Changing Contexts and Shifting Paradigms in Pronunciation Teaching
Changing Contexts and Shifting Paradigms in Pronunciation Teaching
Changing Contexts and Shifting Paradigms in Pronunciation Teaching
369
COMPETING IDEOLOGIES
More fundamentally, pronunciation research and pedagogy have long
been influenced by two contradictory principles, the nativeness principle
and the intelligibility principle. The nativeness principle holds that it is
both possible and desirable to achieve native-like pronunciation in a
foreign language. The nativeness principle was the dominant paradigm
in pronunciation teaching before the 1960s, but its influence was rapidly
diminished by research showing that nativeness in pronunciation appeared to be biologically conditioned to occur before adulthood
(Lenneberg, 1967; Scovel, 1995), leading to the logical conclusion that
aiming for nativeness was an unrealistic burden for both teacher and
learner. Despite extensive ongoing research into a critical period for
acquiring pronunciation, in practice very few adult learners actually
achieve native-like pronunciation in a foreign language. Factors such as
motivation, amount of first language (L1) use, and pronunciation
training are positively correlated with more native-like pronunciation,
but none of these other factors seems to overcome the effects of age
(Flege & Frieda, 1995; Moyer, 1999).
Although an overwhelming amount of evidence argues against the
nativeness principle, it still affects pronunciation teaching practices.
Popularly, the principle drives the accent reduction industry, which
implicitly promises learners that the right combination of motivation
and special techniques can eliminate a foreign accent. In language
classrooms, it is common for learners to want to get rid of their accents
(as one of my recent students expressed it). Many teachers, especially
those unfamiliar with pronunciation research, may see the rare learner
who achieves a native-like accent as an achievable ideal, not an exception.
The second principle is the intelligibility principle. It holds that
learners simply need to be understandable. The intelligibility principle
recognizes that communication can be remarkably successful when
foreign accents are noticeable or even strong, that there is no clear
correlation between accent and understanding (Munro and Derwing,
1999), and that certain types of pronunciation errors may have a
disproportionate role in impairing comprehensibility.
The intelligibility principle implies that different features have different effects on understanding. Instruction should focus on those features
370
TESOL QUARTERLY
that are most helpful for understanding and should deemphasize those
that are relatively unhelpful. This assumption of differential importance
is evident in most intelligibility-based arguments for pronunciation
instruction. For example, the longstanding belief that instruction should
focus on suprasegmentals (e.g., Avery & Ehrlich, 1992) assumes that a
focus on these features leads to better and quicker speaker intelligibility
than a focus on segmentals.
Jenkinss (2000) lingua franca core (LFC), a proposal for intelligibilitybased pronunciation instruction, shares this assumption about intelligibility, albeit with an important difference in communicative context.
Jenkins argues that her approach supports EIL (also called ELF, or
English as a lingua franca) communication, but her recommendations
have caused pronunciation teachers in all contexts to revisit their beliefs
about intelligibility and the primacy of suprasegmentals. Dauer (this
issue) provides an ESL response to the LFC, both praising its renewed
emphasis on segmentals and arguing that its de-emphasis on suprasegmentals will not serve learners well, given that the boundaries between
ESL and EIL communication are more fluid than the LFC suggests.
The LFC also raises issues for EFL contexts, where its recommendations would seem to be most at home. However, because students in EFL
classrooms share the same L1, they converge toward second language
(L2) pronunciation that is heavily influenced by the L1. Thus, the
documented tendency of different L1 speakers to converge toward more
internationally intelligible pronunciation ( Jenkins, 2000) does not seem
to operate in EFL contexts. Walker (this issue) describes a technique
used successfully to help learners who share the same L1 converge
toward pronunciation that will be more intelligible in EIL communication.
Despite the current dominance of intelligibility as the goal of pronunciation teaching, both the nativeness and intelligibility principles continue to influence pronunciation in the language curriculum, both in
how they relate to communicative context and in the relationship of
pronunciation to identity.
371
found Estuary English speech, which they are more likely to encounter
in England, to be often unintelligible. Deterding argues that pedagogical
reliance on prestige models is counterproductive for learners ability to
understand normal speech.
The intelligibility principle carries a sensitivity to context. Intelligibility assumes both a listener and a speaker, and both are essential elements
for communication. Levis (in press) describes the context sensitivity of
intelligibility in terms of a native speakingnonnative speaking (NS
NNS) listener-speaker matrix for assessment (Figure 1). The four quadrants reflect different aspects of intelligibility and suggest different
priorities for language teaching.
Quadrant A has NS speakers and listeners and is usually assumed to be
the standard for successful communication. This assumption implies that
the speakers varieties are mutually intelligible, although it is not clear
just how mutually intelligible native varieties actually are. Research has
shown that understanding in NS communication is often more complex
than one would expect (e.g., Cutler, Dahan, & van Danselaar, 1997).
Quadrant B, with NS speakers and NNS listeners, is a normal configuration for language teaching in an ESL context. It is also the norm for most
language teaching beyond ESL contexts, in which print and audio
materials are based on NS models. However, the ways in which NNS
listeners actually decode and interpret NS speech is not completely clear.
Quadrant C reflects most current research on intelligibility, where NNS
speakers communicate with NS listeners. This model assumes that NSs
already have the ability to communicate and makes NNSs responsible for
communicative success. Quadrant D, where both speakers and listeners
FIGURE 1
Speaker-Listener Intelligibility Matrix (Levis, in press)
LISTENER
Native
Speaker
Native Speaker
Nonnative Speaker
A. NSNS
B. NSNNS
C. NNSNS
D. NNSNNS
SPEAKER
Nonnative
Speaker
372
TESOL QUARTERLY
InnerCircle
SPEAKER
Outer
Circle
Expanding
Circle
Outer Circle
(OC)
ICIC
Expanding Circle
(EC)
ICEC
1. ICOC
(NSNS)
2. OCIC
(NSNNS)
3. OCOC
ICIC
4. OCEC
ECEC
5. ECOC
(NNSNS)
(NNSNNS)
373
The four italicized corners of the matrix reflect the same communicative possibilities shown in Figure 1, but the bolded sections of the matrix
are relatively unexplored. Both Quadrant 1 and Quadrant 2 include
inner-circle and outer-circle interlocutors, and in both cases, the standardized nature of inner-circle Englishes may shift the perceived responsibility for being intelligible to outer-circle interlocutors (Bamgbose,
1998). At this juncture, the communicative context becomes crucial. In
U.S. university settings, for example, graduate teaching assistants from
outer-circle countries such as India are routinely tested for spoken
English proficiency, even when their English proficiency is otherwise
indistinguishable from inner-circle graduate students. It seems evident
that such testing is conducted because outer-circle speakers have unfamiliar accents, not a lower proficiency in English. In an outer-circle
setting, however, an inner-circle interlocutor is more likely to recognize
the validity of the outer-circle accent.
Quadrant 3, in which outer-circle speakers are interlocutors, likely has
the same kind of variation in intelligibility as NSNS communication.
Outer-circle speakers will likely have the same difficulties with unfamiliar
accents and registers that inner-circle speakers have with unfamiliar
dialects.
Quadrants 4 and 5 include outer-circle and expanding-circle interlocutors. These interactions often occur in contexts without inner-circle
speakers. As a result, pronunciation issues may cause breakdowns in
communication similar to those described by Jenkins (2000), who found
that pronunciation caused a loss of intelligibility in NNSNNS communication. It would be surprising, however, if the two quadrants had the
same bottom-up processing difficulties discussed by Jenkins. In general,
the proficiency of outer-circle speakers is more like that of inner-circle
speakers than that of expanding-circle speakers, for whom English is a
foreign language. Thus, an outer-circle listener and an expanding-circle
speaker, as in Quadrant 5, are more likely to negotiate intelligibility
using context or top-down knowledge of English than are an expandingcircle listener and an outer-circle speaker, as in Quadrant 4, where
bottom-up processing constraints are likely to be more severe.
IDENTITY
Both Figure 1 and Figure 2 have a weakness: In judgments of intelligibility, they ignore, on the positive side, the role of language identity,
and on the negative side, language attitudes. Accent is influenced not
only by biological timetables but also by sociolinguistic realities. In other
words, speakers speak the way they do because of the social groups they
belong to or desire to belong to. The role of identity in accent is perhaps
374
TESOL QUARTERLY
375
CONCLUSION
Currently, pronunciation theory, research, and practice are in transition. Widely accepted assumptions such as the primacy of suprasegmentals,
the superiority of inner-circle models, and the need for native instructors
have been rightly challenged. ESOL professionals are recognizing that
judgments of intelligibility involve nonlinguistic as well as linguistic
factors, and that even completely intelligible pronunciation may be
evaluated negatively. Decisions about adjusting accent are not value free
because accents are intimately tied to speaker identity and group
membership. Increasing evidence also shows that the context of instruction directly affects how pronunciation should be addressed. Users of
English who interact professionally in inner-circle contexts may need to
adjust to an inner-circle model, but English users in the outer or
expanding circle may find that inner-circle models are inappropriate or
unnecessary ( Jenkins, 2000). These findings indicate that teaching
pronunciation is only partially a pedagogical decision, and that old
assumptions are ill-suited to a new reality.
THE AUTHOR
John M. Levis is an associate professor of TESOL and applied linguistics at Iowa State
University, Ames, Iowa, USA. His research interests include the intelligibility of
spoken language, intonation, English pronunciation, and varieties of English. He has
published articles in numerous prestigious journals.
376
TESOL QUARTERLY
REFERENCES
Avery, P., & Ehrlich, S. (1992). Teaching American English pronunciation. Oxford,
England: Oxford University Press.
Bamgbose, A. (1998). Torn between the norms: Innovations in World Englishes.
World Englishes, 17, 114.
Celce-Murcia, M., Brinton, D., & Goodwin, J. (1996). Teaching pronunciation: A
reference for teachers of English to speakers of other languages. New York: Cambridge
University Press.
Cutler, A., Dahan, D., & van Donselaar, W. (1997). Prosody in the comprehension of
spoken language: A review. Language and Speech, 40, 141201.
Derwing, T., & Rossiter, M. (2003). The effect of pronunciation instruction on the
accuracy, fluency and complexity of L2 accented speech. Applied Language
Learning, 13(1), 117.
Flege, J., & Freida, E. (1995). Amount of native-language (L1) use affects the
pronunciation of an L2. Journal of Phonetics, 25, 169186.
Hahn, L. (2004). Primary stress and intelligibility: Research to motivate the teaching
of suprasegmentals. TESOL Quarterly, 38, 201223.
Jenkins, J. (2000). The phonology of English as an international language. Oxford,
England: Oxford University Press.
Kachru, B. (1986). The alchemy of English: The spread, functions, and models of non-native
Englishes. New York: Pergamon Press.
Lenneberg, E. (1967). Biological foundations of language. New York: Wiley.
Levis, J. (1999). The intonation and meaning of normal yes-no questions. World
Englishes, 18(3), 373380.
Levis, J. (in press). Pronunciation and the assessment of spoken language. In
R. Hughes (Ed.), Spoken English, applied linguistics and TESOL: Challenges for theory
and practice. London: Palgrave Macmillan.
Lippi-Green, R. (1997). English with an accent: Language, ideology and discrimination in
the United States. New York: Routledge.
Milroy, J., & Milroy, L. (1985). Authority in language: Investigating language prescription
and standardization. London: Routledge & Kegan Paul.
Morley, J. (1987). Current perspectives on pronunciation. Alexandria, VA: TESOL.
Morley, J. (1991). The pronunciation component in teaching English to speakers of
other languages. TESOL Quarterly, 25, 481520.
Moyer, A. (1999). Ultimate attainment in L2 phonology: The critical factors of age,
motivation and instruction. Studies in Second Language Acquisition, 21, 81108.
Mugglestone, L. (1995). Talking proper: The rise of accent as social symbol. Oxford,
England: Clarendon.
Munro M., & Derwing, T. (1999). Foreign accent, comprehensibility, and intelligibility in the speech of second language learners. Language Learning, 49 (Supp. 1),
285310.
Pennington, M., & Ellis, N. (2000). Cantonese speakers memory for English
sentences with prosodic cues. The Modern Language Journal, 84(3), 372389.
Scovel, T. (1995). Differentiation, recognition, and identification in the discrimination of foreign accents. In J. Archibald (Ed.), Phonological acquisition and phonological theory (pp. 169181). Hillsdale, NJ: Lawrence Erlbaum.
Wolfram, W., & Schilling-Estes, N. (1998). American English. Malden, MA: Blackwell.
377