Date of Award
10-2003
Culminating Project Type
Thesis
Degree Name
English: Teaching English as a Second Language: M.A.
Department
English
College
College of Liberal Arts
First Advisor
Ettien Koffi
Second Advisor
Suellen Rundquist
Third Advisor
Gerald Lavoi
Keywords and Subject Headings
English running speech, Japanese listeners, connected speech modification, L2 word recognition
Abstract
In spoken North American English (NAE), sound adjustments called connected speech modification (or post lexical processes), such as linking, palatalization, flapping and /h/ deletion, commonly occur in the stream of speech across word boundaries. This thesis claims that the connected speech modification is one of the causes of non-native English listeners' difficulties in comprehending spoken English. A study was conducted to test the following hypotheses:
Hypothesis 1: Connected speech modification seriously hinders Japanese listeners from recognizing English words.
Hypothesis 2: When more than one connected speech modification is involved in one utterance, Japanese listeners have more difficulty in recognizing English words than when only one connected speech modification is involved.
In this study, a set of cloze dictation experiments was administered to 50 native speakers of Japanese to examine their recognition of individual words and the same words with or without connected speech modification in a sentence. (Rates of correct answers were compared with each other and analyzed.)
If the subjects recognized the words pronounced in isolation but could not recognize the same words in a sentence with connected speech modification, and they did not have difficulties in recognizing words in a sentence without connected speech modification, the results would support Hypothesis 1. If their performance was better in the utterances with one connected speech modification than in the utterances with more than one connected speech modification, the results would support Hypothesis 2.
The overall data positively supported Hypothesis 1. As for Hypothesis 2, however, the results in this study were not consistent although they seemed to support the hypothesis to some extent. The data seemed to be too small to draw any definite conclusion. Further studies would be necessary to examine Hypothesis 2.
In addition, the incorrect answers made by the subjects were examined for their tendency to perceive and segment English speech sounds. Some examples of incorrect answers include "What a my'' in "What am I doing?" "First ball" and "Festival" in "First of all, ... " The data suggest that connected speech modification misleads Japanese listeners into segmenting the sounds incorrectly and activating wrong words even though they can recognize the right words when they are pronounced in isolation.
Recommended Citation
Ito, Kikuyo, "The Influence of Connected Speech Modification on Japanese Listeners’ L2 Word Recognition" (2003). Culminating Projects in TESL. 64.
https://repository.stcloudstate.edu/tesl_etds/64
