Finding contributory factors for enhancing communicative competency continues to be a major interest and a challenge for ESL/EFL instructors. Many thus far have concluded that the suprasegmental level (stress, length, tone, intonation, overall rhythm, and timing), which is collectively referred to as prosody, plays a more significant role than the segmental level (consonants and vowels; e.g., Anderson-Hsieh, 1996; Edwards & Strattman, 1996). The rhythm and timing of an utterance created by emphasis or sentence stress in particular are claimed to serve as signals to provide syntactic, semantic, and discourse information to listeners, who attend to the stressed syllables to determine essential information (Major, 2001). Is it possible to create speech rhythm by adjusting the application of emphasis? Can this adjustment be measured?

When speech is examined acoustically, stressed syllables are produced with physiological force, which can be identified as longer duration, higher pitch, or greater loudness in a perceptual dimension. With the advent of computerized instruments, the auditory data of duration, pitch, and loudness can be synthesized and analyzed acoustically by means of time, fundamental frequency (F0), and intensity.

Phonological Challenges to Japanese English Language Learners

The application of sentence stress to create a good speech rhythm can be a major challenge for Japanese English language learners (ELLs) when the total dissimilarities of the Japanese and English languages are considered. Japanese is mora-timed. The mora is the smallest unit of timing, all of which are pronounced with equal length and loudness (Major, 2001). In contrast, English is stress-timed. The durations between the primary stressed syllables are almost equal in length regardless of the number of unstressed syllables in between. In order to create the regular rhythm from stressed syllables to stressed syllables, unstressed syllables are shortened by vowel reduction (i.e., schwa /ə/; Major, 2001).

Previous Studies

Anderson-Hsieh (1996) claims electronic visual displays show little pattern differences between stressed and unstressed syllables of Japanese ELLs compared to those of native English speakers (NESs). Wennerstrom (1994) concludes that Japanese ELLs’ speech is without much pitch variation or flat compared to NESs. Wennerstrom (2001) also argues in her study of narratives that pitch is used most commonly to reflect a speaker’s priorities to express emotions.

Statement of Problems

Although some findings indicate that Japanese ELLs demonstrate less noticeable stress differences, few have provided objective evidence using the acoustical measurements (i.e., vowel duration, F0, and intensity).

Purpose of the Study

The purpose of this study was to investigate the Japanese ELLs’ ability to change speaking rhythms by applying sentence stress through answering the following two questions.

1. Is there a difference in the acoustic measurements of vowel duration, F0, and intensity of the predetermined target words of the scenario in the general reading between Japanese ELLs and NESs?
2. Is there a difference in these acoustic measurements between Japanese ELLs’ first reading and second reading of the same scenario with an additional instruction to apply feelings?

METHOD

Participants

The ELLs were 30 Japanese adults (15 males and 15 females) who were born and raised in Japan and lived in Kansas, in the United States, at the time of the study. Their ages ranged from 21 to 41 years, with a mean age of 26.7 years for males and 32.8 years for females. Their length of residence in the United States ranged from 1:00 (years:months) to 15:07, with a mean length of 3:02 for males and 6:11 for females.

The NESs were 30 Americans (15 males and 15 females) who lived in Kansas at the time of the study and grew up in the Central Midland region of the United States. They were age-matched with the Japanese ELLs within a 2-year difference.

Reading Material

The reading material was a role-play scenario presented as a voicemail message to a friend. See Table 1. The scenario included five pre-determined target words for possible emphasis application to show the speakers’ feelings. The words were all adjectives in consonant-vowel-consonant combination. Both voiced and voiceless stops and fricatives were used as well as /n/.

Table 1

Scenario

Note. Bold = target words. The target words in the actual narrative provided to the speakers were not in bold so that the speakers would not be influenced by any additional marks.
 

Procedures

Recordings. The participants’ speech samples were recorded individually upon reading the scenario with the instruction to read it as they would generally do. They were not told which words to emphasize. Only the Japanese ELLs performed the second task of reading with feeling.

Acoustic analysis. The target words of good, hot, fun, best, and six were extracted from each sample using PRAAT software. The vowels of /ʊ/, /ɑ/, /ʌ/, /ε/, and /ɪ/ in each target word were then measured for duration (in ms), mean F0 (in Hz), and mean intensity (in dB) using waveform and spectrogram.

RESULTS

Comparison Between Japanese ELLs’ and NESs’ General Readings

Table 2

Descriptive Statistics of the Three Variables in R1 for the Japanese ELLs and NESs for Each Target Word

Note. J = Japanese, A = American, n = 30 possible in each group, R1 = general reading.

A two-way MANOVA was conducted to evaluate the differences between the Japanese ELLs and the NESs on the three dependent variables of vowel duration, F0, and intensity of the target words in the general reading. The two groups did not differ significantly in vowel duration, F0, or intensity in comparison of the target words only. Table 2 summarizes the descriptive statistics of each word for the two groups.

Comparison of the Japanese ELLs’ General Readings and Reading With Feelings

A two-way repeated-measures ANOVA was conducted to evaluate the effects of the two readings and the three variables (vowel duration, F0, and intensity) on each target word within subjects and between genders. There was a significant increase in the three variables on each target word from the general reading to the reading with emphasis for both genders. Overall results are as follows: F (1, 24) = 16.14, p < .05 for good; F (1, 17) = 3.98, p < .05 for hot; F (1, 13) = 36.32, p < .05 for fun; F (1, 16) = 16.91, p < .05 for best; F (1, 24) = 16.14, p < .05 for six. Figures 1, 2, and 3 illustrate the overall changes in each acoustic variable for the target words.

Figure 1. Duration differences from R1 to R2 (the second reading)
in the target words for the Japanese ELLs.

Figure 2. F0 differences from R1 to R2 in the target words for
the Japanese ELLs.

Figure 3. Intensity differences from R1 to R2 in the target words
for the Japanese ELLs.

DISCUSSION

Comparison Between Japanese ELLs’ and NESs’ General Readings

No significant difference was found in vowel duration, F0, and intensity of the target words between the Japanese ELLs and the NESs. The results contradicted Wennerstrom (1994), who found that English productions by Japanese ELLs demonstrated little prosodic variations, most specifically regarding the pitch variation. It could be inferred from the current findings that the Japanese ELLs in this study did apply emphasis to create a rhythm without instruction to do so, which was similar to the NESs in the study.

The difference in the current results may be related to a change of focus in pronunciation instruction from the segmental to the suprasegmental level to improve communicative competency. It could also be caused because of the modern era of globalization, which provides more opportunities to communicate with NESs and to hear English spoken.

Further examination of the mean differences in each acoustic variable indicates that the vowel duration of the Japanese ELLs, as a whole, was slightly longer than the NESs, although the F0 was lower and the intensity was smaller. These findings are also in contrast to the findings by Wennerstrom (2001) that emphasis is most commonly addressed by pitch to add information to an utterance.

Comparison of the Japanese ELLs’ General Readings and Reading With Feelings

The statistically significant changes from the first reading to the second reading demonstrate that the Japanese ELLs were capable of adjusting duration, F0, and intensity of the vowels by applying emphasis. The results confirm that they made overall changes most in duration, then in F0, and in intensity, which is consistent with the outcomes of the first question that the Japanese ELLs’ overall vowel duration was longer than the NESs even though nonsignificant.

The variable they used most to apply varied emphasis was not F0, contrary to Wennerstrom’s (2001) study. Even though the Japanese ELLs could make changes in the three variables, the finding that duration was used more for emphasis could be attributed to the mora system in the Japanese language, which might remain as an overriding influence.

Implications and Future Research

The findings provide an encouraging starting point for enhancing nonnative English speakers’ communicative competency by showing the potential for change. The advancement of technology has increased opportunities for global communication. Young adults in this study appear to have taken advantage of technology and current instructional approaches, which focus more on suprasegmental and conversational aspects.

In the current study, the application of emphasis was thought to be an important contributor for communicability. However, the question remains: Does applying emphasis itself help enhance communicability, or is communicability enhanced only when emphasis is appropriately used?

Also, investigation of the relationship between the ability or willingness to apply emphasis and other factors, such as age and length of residence in the United States, may reveal additional noticeable outcomes. Moreover, the use of human ears as measurement tools to investigate the relationships between comprehensibility and application of emphasis should be explored.

REFERENCES

Anderson-Hsieh, J. (1996). Teaching suprasegmentals to Japanese learners of English through electronic visual feedback. JALT journal, 18, 315–325.

Edwards, H. T., & Strattman, K. H. (1996). Accent modification manual: Materials and activities. San Diego, CA: Singular.

Major, R. C. (2001). Foreign accent: The ontogeny and phylogeny of second language phonology. Mahwah, NJ: Lawrence Erlbaum.

Wennerstrom, A. (1994). Intonational meaning in English discourse: A study of non-native speakers. Applied Linguistics, 15, 399–420.

Wennerstrom, A. (2001). Intonation and evaluation in oral narratives. Journal of Pragmatics, 33, 1183–1206.

Dr. Masako Maeda is a lecturer in the Department of Curriculum and Instruction at Wichita State University, where she teaches linguistics. Her areas of interest include ESL pronunciation, foreign accent modification, phonetics, and applied linguistics.

Dr. Kathy Strattman is an associate professor in the Department of Communication Sciences and Disorders at Wichita State University. As a certified speech-language pathologist, she has taught university courses in accent modification for ELLs and is the coauthor of the Accent Modification Manual.