We’ve heard for some time now that pronunciation is often neglected in both the language and teacher training classrooms. Despite the profuseness of this claim and the awareness of this fact, it nonetheless remains true. Pronunciation is undertaught, likely because of inadequate training, but also because a lack of understanding of how it should be taught, and perhaps a healthy skepticism about its effectiveness. In 2012, Kazuya Saito helped shine some light on this issue with a research synthesis of studies measuring the effectiveness of pronunciation instruction. They found an overall positive effect of pronunciation intervention with a number of considerations for the classroom.
The research synthesis (not meta-analysis) looked at 15 quasi-experimental studies of L2 pronunciation instruction published after 1990. These studies varied in terms of type of pronunciation features, type of instruction, and type of outcome measures:
- type of feature:
- segmental, e.g. /ɹ/ or a range of sounds
- suprasegmental, e.g. intonation, rhythm, syllable structures
- types of instruction
- focus-on-form (FonF), defined as a focus on forms in controlled communicative context
- focus-on-formS (FonFS), defined as a focus on form only in controlled activities such as drills and repetition)
- focus-on-meaning (FonM), defined as a focus solely on communication – this often served as the control group
- type of outcome measure
- controlled response (CR), such as reading a word list in order to elicit target sounds
- free response (FR), such as a picture description that measured a sound during spontaneous production
Each one of the studies included reported positive results for their respective pronunciation interventions except for two. In these studies, the treatment time was extremely short (15-30 mins compared to longer treatments of several hours or 11 or more weeks). Studies that included a FonM control group showed no improvement for this group.
For type of feature, while all studies that focused on segmentals (contained significant improvements, among those that contained free response outcomes (4), only one showed improvement in spontaneous usage. That was a study that focused solely on English /ɹ/ through FonF, with a 4-hour treatment (Saito & Lyster, 2012). The same was true of suprasegmentals, with 1 showing spontaneous improvement after 11 weeks of FonF (Derwing et al. 1998).
For type of instruction, two out of six FonF studies showed improvement on free response tasks whereas no FonFS studies showed improvement on anything but controlled responses (even though three included free response tasks).
First, this research synthesis demonstrates that pronunciation instruction, given adequate time, can have a positive effect on the comprehensibility of language learners. (The focus of all of these studies was comprehensibility not accent reduction nor sounding like a “native”). The authors assert that an ideal time for instruction would be about 1 semester (typically 16 weeks in the US). They also recommend that both segmentals and suprasegmentals can be focused on to the degree that they are important and relevant for the students’ L1, proficiency level, and target language context. In addition, the authors suggest a focus on form in communicative context (FonF) because these studies showed the greatest improvement, especially in free response, came with FonF, and this improvement was often maintained after the study.
In general, this means that pronunciation instruction must be pre-planned and embedded in communicative tasks. Saito supports this with research that found FonF practiced in context can create important sound-meaning relationships. FonF can also “promote proceduralization” and offer opportunities for “contextualized repetitive practice”. Saito points out that communicative FonF can take several shapes, including one or more of these:
- pronunciation-focused tasks – tasks that are designed to elicit or force certain pronunciation features
- corrective feedback during tasks
- metalinguistic information prior to FonF
In summary, pronunciation instruction when explicit, pre-planned, embedded in context, and sustained over time, can lead to lasting improvements in comprehensibility.
Where to Start?
The goal of this post was to simple ask if pronunciation instruction, and if so, what common elements make it so. I hope this post clearly answered those questions. Moving on from investigation into action, some may wonder, knowing that instruction is effective, how to both teach and integrate pronunciation. I hope the resources below provide some guidance:
- Integrating Pronunciation Across the Curriculum – PPT from American English
- Pronunciation in the Classroom: The Overlooked Essential – Free ebook from TESOL
- Overview of Celcia-Murcia’s Framework for teaching pronunciation
- Teaching Pronunciation Using the Prosody Pyramid
- Making pronunciation an integral part of your classroom practice (IATEFL session)
- Research into practice: revisiting some ‘old-fashioned’ notions in pronunciation teaching (IATEFL session)
- Integrating Pronunciation (IATEFL Poland)
- Everything You Wanted to Know about Teaching Pronunciation
Derwing, T., Munro, M., & Wiebe, G. (1998). Evidence in favor of a broad framework for pronunciation instruction. Language Learning, 48, 393–410.
Saito, K. (2012). Effects of instruction on L2 pronunciation development: A synthesis of 15 quasi‐experimental intervention studies. TESOL Quarterly, 46(4), 842-854.
Saito, K., & Lyster, R. (2012). Effects of form-focused instruction and corrective feedback on L2 pronunciation development of /ɹ/ by Japanese learners of English. Language Learning, 62, 595–633. doi:10.1111/j.1467-9922.2011.00639.x