is when to introduce pronunciation, especially since many of the available resources are intended for intermediate or advanced level students. Darcy et al. (2012) suggest that pronunciation instruction should be included “in the first year of extensive exposure to the L2” (p. 94) during the period which Derwing and Munro (2015) have labelled the Window of Maximal Opportunity. Derwing and Munro’s longitudinal study of naturalistic L2 phonological development (in the absence of instruction) indicated that most change happens in the early part of the first year of massive exposure. Thus, it makes sense to capitalize on what Darcy et al. (2012) called the time of greatest pronunciation malleability.
If pronunciation is to be initiated when learners are first intensively exposed to the L2 (as is the case, for instance, with immigrants to a majority English-speaking country), then materials should be available at all levels. Zielinski and Yates (2014) offer sound reasons for implementing pronunciation instruction right from the start, including helping learners with their self-confidence, which, in turn, could lead to increased exposure. Their book of explanations and activities, Give It a Go (Yates & Zielinski, 2009), is freely available and is intended for teachers who have had little or no experience teaching pronunciation: http://www.ameprc.mq.edu.au/__data/assets/pdf_file/0006/276135/interactive_sm.pdf
Intensive instruction at some point, even if it does not continue, also appears to affect future comprehensibility and fluency, suggesting that pronunciation-focused instruction has long-term effects. French et al. (2020) report on an intensive five-month ESL course taken by French L2 learners of English at the beginning of high school with their abilities four years later. Compared to a control group that had not received intensive ESL instruction, the intensive learners had significantly higher ratings of both fluency and comprehensibility. There was no difference with ratings of accentedness.
Although nascent, there are now some studies which probe the success of individual activities on learners’ productions. For instance, Foote and McDonough (2017) examined the efficacy of shadowing recorded dialogues (imitating a model) and found significant improvement. The same learners also produced a picture narrative at three times over an eight-week period, and were subsequently judged by listeners to have improved in both comprehensibility and fluency. The authors attribute these changes to the shadowing activity. More investigations of stand-alone activities such as this will be a benefit to teachers and students alike (see Murphy & Baker, 2015) for a historical review of empirical research in pronunciation teaching).
Pedagogical Implications
Pronunciation teaching and learning includes, like any area of language teaching, a wide variety of activity and exercise types, from minimal pair exercises to discrimination listening exercises, to guided and communicative exercises. How do we know which activities are best suited to promote improvement, and what kind of improvement they promote? It is largely left up to the instructor to determine which ones will work best for their students’ particular needs, taking into consideration not only the type and severity of their problems, but also their age, the nature of the course, their prior education levels, and other aspects of the context. Some activities are relatively simple, while others are complex and require considerable preparation. Here is where research can help the teacher in making choices. In the next section, we address some empirically tested activities that are rarely found in pronunciation texts but which have been shown to have positive results. We have included only suprasegmental activities focused on global speaking and listening skills, in light of the fact that a broad range of segmental resources are more widely available.
Contrastive Stress
Levis and Muller Levis (2018) conducted an experiment with learners of English with a focus on contrastive stress. Over a three-week period, the instructors provided explicit instruction regarding contrastive stress (e.g., pointing out “lexicogrammatical patterns that are used to compare and contrast (e.g., other, first/second, on the left/right)” (p. 149) and offered auditory examples. Students practiced by producing sentences based on pictures that differed (e.g., two houses with different sizes, colors, number of stories, and number of windows, examples of which are available in their open-access publication). Pre- and post-picture descriptions were recorded, and were the basis of a listening study in which 20 undergraduate students rated the descriptions for comprehensibility and fluency from the experimental group and a control group on 9-point scales. The authors found that comprehensibility was improved in the experimental group, although fluency did not change in either group. A simple focus on contrastive stress over a relatively short time period made the speech of these learners easier to understand.
Sitcom Cycle
For many years, expert practitioner Janet Goodwin used clips from sitcoms to aid her students’ awareness of English pronunciation (for details, see Goodwin, 2005). She would take a scene from a popular TV show, such as Seinfeld or Friends, and would ask her students to break into small groups and transcribe the segment. This required multiple listenings. She then had the students mark word stress and major intonation patterns, getting them to focus on prosody. Again, this took multiple listenings. After each of these rounds, she went through the clip with the students, correcting any errors and asking them to listen again to confirm that they could perceive the productions accurately. The next step was to have the students perform the scene, incorporating the body language used by the characters, and using their marked transcriptions as the script. This arc of activities serves several purposes. First, students tend to accept sitcom language as “real” speech, despite the fact that it is scripted, and therefore tends to be more grammatical and coherent than actual “real” speech. The multiple listenings, first for general understanding (transcription) and next for prosody, offer the learners the repetition they need, but with an achievable goal. The teacher can flag areas of difficulty and do some extra work on those areas before moving on. The performance aspect of the activity is carried out only after the learners have heard the clip many, many times, in the company of their classmates. Thus, it is nowhere near as threatening as a speaking activity that is imposed out of the blue. Goodwin video-recorded the learners’ performance, and the whole class would then comment on the strengths and areas for improvement. This intensive focus on a short (2- to 3-minute) clip can lead to significant awareness raising and improvement in production. Goodwin’s activity entails a great deal of recycling of the same material. Repeated passes at the same activity for different purposes can be highly beneficial.
Jazz Chants
Another old, but useful, resource is Jazz Chants (Graham, 1978), which can be employed quite differently from the author’s original intention, which was a focus on grammar in spoken English, along with affective intonation. Prosodic elements can be practiced by going through a chant once with the learners listening, and then having them listen again and tap or clap the beats. The instructor can then substitute a single syllable such as “da” for all of the syllables and have the students first listen and then repeat the chant, using “da” throughout (this eliminates any struggles with consonant clusters or other segmental difficulties, which should not be the focus of a prosodic lesson). The students can be asked to count the number of syllables in each line, and then count the number of stressed syllables. For some learners it is a surprise to find that lines that are meant to be of equal length can vary by several syllables yet share the same stress pattern. Several Jazz Chants are available on the internet for teachers who may not be able to access the printed resource, or who are unsure as to the rhythm of each chant.
TED Talks
McGregor et al. (2016) pointed out that most pronunciation textbooks treat intonation in a mechanistic way that belies its relationship to context. They chose to analyze aspects of intonation (prominence, salience, tone choice) employed in a Ted Talk using both a perceptual
