Why
A movement has been brewing in the English language teaching field to raise awareness about the discrimination against ‘non-native speaker teachers’ and the privilege of ‘native speaker teachers.’ (The labels themselves are problematic, and, as the incredible advocates at TEFL Equity have argued, can perpetuate the hierarchy.)
Apart from the very compelling reasons of fighting against the inequality, racism, linguistic imperialism, etc. of the native-speakerarchy, there is a lack of evidence in linguistics research that ‘native speakers’ are superior language teachers.
As someone who got my start in ELT through the privilege of my ‘native speaker’ status and a silver spoon passport, I feel the urgency to use whatever influence I have to advocate for equal status of all English language teachers. SO consider this the inaugural post in a series that digs into research showing the emperor has no clothes. ‘Native speaker’ teachers have no magical teaching abilities bestowed upon them by the fairy godmother of English.
What
Scott A. Crossley and Danielle S. McNamara (2010) set out to determine if ‘non-native speaker’ lexical production (that just means words, kiddos) changes whether their interlocutor is a ‘native speaker’ or not.
DISCLAIMER: Crossley and McNamara do not pass the Holliday Test. Adrian Holliday has had it up to here with ‘native speaker’ bullshit. He’s decided not to review any research that uses the labels ‘native speaker’ or ‘non-native speaker’ without a critical discussion of such. This work was published 8 years ago and does not reflect the latest thinking on the use of those terms. I am giving them the benny of a doubt, but I do wish they had called their subjects ‘language learners’ or SOMETHING less incendiary.
How
The researchers recruited ‘native speakers’ (NS) and ‘non-native speakers’ (NNS) from an American university, and collected a corpora of spoken language between a boatload of NNS-NS and NNS-NNS dyads. They analyzed the interactions with a computational tool, Coh-Metrix, focusing on lexical diversity and frequency.
I MUST DISCLAIM AGAIN: There is no mention of how these groups were deemed ‘native’ or ‘non-native speakers,’ or what that even means, so take it all with a grain salt. Re-read what Adrian Holliday has to say on the issue, and proceed with caution.
They hypothesized that NS-NNS interactions would provide more comprehensible input to the language learner. Comprehensible input is the language that learners are exposed to via listening or reading that is modified for their level of proficiency. It is generally acknowledged to be critical to language acquisition. Crossley and McNamara suggest that because ‘native speakers’ alter how they speak when communicating with someone they perceive as ‘non-native’ their speech will be extra comprehensible.
A second hypothesis was that NNSs would “produce more varied and more infrequent vocabulary” with NS interaction. This is the aforementioned lexical diversity and frequency. Lexical diversity is calculated by dividing the number of different words by the total number of words in a given (in this case spoken) text. The idea is that higher diversity equals a wider vocabulary, a signal of higher command of a language. Lexical frequency refers to how common a particular word is in corpora. Therefore, “more infrequent vocabulary” just means less common words, again signalling a wider vocabulary.
Allow me to paraphrase this ménage à deux of hypotheticals: ‘Native speakers’ will have a positive effect on language learners’ lexical input and output.
And so…?
What they found regarding output was NNSs produced more lexical diversity when speaking with other NNSs, which is to say they used a wider range of vocabulary. They also found that NNSs used more frequent vocabulary when speaking to other NNSs. The explanation being they likely simplify their speech to be comprehensible to lower proficiency speakers, much in the same way NSs do.
Furthermore, “the study demonstrates that NNSs receive no specific lexical benefits related to lexical diversity and frequency from interacting with NSs.” The input from NNSs and NSs was equally comprehensible vis a vis lexical features.
There you have it, folks. A study which a) employed problematic terms, and b) hypothesized NS interlocutors would come with lexical benefits actually provided evidence to file in our ‘NATIVE SPEAKERS’ ARE NOT HASHTAG BLESSED WITH LANGUAGE ACQUISITION GIFT-GIVING-POWERS FILING CABINET.
Indeed, it shows that NNS interlocutors are fine and good and great. Can I extrapolate that ‘non-native speaker’ teachers are fine and good and great? Try and stop me. Crossley and McNamara conclude that their findings have implications on the use of pairwork and groupwork in EFL classrooms, but I think we can also point to this study to show that ‘non-native speaker’ English teachers are not at a disadvantage when it comes to lexical features of input and output.
Recommended reading for all language teacher bishes and bishes who wish to arm themselves against the native-speakerarchy. BONUS: It’s pretty short and not behind a paywall.
Crossley, S. A., & McNamara, D. S. (2010). Interlanguage talk: What can breadth of knowledge features tell us about input and output differences? Proceedings of the 23rd International Florida Artificial Intelligence Research Society.