ABSTRACT
The Acoustic Phonetic Correlates of Lexical Stress in L2 English - Chapter 10 is devoted to lexical stress in L2 English. The 67 participants produced 7 disyllabic words, for a total of 1,600 measured tokens. The methodology used to identify and rank stressed and unstressed syllables is based on the JNDs of F0, intensity, and duration. A syllable is deemed stressed if its F0 is 1Hz higher, its intensity is 3 dB louder, or its duration is 10 ms longer than any other syllable in the same word. A similar methodology led Fry (1958) to conclude that GAE speakers encode, decode, and rank lexical stress hierarchically as follows: F0 > Duration > Intensity. Subsequent replication studies have uncovered different rankings. Regardless, the most important finding discussed in this chapter is that L2 speakers transfer the acoustic correlates from their L1s into English. This yields the following results:
English: Intensity (85.71%) > F0 (57.14%) > Duration (28.57%)
Arabic: Intensity (57.14%) = Duration (57.14%) > F0 (42.85%)
Japanese: F0 (87.51%) > Intensity (42.85%) = Duration (42.85%)
Korean: Duration (57.14%) > Intensity (42.85%) = F0 (42.85%)
Mandarin: F0 (85.71%) > Intensity (57.14%) = Duration (57.14%)
Slavic: Duration (85.71%) > Intensity (57.14%) > F0 (14.28%)
Somali: F0 (85.71%) > Intensity (57.14%) = Duration (57.14%)
Spanish: Intensity (71.14%) = Duration (71.14%) > F0 (14.28%
Unless substitutions of segments with high RFLs are involved, none of the strategies above interferes with intelligibility because the acoustic correlates of lexical stress operate both independently and interdependently.
