lived in the US for a long time. My husband has lived here for 50 years but he still has a horrible accent, and he spends most of his time watching Japanese TV and Youtube, and talking to his Japanese friends and relatives. Some of his friends that I talk to have very little accent, and they seem to spend most of their time interacting in English. For immigrant children, then, I suspect it is not only their language plasticity that is better than immigrant adults, but also their increased likelihood to mingle more with native children (in and out of school) and their desire to fit in that drives their language absorption.
Yes, but the curve drops off sharply at some point, in my experience around age 12 or 13, and this is what I’m asking about specifically? Is my experience true, and if so, why that age? Is this universal with all languages or specific to some and not others?
I think it depends on the individual and the circumstance surrounding that individual. Arnold Schwarzenegger is something like 77 years of age, and his accent sounds exactly the same to me as it does in the videos of him in his old Mr. Universe days. On the other hand, I have some young Latino parents in my district who have only a trace of an accent in their English.
I have a passing interest in accents, so I did a bit of googling on the subject.
Most of the info I found concerned the ability at different ages to acquire a second language at all, but part of this article specifically addresses accents. According to it, before age 10 is the best time to learn a second language without any trace of an accent; afterward, one can still become fluent in another language, but we lose the ability to differentiate between subtly different phoneme sounds, which is the key to speaking in the native accent. We simply become so used to a certain set of vowel sounds that our brains apparently just map subtly different ones to the closest ones we already know.
Of course, in a case like Schwarzenegger’s, the accent might be intentional - to preserve a trademark sound he’s widely associated with. It’d be interesting to know what he sounds like in private conversation.
Anecdote, So Not Data: Had a very good friend, and roommate. He got a job he couldn’t pass up, so he moved from California to Tennessee. He’s been there about 20 years now. Its been interesting for us, as we call each other every one/two months, to hear how each other sounds. About a year into it, he started hearing the Californian accent, and I let him know he was getting the typical Southern accent. Nowadays, he’s pure South when talking, and he only gets the CA accent back when he’s here for three weeks during the Christmas/New Years holidays. His wife always comments on it when they get back.
I dated a girl many years ago who later went off to do the travel-around-Europe thing. She was early 20’s and ended up spending about 2 or 3 years in Lerwick in the Shetlands before returning to Canada. When she got back she had a distinct trace of English or Scottish type accent.
I also know people who have had very strong accents despite being in Canada for decades, and others with little or no accent (compared to Canadian, eh?). My thought was it varies greatly by individual, but as mentioned, the degree of interaction in a language matters too.
This doesn’t answer the OP’s question, but an amusing anecdote: Back in the 1980s, we saw a performance of something based on Dracula, at a theatre in North Carolina. One character was from Texas but supposedly played by a British actor (I think there was some play-within-a-play thing, or some such). So: an American actor, playing a British actor, doing a bad Texas accent.
That had to have been really tough! I’m not tops on identifying accents, but the fellow was pretty convincing at the British-doing-bad-Texas thing.
That sounds similar to My Fair Lady, for which Audrey Hepburn (who could, of course, speak the upper class British way) had to learn a working class Cockney accent so she could play a flower girl trying to get rid of her Cockney accent.
Your basic premise is correct. My wife had a good friend in college who was 12 when she emigrated from Latvia. She had a slight, but detectable, accent. Her sister, two years younger, had no accent whatever (except whatever you detect of Connecticut), while her 4 years older sister had a fairly strong accent, despite having a PhD in English.
My maternal grandmother was 13 and had a slight accent. My grandfather immigrated at 1 and, of course, had no accent, except for Philly.
I have met two people who learned English as adults and had no accent (except British, where they lived). One was Hungarian (and his wife had a nearly incomprehensibly thick accent) and the other was Danish. The former was a cousin of Erno Rubik. But these two were very clearly exceptional.
I had a Chinese professor in college who spoke and wrote English exceptionally well, but his speech was strongly accented.
I knew a young lady who was from Czechia. I don’t recall how long she had been in the US, but she did have a slight–and I do mean slight–accent. I saw her a few years ago after 10+ years, and I completely didn’t recognize her. After speaking to her for a minute, I realized who she was.
She looked exactly the same, but she had no accent that I could detect, and an accent probably would have tipped me off when she first spoke.
A few years ago This American Life had a Chinese-American woman on the show who claimed she could tell how old someone was when they first came to America based on how they speak English:
One argument I’ve read about why 2nd language learning ability drops off a cliff around that age is that it’s related to puberty.
Specifically, Nature “wants” the soon-to-be parents to teach the baby adult language, not have the baby teach the parents a new baby-language. Making the parents mostly new-language-proof helps with that.
I’m in no position to evaluate the merits of this idea, but it came from a legit academic source, not some rando on FaceBook.
Critical period hypothesis states that after a certain age, children’s brains filter the sounds that are not found in the languages spoken around them because they’re not useful to them. Presumably it takes too much energy keep the ability to distinguish and reproduce phonemes that you never come across.
It seems that there is a wide variation as to what that age is supposed to be. I remember reading back in the 90s that it was around 8-10 years old. After that, unknown phonemes are filtered and reinterpreted following a “close enough” method. Zis is wai ze FRench spik laik zis.
Of course, it’s not a hard and fast rule. Some people are very good at imitating what they hear very precisely, others are just hopeless no matter how long they try. Personally, I’ve never had too many problems pronouncing foreign sounds except for the the rolled r you find in Italian and Russian, and various uvular, pharyngeal, and glottal consonants in Arabic. I’ve gotten better at the former, but for the latter, I still don’t even hear the differences. Apart from those, I’ve been able to give relatively faithful imitations of the new sounds I’ve been confronted with in the dozen languages I’ve learned (to wildly different levels of proficiency, though).
Note that this doesn’t hold for other aspects of second-language acquisition. It is absolutely possible to have a broad vocabulary in a foreign language you learn as an adult, and grammar is possibly more easily and more quickly learned as an adult, due to age-related differences in the way humans process and synthesize new information.
Lincoln Steffens - Wikipedia reported that when he returned to Germany after spending time in France, his German got much better.
Something similar happened to my sister in her 30’s. She lost most if not all of her original American accent after returning to Melbourne from 5 years in Tennessee.
Also note that it takes something like 18 years to reach full proficiency in English, and that’s studying full time. As demonstrated by talking to and listening to younger people. With the best skill in the world, it takes a long time to really learn a language at native level. Even 10-12 yr olds haven’t achieved it.