Accent trait or pronunciation error?

In the past few years, I’ve seen a few teachers in Brazil make a point of differentiating accent traits from pronunciation errors. You can listen to Daniel Bonatti explain it here for CanalRh or Vania Below from ManagedEnglish tackle it here. The implication, as I understand, is that diversity in accents should be celebrated, but pronunciation errors should be corrected.

While I totally agree with the sentiment (yay, diversity!), I find the practicalities of it a little difficult to wrap my head around. My main question is – and this is indeed an honest question, one I don’t know the answer to –: when you subscribe to that view of accent trait vs pronunciation error, how do you decide what to correct in your students’ pronunciation? Similarly, when you assess somebody’s pronunciation, such as in a proficiency test, what do you chalk up to accent and what do you penalize?

Mind, if you base your feedback to students on intelligibility as a parameter, it’s a bit easier, although still not a rested case by any stretch of the imagination. Yet when I try to include that distinction (accent trait vs pronunciation error) to my repertoire, my mind just screeches to a halt.

Allow me a few examples. Would you correct your students if they…

A) … pronounced bus with the vowel of put?

B) … vocalized the final L in Brazil, saying BraziU?

C) … changed the TH in think for a T or an F sound?

D) … used the flap T in internet and better, making it sound something like innernet and berer?

E) … spoke each syllable at seemingly regular intervals and avoided muffling unstressed vowels?

Lo and behold, if you’re using intelligibility as a parameter according to Jenkins (2000), your only problem is letter D. Yes, ladies and gentlemen, her study indicates that the flap T of Standard American English, which so many of us have tried to emulate, actually reduces intelligibility in an international community. Nonetheless, even though I subscribe to the intelligibility view, I’ve never found it in my heart to correct a student that has done the flap T. It is, after all, a feature of a native variety, and a powerful standard variety no less.

So is it native varieties we would use as a parameter then? That’s sort of what I understood from the videos on the difference between accent and pronunciation mistakes, since every example of accent there was a native variety. It might have been a coincidence or a poor choice of examples, but could it be they mean “accent” refers to what’s possible in a native variety, but we should classify as “error” everything that isn’t? In that case, surprisingly, all of the items above are fine: A is a feature you’ll hear in Northern England; B is a characteristic of Estuary English (around London); C is possible in Cockney and in African American English; D of course is Standard American English; and E seems to be the norm in Indian English and possibly also Welsh English.

If you go by native varieties, then, most of what we usually consider as errors go down the drain. In addition, that view poses all kinds of problems. First, not all native varieties are standard, of course, and I reckon most of our students would rather refrain from sounding non-standard, in that it could hinder their acceptance in their lines of business or even social interactions. Also, there is no way we teachers can keep up with all the features of the ever-changing varieties of the English language. Most importantly, though, using native varieties as yardsticks would be a major step back, and I can’t stress it enough: please let us not revert to that. Non-native varieties can be very intelligible and, as they are possibly associated with identity, it could be an intrusion to try and “fix” them. Just listen to Gisele Bündchen speaking English with her beautiful and clear Brazilian accent: who on Earth would want to fix THAT? And with what excuse, right? “No, Gisele, if you speak more like an American, and pronounce the final L as an actual L at the back of your mouth, then you can be MORE successful.” As if!

Basically, that’s where my confused mind is at the moment with this differentiation of accent traits from pronunciation errors: I wouldn’t know how to go about it. If you support this view, I would love to hear from you. I’ll also thank anyone with further references on these terms. I sure cannot understand them on my own.

Hat tip: Thank you, Amanda Estrambek and Carina Alves, for bringing up the subject on BrELT and motivating this post.



Crystal, D. (1995). Documenting rhythmical change. In J. W. Lewis (Ed.), Studies in general and English phonetics: Essays in honour of Professor J. D. O’Connor (pp. 174-179). London: Routledge.

Jenkins, J. (2000). The phonology of English as an International Language. Oxford: Oxford University Press.

Tench, P. (1990). The pronunciation of English in Abercrave. In N. Coupland (Ed.), English in Wales: Diversity, conflict and change (pp. 130-141). Clevedon: Multilingual Matters.

Plus the many links in the article. =)

Natália Guerreiro

Natália Guerreiro has been a teacher since the year 2000 and currently works in Aviation English assessment and teaching for the Brazilian Air Force. She holds a CELTA, a B.A. in English & Portuguese from UFRJ, and an M.A. in Applied Linguistics from the University of Melbourne. She's been elected BRAZ-TESOL's Second Vice President for the 2019-2020 term.

No Comments

Post A Comment