Disbarring something going horribly wrong between now and December, the job I’ve got in HK is teaching English to young ESL students. Although I’ve been planning on becoming a professional teacher for a while, I don’t have a TEFL qualification, so getting the job was partly on the strength of my A-level in English language (which covered first- and second-language acquisition) and partly also because I tutored my Chinese uni friends English while I was at uni.
However, this sudden transition into formally teaching English-as-a-second-language has forced me to very hastily bone up on all the theory I’d learnt five years ago. Amongst the books I acquired was “Teaching Languages to Young Learners” by Lynne Cameron (Cambridge University Press 2001) – it was great to re-familiarise myself with the various different language acquisition theories and practical uses thereof. Although it’s unsurprisingly geared towards teaching younger children, there are still passages in there that are relevant to older L2 learners. I thought the following section was particularly illuminating:
It has long been hypothesised that children learn a second language better than adults, and this is often used to support the early introduction of foreign language teaching. The Critical Period Hypothesis is the name given to the idea that young children can learn a second language particularly effectively before puberty because their brains are still able to use the mechanisms that assisted first language acquisition. The Critical Period Hypothesis holds that older learners will learn language differently after this stage and, particularly for accent, can never achieve the same levels of proficiency. While some empirical studies offer support for the Critical Period Hypothesis, other studies provide evidence that there is no such cut-off point for language learning. Lightbrown and Spada (1999) present some of the evidence for and against the Critical Period Hypothesis, and remind us to attend to the different needs, motivations and contexts of different groups of learners. They suggest that where native-like proficiency in a second language is the goal, then learning benefits from an early start, but when the goal is communicative ability in a foreign language, the benefits of an early start are much less clear.
I would tend to agree with the conclusion drawn by Lightbrown and Spada; additionally, I’d label the entire Critical Period Hypothesis (CPH) as being fundamentally baseless.
The CPH was first introduced by Wilder Penfield back in 1959 and broadly states that native-level language acquisition cannot occur after a certain cut-off point. In 1964 Lenneberg postulated that this was because the brain lost plasticity after that point. One of the main bodies of evidence used to support the CPH was the case of Genie: she’d been tied up in a darkened room and deprived of an opportunity to learn language until she was twelve by a mentally unstable father, and so she was taken a test subject to see if primary language acquisition was possible after this cut-off point.
Experimental evidence from her case concluded that she was unable to master English fully, although there was a fair degree of controversy concerning the methodology. I criticise not only the methodology (I believe she was often taken out of a natural learning environment that normal children would enjoy in favour of laboratory testing, presumably causing stress) but also the lack of consideration given to the apparently poor brain development owing to the absence of social and physical interaction.
The conclusions of two further linguists support these criticisms: Piaget claims that the mental development of young children is influenced by their interaction with the physical world – for example, by learning how to convey food to mouth via a spoon, or playing with building blocks. Vygotsky goes further and says that the ultimate level of development attainable is more dependent on social interactions a child experiences. Although they factor language acquisition into mental development to varying degrees, the fact is that Genie had neither physical nor social interaction during the Critical Period and so is empirically atypical: it’s patently unscientific and illogical to base more general theories of first/second language acquisition on this type of case.
Although, as Lightbrown and Spada pointed out, there is evidence both for and against being able to speak L2 at a native level, I would suggest that the evidence against is based largely on data that makes fundamentally incorrect assumptions and uses incomparable data (regarding L1 vs L2 acquisition). The subtext of “[we must remember] to attend to the different needs, motivations and contexts of different groups of learners” is “students learning L2 for one hour per week are never going to be as good in L2 as they are in L1”. The implication is that there is a strong possibility of L2 students being able to achieve native-level fluency.
The final part of the passage is equally interesting:
Further support for making this key distinction comes from a recent study into brain activity during language processing (Kim et al. 1997). This study discovered that the brain activity patterns of early bilinguals, who learn two languages at the same time from infancy, differ from those of learners who being learning a language after about 7 or 8 years of age; different parts of the brain are used for language recall and activation. Foreign language learning… is thus an essentially different mental activity from early simultaneous bilingualism and from L1 acquisition.
It’s difficult to construct an argument without knowing the full details of the study, so I’m going to do some further research into what exactly Kim et al.’s study involved and the background of the participants. I suspect that it may again fall foul of Lightbrown and Spada’s observation noted above, that the data collected was fundamentally incomparable. If this turns out to be the case, then it is critical to undertake further research wherein older L2 learners are exposed to their target language under the same conditions that an L1 child would receive.
One final point I’d like to draw out (from my recent trip to Hong Kong) was how most people I spoke to were unhesitating in replying to me in Cantonese, provided I asked them questions in Cantonese. This was not limited to monolingual speakers either (it turned out after I’d been unable to fully follow their directions). The conclusion I would draw from this is that even though my overall command of the language leaves a lot to be desired, even at this stage my accent was sufficiently good to pass for a fluent/native speaker. (All those listening SRS cards really paid off.) This is again incongruent with the suggestion that “older learners… particularly for accent, can never achieve the same levels of proficiency [as a native speaker]”. It’s more to do with the methodology used, the time invested and the amount of feedback elicited from experts (i.e. natives).
I’d like to draw the tentative conclusion that there’s no particular reason to start children learning a second language sooner rather than later and that more pertinently there’s nothing wrong with starting to learn a language well after the limits defined by the Critical Period. Furthermore, I’d suggest that more research is needed that juxtaposes comparable data (where both L1 and L2 are acquired under the same conditions) since many contemporary papers suffer from conclusions rooted in potentially incorrect assumptions.
There’s only one way to prove it one way or the other though, and that’s for as many people as possible to aim for native-level fluency in learning a second language.
Anyone feel like sponsoring a PhD?
 It all goes to show the importance of taking a balanced set of subjects – my degree was in Physics, and (if I may say so myself) there was a distinct lack of literacy in the other students who’d taken exclusively science and maths subjects before university.