Wayne Dickerson, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA

Levis (2018) begins his analysis of research into intelligibility with this statement: “Intelligibility is widely agreed to be the most important goal for spoken language development in a second language – both for listening and speaking – no matter the context of communication” (p. 15). For the purposes of this article, let us assume that you agree with this goal for teaching pronunciation. Furthermore, you are in the market for a textbook developed to improve your students’ intelligibility. What would such a textbook look like? I submit that it would closely follow a framework that describes what a listener must hear in a speaker’s phrases to judge them as intelligible. It is a framework that organizes that text and informs every lesson.

The framework proposed here is a higher order construct. It is nevertheless grounded in the production and perception of speech as complementary roles based on a shared understanding of how to encode and decode speech for intelligibility. The central tenet of teaching for intelligibility is the Golden Rule for Communication: Speak to listeners as listeners would wish to be spoken to. Five tests or indicators can confirm that our instruction prioritizes the listener's perspective. Teaching learners to speak this way makes the listener’s perspective a priority.

Indicator 1: Choice of Speech Style as the Target

Spontaneous speech is the currency of everyday verbal exchanges and the basis for all other oral language. This choice of pronunciation target helps to identify key characteristics of language that speakers and listeners find most comfortable. Among its defining features is a phrase length that is constrained by the practical capabilities of listeners to catch and interpret a stream of airborne sound waves. Fieldwork has shown that most spontaneous phrases are three to four words long. Furthermore, the frequency of phrases longer than seven words decreases sharply (Pickering, 2018).

From the start, instruction should teach learners about the seven-word phrase limit, its importance to listeners, and how to segment speech accordingly. Furthermore, if learners are to become users of listener-centered language, all phrases in their practice materials should model the one-to-seven-word range.

Indicator 2: Bipolar Signals of Importance

English speakers use phonetic tools to draw listeners’ attention to important parts of a phrase. In this excerpt from a conversation between two colleagues, Speaker B wants Speaker A to notice the phrase, the Blue Bonnet State (Texas). So Speaker B adds phonetic length to, and changes the pitch of, the [uw] of Blue, making it a pitch accent. Speaker B also minimizes the length and drops the pitch and volume of the vowels surrounding [uw]—to the and Bonnet State.

Speech Example A

A: So where do you vacátion these days?

B: Well, our first trip’s usually in Márch | to the Blúe Bonnet State.

How does modulating the suprasegmentals of [uw] and nearby vowels signal importance? The answer is, indirectly. On hearing the [uw] of Blue, Speaker A correctly interprets it pragmatically, not literally, and follows pragmatic rules that say (a) “Here is a cue of importance” and (b) “Don’t look for meaning in [uw], nor necessarily in Blue, nor necessarily in Blue Bonnet, but in the largest construction that [uw] is a part of, namely, in the compound noun Blue Bonnet State.”

Speakers A and B unconsciously engage pragmatic rules that native English speakers and listeners play by. Learners of English, unaware of English norms, play by different rules and listen for different cues. Consequently, they do not notice these suprasegmental contrasts nor recognize their significance to communication. This gap in their knowledge and skill makes them poor communicators in English.

Producing, noticing, and interpreting contrasts in vowel suprasegmentals as signals of importance are foundational skills that must be taught explicitly and early. Furthermore, because English, more than most languages, requires a larger contrast between a pitch accent and surroundings vowels, it is not enough to maximize vowel cues. To meet listeners’ expectations of what a prominence sounds like, instruction should always emphasize maximizing and minimizing vowel cues.

If you have taught pronunciation for a while, you are likely to recognize that I have been describing what many pronunciation textbooks call the focus, a pragmatic term associated with the primary pitch accent. It is the product of applying pragmatic rules to interpret the phonetic cues of prominence in a phrase.

Pronunciation textbook writers use the concept of focus to answer the key question we began with: What must English listeners hear in speakers’ phrases to judge them as intelligible? They explain that speakers’ primary pitch accent directs listeners’ attention to the most important meaning of a phrase.

Indicator 3: The Expanded Focus Within the Two-Peak Profile

The definition of focus as derived from the primary pitch accent was uncontested until spontaneous-speech researchers discovered a second meaningful pitch accent in many phrases (Wells, 2006, pp. 8, 209ff). A circumflex identifies this accent:

Speech Example B

A: So └whêre do you vacátion┘ these days?

B: Well, └our fîrst trip’s usually in Márch┘, | └to the Blúe Bonnet State┘.

Departing from the classic TESOL hypothesis that every content word in a phrase carries an accent, we now know that another characteristic of nearly all unrehearsed phrases is that they have only one or two pitch accents. That is, some content words are unaccented, like trip and usually in Well, our fîrst trip’s usually in Márch.

We call this model of rhythm the two-peak profile (see Figure 1). The metaphor of a mountain range in profile led us to call the first pitch accent the anchor peak and the second the primary peak. All nonpeak words are in the valleys of this model (Dickerson & Hahn, in press).

Figure 1. The two-peak profile.

This more accurate conception of English rhythm led to the conclusion that, because pitch accents signal important pragmatic meaning, the cue for the focus actually begins with the first pitch accent (the anchor peak, if there is one) and ends with the last pitch accent (the primary peak). Furthermore, to make sense of these two peaks as a single thought, the expanded focus must include any unstressed syllables before and after these peaks and all words in between. A phrase with only a primary peak includes any unstressed syllables before and after it. Syntax provides the necessary cohesion (Pickering, 2018). The expanded focus is bracketed (└ ┘) in Speech Example B. Compared to a focus based on the primary peak alone in Speech Example A, the expanded focus makes the core meaning of each phrase much clearer. We are now closer to defining what listeners must hear in speakers’ phrases to judge them as intelligible.

In this framework, the presentation of the two-peak profile starts near the beginning of instruction so learners have the maximum exposure to and practice with the rhythm of spontaneous speech. We start with the anchor-placement rule. To give learners practice placing the anchor, we mark the primary peak in their materials. Then, when they practice phrases with the anchor they have predicted, they produce the complete two-peak profile.

Because many of the phrases they use have a valley between the peaks, instruction on how to handle valley syllables comes next.

After that, we show learners how to predict the primary peak to signal the presence of new information, contrasts, and emphasis. With this addition, learners can predict the entire two-peak profile on their own.

Indicator 4: Valley Compression for Listeners

Valleys deserve special attention because they play a critical role in intelligibility. Valleys are characterized by extreme compression, another feature of spontaneous speech. For many years, we believed that we rush through valleys to keep peaks coming at a regular pace. Then research showed definitively that English is not stress-timed (Cauldwell, 2002). So why hurry? Of the three potential valleys in the two-peak profile, only the one between the peaks is consistently part of the extended focus. It seems likely that speakers compress its syllables to help listeners understand the focus.

Semantically, the expanded focus is a single thought. If a listener is to grasp it as a meaningful unit, the speaker must deliver it quickly and in an unbroken string. Compression of valley syllables speeds up this string so the listener can catch it in one go. Failing to compress syllables, introducing hesitations, and adding more peaks lengthen this valley and strain the listener’s memory. Any delay in understanding one phrase can mean that the listener will miss all or part of the next phrase, damaging intelligibility.

We compress valleys in a variety of ways—through assimilation, vowel reduction, trimming consonant and vowel sounds, and linking consonant and vowel sounds to each other. If our goal is to meet listeners’ needs, then none of these compression techniques is optional for speakers. They require instruction as soon as learners begin creating interpeak valleys.

Indicator 5: Intonation to Interpret the Expanded Focus

Besides the words of the expanded focus, intonation makes its own contribution to intelligibility. The intonation pattern associated with the primary peak is especially important. The primary peak announces the start of the final intonation pattern, alerting listeners to significant upcoming pitch information. By tuning in to the complete intonation pattern, listeners learn how to interpret the expanded focus: Is the speaker concluding or preparing to say more? Is the speaker making a statement or a query? Intonation is part of what listeners must hear in order to understand a speaker’s phrase.

Conclusion

The framework for intelligibility-based pronunciation instruction presented here is a higher order pragmatic structure. It encompasses the words of the extended focus and its complete intonation pattern. It describes what listeners are listening for in a speaker’s phrases and helps teachers and students keep their eyes on the goal of all their pronunciation efforts. It also motivates all work to control lower level building blocks of the framework, such as consonant and vowel articulations, consonant clusters, key allomorphs, and the stress of words and constructions. They, too, are critical for the intelligibility of a phrase.

References

Cauldwell, R. (2002). The functional irrhythmicality of spontaneous speech: A discourse view of speech rhythms. Apples 2, 1–24.

Dickerson, W., & Hahn, L. (in press). Speechcraft: Discourse pronunciation for academic communication (2nd ed). Ann Arbor, MI: University of Michigan Press.

Levis, J. (2018). Intelligibility, oral communication, and the teaching of pronunciation. Cambridge, England: Cambridge University Press.

Pickering, L. (2018). Discourse intonation: A discourse-pragmatic approach to teaching the pronunciation of English. Ann Arbor, MI: University of Michigan Press.

Wells, J. (2006). English intonation: An introduction. Cambridge, England: Cambridge University Press.