[ advanced search ]

Join TESOL Membership Benefits   A |  A |  A

Login
Please allow pop-ups
Email Address:

Password:


    Need Help?
   Forgot Password?

TESOL Matters Vol. 9 No. 3     (June/July 1999)

From the column of the TESOL Speech/Pronunciation Interest Section
by John M. Levis

Judy Gilbert, in a 1998 message to an Internet discussion group, observed, "I have noticed that when British pronunciation people get together to list suitable topics for a colloquium, the subject of What Model? always comes up, often with an air of urgency. But when American pronunciation people get together for the same purpose, it never comes up at all" (from an e-mail discussion list). A look at the kinds of articles published by British and U.S. writers supports Gilbert's observation. British writers such as David Crystal, Jennifer Jenkins, and Bryan Jenner regularly examine the issue of international standards. This is largely driven by two interacting dynamics: the need to recognize the validity of nativized varieties of English, such as Indian English, Singaporean English, and Nigerian English, and the fact that communication between speakers of nativized and native varieties can more easily involve intelligibility difficulties.

U.S. discussions about pronunciation reflect little interest in the issue of international standards. Another closely related issue, however, strongly affects teaching and materials writing. This issue is what to do about acceptable variations in pronunciation. By variation, I mean the differences that mark speakers as being from specific dialects but do not interfere with understanding. There is widespread awareness of some well-known variants in pronunciation, such as either /iD«r/ or /aiD«r/ or tomato /t«meIto/ or /t«mato/. There is far less awareness that patterns of acceptable variation, such as the difference between the so-called r-less dialects of Boston and some areas of New York versus the non-r-less dialects elsewhere in North America, are widespread in English pronunciation.

One common variation, the vowels of the words cot and caught, constitutes one example of how differences in pronunciation should be treated. For many speakers of American English, these two vowels are distinct. A very large number of educated native speakers, however, use the same vowel for both words. I happen to use both vowels. For me, the names Don and Dawn are as different as, well, men and women. My wife, on the other hand, born and bred in the western U.S., hears and uses the same vowel for both. Should she need to learn to pronounce English correctly (meaning the way I do)? Though this question may appear absurd, it is important, judging by the number of computer-based accent-reduction programs that include the distinction between these vowels. Are these programs, with their prescriptive vowel distinctions, to be accepted as the one and only model for intelligible American English speech?

Clearly, my wife and I both have correct pronunciation in English, although we differ from each other. When teaching pronunciation, we take a practical and appropriate response to the variation. We model what we say but accept either variant as correct. By using this principle of acceptability, our students are given the same respect we give each other. Students learn that that both variants are acceptable, though one variant may be used more frequently in certain areas of the U.S., and that their speech will be clear if it is within the acceptable range for the listener. This is the approach taken by a recent textbook, Pronunciation Plus (Hewings and Goldstein, 1999), which presents this vowel difference as normal dialect variation in North American English.

What about other variations that are not so widespread or that do not have the same social status? I once was discussing training teachers with a speech pathology professor at North Carolina State University. She said that she taught all her students from the southern U.S. to say a very clear /ai/ sound for words like I, hide, and why. When I asked why, she explained that the normal southern pronunciation of those words /a:/, /ha:d/, /wa:/ was not proper and that the students needed to use a correct model for their clients. I (who am not from the South) objected that southern English was indeed a proper model. She (who is from the South) disagreed. The conversation moved onto other topics, and her /a:/ monitor turned off. Within 10 words she was consistently saying /a:/ for I and /ba:/ for bye. Her speech was perfect southern educated English -- and perfectly comprehensible and acceptable.

Clearly, decisions about acceptability in pronunciation must be tempered by a knowledge of students' goals and the context of instruction. Sociocultural judgments about which variants are acceptable will not always be the same. Given that my students are in the South, I would never correct their use of /a:/ where it is the variant used by educated speakers, even though I do not use that variant personally. By implication, it seems that teachers of pronunciation should be far more willing than they are to question sociolinguistic judgments in determining what to teach. Many variants that are not part of the prestige dialects are easily understood and relatively neutral in sociolinguistic effect and should not be marked as targets for correction.

Practically speaking, in most contexts, simply being aware of dialect differences can help teachers and students make wise decisions about variation. In a recent article about learners' pronunciation rights, Gomes de Matos and Celce-Murcia (1999) echo this position when they say learners have "the right to be exposed to different dialects" and "the right to be sensitized to variations in speech" (pp. 1-2). Variation exists everywhere in pronunciation, and teachers must come to terms with it in order to avoid burdening students with unimportant distinctions. It really does not matter whether our students say Barney or Bonnie if the interlocutors know who is being spoken about, as both are acceptable variants. Intelligibility is not affected.

In conclusion, I make three recommendations for preparing teachers to teach pronunciation:

1. Teachers should be aware of the most common systematic variations of pronunciation in common educated dialects of English. This is a teacher education issue in that TESOL degree programs should help teachers become aware of both their own pronunciation and that of speakers of other dialects.
2. Teachers should be encouraged to model their own dialect but should allow students to use any acceptable variation in pronunciation. Like the first recommendation, this one has implications for how teachers are educated.
3. Students should have access to more pronunciation models than they can get from the teacher or the textbook alone. This recommendation has implications for materials development.

Although it remains very difficult, in a general sense, to determine what errors or combinations of errors impede intelligibility, training teachers to recognize the range of features of their own and other major dialects will make them better able to concentrate on the speech characteristics, prosody, and sounds that can cause misunderstanding and lack of clarity while not burdening students with distinctions that teachers do not require of each other.

References

Gomes de Matos, F., and Celce-Murcia, M. (1999). A new frontier in TESOL pedagogy: Learner's pronunciation rights. Materials Writers Interest Section Newsletter, 12(2), 1-2.

Hewings, M., and Goldstein, S. (1999). Pronunciation plus: Practice through interaction: North American English. Cambridge: Cambridge University Press.

John M. Levis is an assistant professor of ESL in the Department of Foreign Languages and Literatures at North Carolina State University in Raleigh, North Carolina.