Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Japanese pronunciation by English speaking students and English-Korean speaking students: contrastive and error analysis
(USC Thesis Other)
Japanese pronunciation by English speaking students and English-Korean speaking students: contrastive and error analysis
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
JAPANESE PRONUNCIATION BY ENGLISH SPEAKING STUDENTS AND ENGLISH-KOREAN SPEAKING STUDENTS: CONTRASTIVE AND ERROR ANALYSES by Kanzo Takemori A Thesis Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF ARTS (East Asian Linguistics) August 1995 Copyright 1995 Kanzo Takemori UNIVERSITY OF SOUTHERN CALIFORNIA TH E GRA DUATE SC H O O L U N IV ERSITY PARK LOS A N G ELES. C A LIFO R N IA SOOQ7 This thesis, written by Kanzo Takemori under the direction of A.iJi Thesis Committee, and approved by alt its members, has been pre sented to and accepted by the Dean of The Graduate School, in partial fulfillment of the requirements for the degree of Dtmm THESIS C O M M I T T E E CidrsM ........ Table of Contents Introduction..............................iii Chapter 1. Japanese as the Target Language The Sound System of Japanese......... 1 2. English as the Base Language The Sound System of English......... 15 3. Korean as the Base Language The Sound System of Korean.......... 28 4. Contrastive Analyses of the Sound Systems of Japanese and English, and Japanese and Korean.................. 42 5. Error Analyses of Unacceptable Pronunciations by English and Korean Speaking Students.................... 66 6. Discussion and Conclusion...........125 Bibliography 132 Introduction The students in Japanese language classrooms in the United States used to be mostly those whose first languages were English. In 1991, National Foreign Language Center researched the distribution of first languages of students of Japanese. They studied 624 postsecondary students and established that English was by far the most dominant first language in Japanese language classrooms. The result of the research is summarized in the following chart. First language of Japanese Learning Students (1991) First: language Number Percent < %) English 541 86.7 Japanese 6 1.0 Chinese 23 3.7 Korean 6 1.0 Other Asian Language 34 5.4 French 1 0.2 German 1 0.2 Spanish 1 0.2 Other European Language 12 1.9 Non-European/non-Asian Languaqe 11 1.8 Total 624 100 In fact this conventional assumption may not hold true in a certain geographical area. For example, in 1994, in Southern California, this assumption was inaccurate. * * • in In five beginning Japanese classes at University of Southern California in fall 1994, I interviewed the instructors of the classes and asked them to count the number of the students and their first languages. Among the total of 78 students, 32 spoke English as their first language (41% of the students) and 46 spoke other languages (59%). Among the 46 speakers of other languages, there were 24 Chinese speakers (30.7%), 16 Korean speakers (23.1%), and 4 speakers of other languages (5.2%). Although English Bpeakers formed the largest group, Chinese and Korean speakers constituted a near majority. In the research conducted by the National Foreign Language Center and my observation, the languages, English, Chinese, Korean, and other languages are referred to as ‘first languages', which mean the languages that have been acquired naturally during childhood and used as foundations to learn a new language. For an English speaking student, English is the first language, however, for a bilingual student of English and other language, it is not very clear which is the first language, English, the other language, or both. In this study, a language that a student ubob as a foundation to learn a new language is referred to as a 'base language'. An English monolingual student has one base language, English, and a bilingual student of English and Korean, for example, has two base languages, English and Korean. Although there are varieties of base languages in Japanese classrooms, teaching methodologies used in Japanese language classrooms are mostly designed for speakers of English as the base language. The unique characteristics of other base languages are not included and the bilingual students are taught in the same way as English speakers are. Even with respect to the teaching of pronunciation of Japanese, not much attention are paid to the unique features of the base languages. As a result, the students are found to speak Japanese with a foreign accent which is unacceptable to the ears of native Japanese speakers. If communication can not be made because of the foreign accent, the teaching of pronunciation has to be improved. This study will focus on pronunciations of Japanese by the speakers of English and Korean as the base languages. I selected Korean besides Japanese and English because I thought that Korean was becoming a major base language in Japanese classrooms and soon may form the one of the largest groups, also not many researches studied this topic. The reason I decided to study pronunciations is that I strongly v believe that language is spoken, and in order to speak a language, the importance of pronunciation should not be overlooked. Language instructors' responsibilities should include teaching proper pronunciation, so that the students will be able to orally communicate well in the target language. Traditionally linguists applied a contrastive analysis of two languages to predict possible difficulties in foreign-language learning. Later, an error analysis was thought to be more reliable measurement for identifying the unacceptable forms produced by a student studying a foreign-language. During the study of this paper, I hope to find answers for the following questions. X) Is the contrastive analysis necessary? Is the error analysis sufficient without contrastive analysis? In this study, both analyses, first the contrastive analysis, then the error analysis, will be used and checked how they supplement each other. 2) Does the types and number of errors differ among the students? The students who have the same base language may have the same types and number of errors. On the vi other hand, the students whose base languages are different may have quite different types and numbers of errors. 3) Do English-speaking and English-Korean speaking students make same types of errors? The differences in the sound systems in the base languages may result in the differences of types of errors. 4) What are the frequent errors of English speaking and English-Korean speaking students? There may be some types of errors that occur frequently. 5) Do the students' errors occur randomly? When the students make errors, they may pronounce any sounds instead of the correct sounds, or pronounce certain sounds which have similar qualities to the target sound. 6) What is the factor that causes errors? Is there any relationship between the target language and the base languages? In Chapter 1, I will examine the sound system of Japanese, which is the target language under consideration. In Chapter 2, the sound system of English, and in Chapter 3, the sound system of Korean will be presented. These three chapters will be the foundation for the analyses that follow. In Chapter 4, contrastive analyses of sound systems vii of Japanese and English, and Japanese and Korean will be studied. Based on these analyses, pronunciation difficulties will be predicted by means of a hierarchy of difficulty. In the prediction, examples of predicted difficulties by English speaking and English-Korean speaking students will be presented. In Chapter 5, error analyses of Japanese pronunciation by English speaking and English-Korean speaking students will be presented based upon an experiment which was conducted to collect data on the students' actual errors with the help of ten native Japanese speakers. The data was analyzed with a view toward finding types of errors that distinguish English speaking students from English-Korean speaking students. Also, the number and the percentage of error occurrences are calculated to show which type of errors is common. Finally, predicted difficulties and actual errors by the students are compared to check which prediction is correct. The following techniques are employed in the transcription of the examples in this paper: International Phonetic Alphabets are used in brackets to indicate a phonetic transcription, as in [fuitsuiui]. Two slant lines are used to represent the phonemic interpretation, as in viii /hutuu/. Quotation marks are used to indicate the Japanese expressions and words, as in "futsuu" written in Hepburn Romaji. Parentheses are used when the English translation of Japanese is given, as in (usually). ix Chapter 1 The sound system of Japanese: Japanese aB the Target Language As a first step in contrastive and error analysis » it is important to establish the nature of the sound system of Japanese, the target language; therefore, I present the basic Japanese sound system in this chapter. The language described here is common Japanese, which is spoken by educated speakers who are native to Yamanote ('uptown') areas in Tokyo. It is used in public communication such as T.v. and radio announcements. This is the speech model presented in Japanese languge classes. The sounds of Japanese have been studied by many: Bloch, B. (1950) in Studies in Colloquial Japanese TV: Phon&nics; Han, M. S. (1962) in Japanese Phonology: An analysis Based Upon Sound Spectagrams; and Vance, T. J. (1987) in An Introduction to Japanese Phonology among others. No major disagreements as to segmental phonemes have been found among various scholars though the theoretical views present slightly different results. Suprasegmental features, such as segment duration, tone or pitch accent, and intonation present diverse views, but such features are not within the scope of this study. The speech 1 sounds that occur in Japanese may be classified into vowels and consonants. A vowel may occur independently, forming a meaningful utterance, whereas a consonant may not. A vowel forms the peak of a phonetic syllable, whereas a consonant functions as the margin of a syllable. The quality of each vowel and consonant is actually a bundle of two or more distinct components, such as point of articulation, manner of articulation, position of the glottis, and position of the velum. In the first part of this chapter Japanese vowel phonemes are described according to their phonetic characteristics. The discussion of the duration of vowels and the devoicing of vowels follows. The second part of this chapter presents a description of Japanese consonants. The consonant phonemes are introduced in a chart, and allophones and their distribution are explained. The second chart shows phonetic consonants. Then minimal pairs are given to confirm the contrasts among the consonants. There are five vowels in Japanese. Each vowel is described below. Vowels Japanese has a relatively simple vowel system with only 2 five vowel phonemes, /i/, /e/, /a/, /o/ and /ui/. £ i J is a high, front, unrounded vowel which occurs in the initial, medial and final positions of an utterance as , v in the following examples: [iki] (chic), [nirui] (to boil), and [kij (tree). [e] is a mid, front, unrounded vowel which occurs in the initial, medial and final positions of an utterance as in the following examples: [eki] (fortune telling), [ne^vu] (to sleep), and (kej (hair). [a] is a low, central, unrounded vowel which occurs in the initial, medial and final positions of an utterance as in [aki] (vacancy), [na&u] (to sound), and [ka] (mosquito). (o] is a mid, back, rounded vowel which occurs in the initial, medial and final positions of an utterance as in [oki] (off-shore), [noruij (to ride), and [ko] (child). [uij is a high, back, rounded or unrounded vowel which occurs in the initial, medial and final positions of an utterance as in [mki] (float), [nui&n] (to paint), and [km ] (suffering). Japanese speakers tend to pronounce the high back vowel with lip rounding in a careful speech. This rounding becomes very weak, and is sometimes absent in connected speech at a normal conversational tempo. 3 The following chart summarizes the phonetic characteristics of the five vowels, [i], [e], [a], [o] and [ui]. Front Central Back High i ui Mid Low Duration of vowels In Japanese, all five vowels can be either short or long. A difference in vowel length corresponds to a difference in meaning. The following examples involve minimal pairs in which a short vowel contrasts with a long vowel of the same quality. [i_] (stomach) contrasts with [iij (good) [e] (picture) contrasts with [ee] (yes) [od^isaip](uncle) contrasts with [od^iisafr] (grandfather) [suiteki](splendid) contrasts with [siyteeki] (beefsteak) [ba] (place) contrasts with [baa] (bar) [botai] (mother's body) contrasts with [bootai] (bow tie) 4 [ijjo] (together) contrasts with [i$$oo] (lifetime) • * • If the actual measurement of the duration of a short vowel is taken as 1, the measurement of the duration of a long vowel pronounced in the same environment ranges between 2 and 3 (Han, 1962). According to Han, when a short vowel is not preceded by a consonant as in the case of [i] and [iij, the ratio of 1 to 2 is obtained. When a short vowel and long vowel respectively is preceded by a voiced consonant as in [bo] and [boo], the ratio is approximately 1 to 2.5. When the preceding consonant is voiceless, the ratio is 1 to 3. The specific length of a long vowel varies somewhat according to its environment. Devoicing of Vowels There are voiceless vowels in Japanese but they do not contrast with their voiced counterparts. Specifically, the vowels [i] and [ui] become devoiced [£) and [ly] between voiceless consonants in unaccented syllables (Han, 1962). For example, the word "kushi" (comb) is pronounced [kiu£i], not *[kuj£i] and "kishi" (shore) is pronounced [kiji] and not *[kiji]. 5 Consonants The following chart gives an overview of the twenty consonant phonemes in Japanese including the two syllabic consonants which are discussed separately later. bilabial alveolar palatal velar glottal stop voiceless P t k voiced b d 9 affricate voiceless * voiced if fricative voiceless s h voiced z nasal voiced m n 9 flap voiced r semi-vowel voiced w y syllabic consionants N, 0 The consonant phonemes, their allophones and their distribution are discussed below. Stops /p/ is realized as the voiceless bilabial stop [p] and occurs before [ i e a o ua ] and [yj. [p] is weakly aspirated or unaspirated in the word initial position. /b/ is realized as the voiced bilabial stop [b] and occurs before [i e a o ui ] and [ y ]. 6 /t/ is realized as the voiceless dental stop [t] before [e a o], and as the voiceless alveolar affricate [ts] before [vu]. [t] is weakly aspirated or unaspirated in the word initial position. /d/ is realized as the voiced dental stop [d], and occurs before [e a o]. /k/ is realized as the voiceless velar stop [k] and occurs before [ i e a o vu ] and [y]. [k] is weakly aspirated or unaspirated in the word initial position. /g/ is realized as the voiced velar stop [g] and occurs before (i e a o ui ] and [y]. Affricates is realized as the voiceless pre-palatal affricate [t£] and occurs before [i e a o ui]. /]f/ is realized as the voiced pre-palatal affricate [d^] in the initial position. In the medial positions before [i a o vu], [d^] is in free variation with the fricative This is discussed in some detail in the section that follows. Fricatives /s/ is realized as the voiceless alveolar fricative [s] and occurs before [e a o vu]. 7 /z/ is realized as the voiced alveolar affricate (dz] in the initial position before [e a o vu ] , and as the voiced alveolar fricative [z] in the medial position before [e a o vu ] . /J/ is realized as the voiceless prepalatal fricative [^] before [i e a o in]. /h/ is realized as the voiceless bilabial fricative [$] before [vu], as the voiceless pre-velar fricative [x] before [i] and [y], and as the voiceless glottal fricative [h] before [e a o]. Nasals /m/ is realized as the bilabial nasal [m], and occurs before [i e a o in]and [y]* /n/ is realized as the dental nasal [n] and occurs before [i e a o ui ] and f y 1 * /rj/ is realized as the velar nasal (tj J and occurs before [ i e a o in ] and [y] in the medial position, [rj] does not occur in the initial position. F l a P v /r/ is realized as the alveolar flap [r] and occurs before [i e a o ui] and [y]. 8 Semi-vowels /w/ is realized as the bilabial semi-vowel [w] and occurs before [a]. /y/ is realized as the pre-palatal semi-vowel [y] and occurs before [a o ui ]. Syllabic Consonants /N/ is actualized as I*]. I?], [fo] or [rj] depending on its environment, and occurs in the medial and final positions. For example, /hoN/ (book) is pronounced as [hold], i- /hoN mo/ (book too) as [horn mo], /hoN da/ (it is a book) as [hon da], and /hoN ka/ (is it a book?) as [hoi£ ka]. /Q/, when followed by an appropriate voiceless stop, affricate or fricative, makes geminate consonants. Phonetically, the geminate consonants are transcribed as the double symbols [^p], [tt], [kk], [ss] or [ ]. Many examples of single and geminate consonants contrasting with each other can be found in daily speech, as the following minimal pairs illustrate. [si£ipai] (spy) [svugpai] (sour) [ita] (existed) [ i*ta ] (went) [it^i] (one) [itt^i] (agreement) [iki] (breath) [ikki] (one aircraft) (isoo] (to convey) [iggoo] (still more) [i^ui] (different kind) (i^j* ua ) (a kind) On the average the duration of single and geminate consonants is found to be 1.0 to 2.8 (Han, 1992). The following chart includes all consonantal allophones and describes the phonetic consonants which occur in Japanese. bilabial dantal pre- palatal pre- velar velar Lnocaplata nltt glottal ■top ▼oiaelaee p t k volcad b d g affricate voiotlMi ts •J voiosd dz d? fricative voloeleea ♦ B X h voiced Z 3 nasal voiced m n 3 flap voiced t ■emi-vowel voiced w y Among the consonants, [ts][t], [dz][z], and [dj][j] need additional clarification. [ts] and [t] are in complementary distribution, fts] occurs only before [ui] as in [tsvuma] (wife) or [ tsuiyui J (rainy season). [ts] is an allophone of /t/: it does not contrast with [t], the other allophone of /t/r [t] not occuring before [in]. For most speakers, [dz] and [z] are in complementary distribution and, therefore, allophoes of the same phoneme /z/. [dz] occurs initially as [dzuiibuifo ] (considerably) and [dzoo] (elephant), while (z] occurs in the medial position as in [suirua^ii] (cool) and [^izuika] (quiet). For some speakers, however, [dz] and [z] are allophones in free variation. ] do not contrast in most environments. [j ] is rare in Japanese. In some words, particularly when the preceding syllable has an [s] or [|], [j] may occur medially as in [ 9Ui^i ] (vein) or [ in those positions does not contrast with [dj]; furthermore, the same speaker may use [dj] instead without noticing the change. They are allophones of one phoneme (Han, 1962). [g] and |rj] contrast in a few words. For example, the word for "*fter death'* is [^ijo] and not *[^igo]. [£igo] is for the word "four by five". Examples of consonants that contrast with each other in the initial and medial positions are listed below. Examples are listed in the order of manners of articulation, from stops to semi-vowels, as shown in the above chart. For each 11 5 T imi] (a kind of shellfish) manner of articulation, each example ia ordered according to the place of articulation, from bilabial to glottal. Minimal pairs which show initial contrasts: [gai] (pie) [bai] (double) [tai] (sea bream) [dai] (big) [kai] (shell) [£ai] (harm) [dzai] (fortune) [tja(iro) ] (brown) [djai] (wicked mind) [sai] (difference) [Jail (gratitude) [hai] (yes) [mai ] (counter for sheets) [nai] (inside) [rai(ui) J (thunderstorm) [wai(ro)] (bribery) (^ai(ba)1 (blade) 12 Minimal pairs which show medial contrasts: [ si£i£ari ] (clear cut)1 [yokuiba£i] (greedy)* [iba(ifu2) ] (boast) [itai) (hurt) (idai) (medical school) [ika] (cuttle-fish) [icta] (burr) [ it|J a( tsijikiii) ] (flirt) [ idjakua ] or [i^akui] (weak-stoma [isa(ribi)] (fishing-fire) [iza^i] (cripple) [ija] (doctor) [ihai] (family tablet) (ima] (living room) [inai] (is not) [d^uiuirjo] or '3 uiuirjo ] (behind 1, 2 There seems to be no examples which show the consonants [pj in the medial position following the initial [ij. Therefore, the example [sui£ari] is given to show the medial [p]. [yokuibari] is given in order to show an example of [b], the voiced counterpart of [pj. 3 This example is given because there seems to be no (ft. 3 to be continued on the next page) (ifaij (rely on) [iwai] (celebration) [i^ai(ya)] (unwillingly) In this chapter, I have established a basic overview of the sound system of Japanese which is the target language system in the contrastive and error analyses that follow. examples which show [rj] following the initial [i] and forming an accented syllable with [a]. 14 Chapter 2 The Sound System of English: English as the Base Language This chapter describes the sound system of American English. American English is described here since almost all students who speak English as a base language speak American English rather than British English. There are various ways in which American English sounds are described by different scholars, although their basic concept is quite similar. Some researchers who have studied this issue are Trager, G. and Smith, H. L. (1951) in An Outline of English Structure, Prator, C. and Robinette, B. (1973) in A Manual of American English Pronunciation, 3rd ed, and Ladefoged, P. (1975) in A course in Phonetics. In the first part of this chapter, each American English vowel is described according to its phonetic characteristics, and any unique characteristics of English vowels are discussed. A chart is included which summarizes the vowel system of the language. The inventory of the consonants of the language is described in the second part of this chapter. First, a chart introduces the consonant phonemes. Allophones and their distributions are discussed, then more complex 15 consonants are summarized in a chart. Finally a list of minimal pairs is provided to illustrate the possibility for each consonant occurring in the initial, medial and final positions. Symbolization of vowels The vowel symbolization is adopted from Prator and Robinette’s An Manual of American English Pronunciation (1973) and the consonant symbolization is from the book, Linguistics: An Introduction to Language and Communication, by Akmajian, A. et al (1985). 16 Vowels The English vowel system, consisting of fourteen vowels, is complicated. ti(y)1s a tense, high, front, unrounded vowel (with an accompanying [y] glide), as in [biyt] (beat). (I]: a lax, high, front, unrounded vowel, as in [bit] (bit). l®(y)]s a tense, mid, front, unrounded vowel (with an accompanying [y] glide), as in [beyt] (bait). [£]; a lax, mid, front, unrounded vowel, as in [b£tj (bet). [»]: a lax, low, front, unrounded vowel, as in [bet] (bat). [a]: a lax, low, back, unrounded vowel, as in [pat] < pot). [a(w)]: a tenBe, low, back, unrounded vowel (with an accompanying [w] glide), as in [bawt] (bout). [*(y)]: a tense, low back, unrounded vowel (with an accompanying [y] glide), as in [bayt] (bite). [^]: a lax, mid, central, unrounded vowel, as in [b^t] (but). [;>]: a lax, mid, back, rounded vowel, as in [b2t] (bought). 17 [o(w)Jj a tense, mid, back, rounded vowel (with an accompanying [w] glide), as in [bowt] (boat). [o(y)]i a tense, mid, back, rounded vowel (with an accompanying [y] glide), as in [boy] (boy). [U]i a lax, high, back, rounded vowel, as in [pUt] (put) . [u(w)]t a tense, high, back, rounded vowel (with an accompanying [w] glide), as in [buwt] (boot). Tense and lax vowels The vowels of English can be divided into tense and lax vowels. In order to characterize the differences between them we can consider some of them in pairs, each pair consisting of a tense vowel and the lax vowel which is nearest to it in quality. Three pairs of this kind are: [iy, I] as in "beat, bit"; [ey,£] as in "bait, bet"; and [uw, U] as in "boot, foot.” In each of these pairs the lax vowel is shorter. However, not all vowels are paired with tense vowels. There are no vowels that are very similar in quality to the remaining two lax vowels, [*] as in "bat" and [9] as in "but” (Ladefoged, 1975). 18 Diphthongs English has a number of diphthongs. Diphthongs are vowels that show a change in quality within a syllable. They are due to tongue movement away from the initial vowel articulation toward a glide position. The vowels as in beat, bait, bite, bout, boat, boy and boot are all diphthongs. Diphthongs are transcribed as a vowel-glide sequence (O'Grady, et al.1993). The following chart summarizes the American English vowel system (diphthongs are not included). Front Central Back High Mid l_ 2 - P 4 LOW a t*na« (Language Files. 5th.ed .1991) Consonants The following chart shows the phonemic consonants in English. 19 bilabial labio dental inter- dan tal alveola r palatal velar glottal stop toI i el— P t k voload b d 9 aff riots TO 10*1 Ml •5 mind d3 fricative woiaaleaa I e s J h Toioed V * z 3 naaal m n 3 lateral- liquid 1 retroflex- liquid r semi vowel mioalMB w • voiced w y The following discussion summarizes the consonant phonemes, their allophones and their distribution. /p/ is realized as the aspirated voiceless bilabial stop [ph] in the syllable initial position and as the unaspirated voiceless bilabial stop [p] elsewhere. These two consonants are allophones of /p/, in complementary distribution. /b/ is realized as the voiced bilabial stop [b] and occurs in the syllable initial, medial and final positions. 20 /t/ is realized as the aspirated voiceless alveolar stop {th] in the syllable initial position, as the unreleased voiceless alveolar stop [?t] in the syllable final position, as the voiceless glottal stop [?] before a syllabic [n], as the flap [D] between vowels when the first vowel is stressed, and as the unaspirated voiceless alveolar stop [t] elsewhere. [th], [?t], [?] and [D] are allophones of /t/, in complementary distribution. /d/ is realized as the voiced alveolar stop [d] and occurs in the initial, medial and final positions. /k/ is realized as the aspirated voiceless velar stop (kh] in the syllable initial position and as the unaspirated voiceless velar stop [k] elsewhere. These sounds are allophones of /k/, in complementary distribution. /g/ is realized as the voiced velar stop [g] and occurb in the initial, medial and final positions. Affricates /*$/ is realized as the voiceless alveo-palatal affricate [t£] and occurs in the initial, medial and final positions. /dj/ is realized as the voiced alveo-palatal affricate [dj] and occurs in the initial, medial and final positions. 21 Frisfltiygg ft/ is realized as the voiceless labio-dental fricative [f] and occurs in the initial, medial and final positions. /v/ is realized as the voiced labio-dental fricative [v] and occurs in the initial, medial and final positions. /©/ is realized as the voiceless inter-dental fricative [0] and occurs in the initial, medial and final positions. 'S' is realized as the voiced inter-dental fricative [JJ-] and occurs in the initial, medial and final positions. /s/ is realized as the voiceless alveolar fricative [s] and occurs in the initial, medial and final positions. /z/ is realized as the voiced alveolar fricative [z] and occurs in the initial, medial and final positions. f\l is realized as the voiceless palatal fricative [J] and occurs in the medial and final positions. occurs in the initial, medial and final positions. /h/ is realized as the voiceless glottal fricative [h] and occurs in the initial, medial and final positions. /m/ is realized as the bilabial nasal [m] and occurs in the initial, medial and final positions. / is realized as the voiced palatal fricative and Nasals 22 /n/ is realized as the alveolar nasal [n] and occurs in the initial, medial and final positions. /tj/ is realized aB the velar nasal the medial and final positions. 'n' before 'g' is pronounced [ ij ]. Liquids /l/ is realized as the alveolar lateral [1] and occurs in the initial, medial and final positions. /r/ is realized as the alveolar retroflex [r] and occurs in the initial, medial and final positions. Semi-vowels /w/ is realized as the voiceless bilabial glide [w] and occurs before the letter 'h'. /w/ is realized as the voiced bilabial glide [w] occurs after [a], [o] or [u] to form diphthongs. The consonants which are found in American English are summarized in the following table. and occurs in 23 bilabial labio dental inter dental alveolar palatal velar glottal stop voiceless PiP*1) tn] (front, before) [SSal] (rice) [ san ] (mountain) (Sin] (shoes) [him] (force) [muge] (weight) [nam] (another person) The following list provides examples of consonants in the medial position. 38 [ag^e) (in front of) [opPPa] (older brother) [3£TTa] (do not have) (pubu] (married couple) [ sathantj] (candy) [sikTTarj] (restaurant) [kat&hi) (together) (alda) (know) [3tTTakkhe] (how) [ipKKu] (entrance) [kukKKi] (national flag) [segeJ (world) [katchi] (together) [nat£cam] (nap) [abDji] (father) [patSSak] (closer) [kam5aj (potato) [kanda] (go) [ilbon] (Japan) [sa£am] (people) The following list demonstrates the consonants in the final position. [i£] (mouth) 39 [ot] (clothes) {ak] (evil) Inom ] (fellow) [suri) (pure) [karp (river) [ornulj (today) Among the consonants described above,[p], [t], [k], [<£], [h) and [1] need additional discussion. In Korean, the lax voiceless stops and affricate such as tp], [t], [k] and [J5] become voiced intervocalically — [b], [d], [g] and [3f] respectively. For example, the word /aka/ (baby) is pronounced [aga], because [k] between voiced sounds becomes [g]. [p] and [b] are allophones of /p/ in complementary distribution. Similarly, {tJ and [d] are allophones of /t/, in complementary distribution, [k] and (g] are allophones of /k/, in complementary distribution, and [) and [3f] are allophones of /c/, in complementary distribution. The stops [p], [t] and [k] become [m], [n] and [g] respectively before nasals. For example, [£agn\inta] (catch) becomes [ Samnlinda ], [ patniinta] (receive) becomes [pannlinda ] and [kukmul] becomes [kutpul]. 40 The lax glottal fricative [h] is deleted. For example, [ cohvin ] (good) becomes [ covin ] . The lateral [1] becomes [n] after the nasals [m], [tj] and the stops [p], [t] and [kj (Nam-Kil Kim, 1990). For example, [kamlo] (sweet) is pronounced [kanrno], [aplygk] (press) is pronounced [apnygk], [matlyagpan] (first son) is pronounced [matnyajjban] and (p£Bkli] (one hundred li) is pronounced [p®kni]. In addition, the last three examples undergo the above mentioned process of nasalization in which the stops [p], [t] and [k] become [m] , [n) and [rj]. Therefore, [agnydk] becomes [amnySk], (matnyarjban] becomes [mannyarjban ] and (paBkni) becomes [paBijni]. 41 Chapter 4 Contrastive Analyses of the Sound Systems of Japanese and English, and Japanese and Korean The Contrastive Analysis Hypothesis claims that the principal barrier to second language acquisition is the interference of the first language system with the second language system, and that a scientific, structural analysis of the two languages in question will yield a taxonomy of linguistic contrasts between the two languages. This in turn will enable the linguist to predict the difficulties a learner will encounter. (Brown,1987) Contrastive analysis involves four different procedures. The first of these is descriptions the linguist or language teacher, using the tools of formal grammar, explicitly describes the two languages in question. Second, a selection is made of certain forms for contrast, since it is virtually impossible to contrast every possible facet of two languages. Forms such as linguistic items, rules and structures are contrasted. Whitman (1970) admits that the selection process "reflects the conscious and unconscious assumptions of the investigator," which in turn affect exactly which forms are selected. The third procedure is the contrast itself, the mapping of one linguistic system 42 onto the other, and a specification of the relationship of one system to the other which, like selection, "rests on the validity of one's reference points." Finally, one formulates a prediction of error or of difficulty on the basis of the first three procedures. That prediction can be arrived at through the application of a hierarchy of difficulty. (Whitman, 1970) Such a hierarchy of difficulty has been developed by Prator (1967), who introduced a hierarchy of six categories of difficulty. The six categories are listed below from level 0 to level 5 in ascending order of difficulty. Level 0 — Transfer No difference or contrast is present between the base language and the target language with respect to a particular item or structure. Students can Bimply transfer the item or structure from the base language to the target language. Level 1 — Coalescence Two items in the base language become coalesced into essentially one item in the target language. This requires that students overlook a distinction they are accustomed to in the base language. 43 Level 2 — Underdifferentiation An item in the base language is absent in the target language. Students must avoid that item. Level 3 — Reinterpretation An item that exists in the base language is given a new distribution or modified in some other way. Level 4 — Overdifferentiation A new item, bearing little if any similarity to the base language item, must be learned. Level 5 — Split One item in the native language becomes two or more in the target language, requiring students to make a distinction that does not exist in the base language. Using the hierarchy of difficulty presented above, difficulties that students will encounter can be predicted in so far as contrastive analysis is valid. The first two stages of contrastive analysis, description and selection, have already been completed in the previous chapters which described the Japanese, English and Korean sound systems. In the third stage of the 44 analysis, which is the contrast itself, one language is mapped onto the other. Japanese-Engliah Contrastive Analysis The following chart indicates the sound contrasts between Japanese and English vowels. Placed in parentheses are English vowels and sounds without parentheses are Japanese. Japanese and English vowels Front Central Back High i (i,I) ui (U,u) Mid e (e,t) (9) o (3,o) Low (a) a (a) Prediction Given the contrasts and similarities in the Japanese and English vowel systems, we can predict the following types of difficulty under the Contrastive Analysis Hypothesis. Level 0 — [i], [e], [a], [o] and (ui], although they are not exactly equal to the English counterparts, [i], [e], 45 1*1» l°] and [u] respectively due to slight differences in tongue position and lip rounding, they are regarded here as essentially the same since these slight differences are not noticeable to the average native speakers of Japanese. Level 1 — No instance of coalescence is found. Level 2 — [I], [£,], [*], (9], [;>] and (U] are absent in Japanese. They must be avoided. Level 3 — No instance of reinterpretation is found. Level 4 — No instance of overdifferentiation is found. Level 5 — Japanese has a clear distinction between short and long vowels, which English does not. Students must learn it. He see that Japanese vowel quality should be rather easy for English speaking students as long as they can avoid the vowels mentioned in Level 2. What might however cause difficulties to English speaking students is the duration of vowels in Level 5. In Japanese vowel length is distinctive 46 and vowels can be either short or long. A long vowel is approximately 2.5 times as long as a short vowel. It was pointed out that in English tense vowels are longer than lax vowels but the duration of the longer tense vowel is not as long as the corresponding Japanese long vowel. Therefore, English speaking students may have problems when pronouncing Japanese words like the following which involve long vowels. The anticipated difficulties for English speaking students are given on the right side of the arrow. A vowel followed by a plus sign indicates a vowel somewhat lengthened but not lengthened enough to sounds like a long vowel in Japanese. word (meaning) correct anticipated The following chart indicates the differences and similarities between Japanese and English consonants. Placed in parentheses are English consonants and those without parentheses are Japanese. chiisai" (small) pronunciation pronunciation — > I *£ i s a * 1 o r tt$i+sai] "suugaku" (math) (smmyakm ] — > [svurjakui 1 or [ sui+Tjakui ] "sayoonara" (goodbye) fsayoonaEa1 — > [sayona£aJ or [sayo+na£a] 47 Japanese and English consonants b ilab ia l la b io d e n ta l i n t e r d e n ia l d e n ta l a lv e o la r DM p a la ta l p a la ta l p v t- velar velar in c o m p le te velar tJolL sI slop voiceless p < p . p h ) t k ( k , k h ) (7) voiced b ( b ) d ( d ) 8(a) affricate vo ic eless 15 « <4> vow ed dz (t^ ) fricative v o ic e le st ♦ (f> (e) s ( s ) % h ( h ) v oiced (v) ( * 7 (7 ) V nasal voiced m ( m ) n (n) 3 <3 > V3 flap v oiced /(I? ) lateral- lmuid voiced ( ! ) ret n'flex- liquid voiced (r) semi vowel v o iceless <w) voiced w (w) y (y) Prediction A prediction of the learning difficulty of consonants can be accomplished by subjecting the above contrastive description to the hierarchy of difficulty. Level 0 — [p], [bj, [k], [g], [sj, [zj, [g], [m], [h] and [w] in Japanese are the same as in English. These sounds should cause no troubles for the English speaking students of Japanese since they will be directly transferred from the base language to the target language. Level 1 — English [r] and [1] are coalescenced into [*]. 48 Level 2 — The English [?], [f], [v], [0J, [J), [DJ and [y] are absent in Japanese. They must be avoided. Level 3 — The English [t], [d], [tj], [dj], [j] and [n] have to be pronounced with a different point of articulation to correspond to their Japanese counterparts. However, native Japanese speakers may not be able to distinguish them from Japanese sounds, even with the different points of articulation so these Bounds might be easier for the English speaking students of Japanese than a level 3 classification of difficulty might imply. Level 4 — [ts], [dz], [$], [x] and [fe] are entirely new to the English speaking students and thus must be learned. Level 5 — No instance of split is found. The sounds in Level 0, 2 and 3 are not problematic. The students can simply transfer their base sounds in Level 0 and 2 to Japanese, and drop the sounds in Level 3 when speaking Japanese. The sound in Level 1 does not seem to cause troubles, but there may be small troubles which are discussed later. 49 The sounds in Level 4 will probably cause trouble for the student. To see why this is so, consider the following discussion. Level 1 [£], a flap does not occur in English. Instead, there are two sounds similar to [?]: [1] and [r]. To imitate the sound [f], English speakers may pronounce [1] or [r]. The following kinds of pronunciations are anticipated. word "ryoohoo" ’ ’hiragana' "yukkuri" (meaning) correct (both) pronunc iat ion (ryoohoo] anticipated pronunciation . v — > Ilyoohoo] or [ryoohoo] (kana [hiragana] — > [hilagana] or syllabary) [hiragana] v (slowly) [yuikkmri] — > [yuiljkiuli] or [yuikkuiri] Level 4 English has no voiceless alveolar affricate [ts] in the word initial position. It has the sound only in the word final position. English speakers could deal with this problem via positive transfer of sounds that already exist in the base language. For instance, the students may substitute [t], [s] or [tj] for the sound. [t] and [s] share the same point 50 of articulation/ alveolar, with [ts], and [t^J shares the same manner of articulation, affricate, with [ts]. So, this kind of partial transfer is quite expected under the contrastive analysis hypothesis. In the case of [dz], the sounds which may be substituted for [d2] are [d], [z] and [d^], which are the voiced counterparts of [t], [s] and [t£] respectively. The following examples illustrate the expected pronunciations of these sounds with English as a base language under the Contrastive Analysis Hypothesis. word (meaning) correct _anticipated pronunciation pronunciation "tsukue" (desk) [tsuikme] — > [tiykme], [si^kuie] or ftjiykuiel "natsu" (summer) [ natBiu ] — > [ natui ], [ nasrn ] or [ natSui ] "zero" (zero) (cfee^o ] — > [deto], [zeifo] or [d^efo] "zoo" (elephant) f cfaoo 1 — > [doo], [zoo] or [djoo] [<|»], a voiceless bilabial fricative, does not occur in English. English speaking students may substitute [f], a voiceless labiodental fricative, or [h], a glottal, fricative, instead of [+]. 51 word (meaning) correct _anticipated pronunciation pronunciation [gm^ansui] — > [ fuiransui ] or [ huiJa^sm ] [ sai£igi ] — > [ sai J f ] or [ saihi^ ] [jn^tsuika] — > [fi^itstuka] or (htptsuika] furansu" (France) saifu (wallet) futsuka" (the second day) English has no [x], a voiceless prevelar fricative, [h], a voiceless glottal fricative, may be substituted for the sound instead. This is because both sounds share the same manner of articulation and a similar place of articulation. word (meaning) correct anticipated The incomplete velar nasal (syllabic nasal) [JQ] does not occur in English. [m]r [n] or [Ip may be substituted for it since these sounds at least share the same manner of articulation. hiki (counter for animals) (fire) (comparison) pronunciation pronunciation [xiki] — > [hiki] [xiki] taihi [xi] — > [hi] [taix^] — > [taih^] 52 word (meaning) correct — > anticipated pronunciation pronunciation "nihon" (Japan) [nihojO] — > [nihoj], (nihonJ or (nihoj] "nihonshoku" (Japanese [nihofojokui ] — > [nihom£oktu), food) [nihon^okui] or (nihor^Jokm ] "sensei" (teacher) [se&see] — > [semsee], [sensee] or [semsee] In addition to the problems predicted above, Japanese syllabic consonants, phonemically transcribed /Q/ and /N/, may cause trouble for English speaking students. As it was mentioned in Chapter 1, /Q/ makes a geminate consonants when followed by an appropriate voiceless stop. According to Han (1992), native Japanese speakers distinguish between the single and geminate consonants and single consonants by controlling the stop closure duration in the ratio of 1.0 to 2.8, while English speakers miss the contrast or pronounced them with the ratio of approximately 1.0 to 2.0. Therefore, the following kinds of problems can be predicted. word (meaning) correct _anticipated pronunciation pronunciation "Buppai" (sour) [smppai] — > [suipai] "itta" (went) [itta) — > [ita] 53 "yokka" (the fourth [yokka] — > [yoka] day) ~ /Q/ is nevr to English speakers, and the controlling of stop closure durations associated with single and geminate consonants may present Level 5, split category problems. The syllabic nasal /N/ is new to English speakers. In particular, the control of duration of /N/ at the syllable final position is a new phonetic skill the English speakers must learn. The syllabic nasal /N/ presents Level 3, reinterpretation, category problem. Sato (1993) reports the ratio of longer syllable final nasals to shorter syllable final nasals is 1.0 to 1.48 in Japanese, and 1.0 to 1.27 in English. Therefore, when English speakers pronounce Japanese syllabic nasals, they may pronounce them too short. word (meaning) correct — > anticipated pronunciation pronunciation "manpai" (full) [maqipai ] — > [mapai] ”nandai" (difficult [nandai] — > [nadai] problem) "gangai" (steep cliff) [garjgai] — > [gagai) appointee) • 54 Japanese-Korean Contrastive Analysis The following chart summarizes the contrast between Japanese and Korean vowels. Placed in parentheses are Korean vowelB, and those without parentheses are Japanese. Japanese and Korean vowels Front Central Back unrounded rounded unrounded unrounded rounded High i (i) (u) ui {ui) (u) Hid high e(e) (o) (9)o(o) Mid low (t) O) Low a (a) Prediction Given the contrasts and similarities for vowels indicated above, we can expect the following kinds of difficulties. Level 0 — The vowels [i], fej, [ui] and [a] are the same in Japanese and Korean. [u] should also be included in this level also since rounded high back vowel occurs in Japanese. Therefore, these sounds can be simply transferred 55 from the base language to the target language. Level 1 — No instance of coalescence is found. Level 2 — (ii], [d], [0], [£j and [0] are absent in Japanese. They must be avoided. Level 3 — Although there is a distinction between long vowels and short vowels in Korean, it is not the same as Japanese. Students must learn it. Level 4 — No instance of overdifferentiation is found. Level 5 — No instance of split is found. He see then that all the Japanese vowels are included in the Korean vowel system. If Korean speaking students are able to avoid using the vowels categorized in Level 2, the pronunciation of Japanese vowels will bring them little difficulty. (Beyond the contrasts of the vowel systems,) Korean speaking students may however face problems when they try to pronounce Japanese long vowels as mentioned in Level 56 3, because the precisely the same distinction does not exist in Korean. As discussed in chapter 3, only extra duration /:/, not an additional vowel, is given to a long vowel in Korean to distinguish it from a short vowel. That means Korean long vowels are shorter than Japanese long vowels. Therefore, the following kinds of examples can be predicted as possible problems. A vowel followed by a plus sign indicates a vowel somewhat lengthened but not lengthened enough to sounds like a long vowel in Japanese, word (meaning) correct anticipated The contrasts between Japanese and Korean consonants are summarized in the chart below. Sounds in parentheses are Korean consonants and those without parentheses are Japanese. Moreover, parentheses are used to indicate "aspirated, "tense" and "lax" to indicate that they occur only in Korean. pronunciation pronunciation H yasashii" (easy) "sensei" (teacher) (seOsee1 doo (how) (doo] [yasaji] — > fsetose*) or (se)dse ] • — > [do*] or [do] 57 Japanese and Korean consonants bilabial dental alveolar pre- palalal palatal pie- velar vdar incomplete velar glottal stop (aspirated) (ph) (th) (kh ) (tense) (PP) (TT) (KK) (lax) voiceless P (P) t (t) k (k) (lax) voiced b (b) d <d) a (») affricate (aspirated) * h> (tense) ( W ) (lax) voiceless ts s (?) (lax) voiced dz d? (?) fricative (tense) (SS) (lax ) voiceless ♦ s (s) s (?) X h (h ) (lax) voiced z nasal m(m) n (n) 3 (3> id lateral (1) flap V V r (r) sem i-vow el w (w) (y ) y Prediction Given the contrasts and similarities between Japanese and Korean consonants indicated above, we can anticipate the following kinds of difficulty. Level 0 — Japanese and Korean share seme of the same V consonants: [p], [bj, [k], [g], [s], [z], [h], (m], [g]f (r] and [w]. The Korean speaking students of Japanese can simply transfer sounds from the base to the target language. 58 Level 1 — No instance of coalescence is found. Level 2 — Korean (Ph3, [PP], [th], [TT], [kh], [KKJ, [ £h ], [£(? ], 153 J and [1] have to be avoided. They do not exist in Japanese. Level 3 — Korean [t], [d], [J5], [*], {n] and [y] need to be reshaped to form their Japanese counterpart. However, because of their similarity to the corresponding Japanese sounds, it is plausible that native Japanese speakers may hear them as Japanese sounds anyway since they have very close point of articulation; Korean and [5] are palatal, and Japanese [t^] and [d^] are prepalatal. (The counterparts of Korean [cj and [J] are [t^] and [dj] respectively in Japanese.) Level 4 — Japanese [ts], [dz], [+], [z], ($], [j], [x] and [Id] are new to the Korean speaking students. Level 5 — Korean [h] may be substituted for Japanese 14*] or [x] in Japanese. 59 The sounds in Level 0, 2 and 3 should cause few problems. For the soundB in Level 0 and 3, Korean speaking students can simply transfer their base sounds to Japanese. The sounds in Level 2 need to be dismissed. Level 4 and 5 have the possibility of causing trouble for Korean speaking students. Anticipated problems are discussed below. Level 4 [ts], a voiceless pre-palatal affricate and [d2], a voiced pre-palatal affricate, do not occur in Korean. When trying to pronounce this sound, Korean speaking student may pronounce the affricate [c] or the fricative [s] instead of [ts], and [d ] or [j] instead of [dz], since they share some similar features with target sounds. Some anticipated unacceptable pronunciations are given next, word (meaning) correct — > anticipated "zeroM Tsuda" (a surname) empitsu" (pencil) futsuu" (usually) (zero) pronunciation pronunciation [empjteui ] [ emp^SLu ] [frytsmuj ] — > [4*fi&uiuj] or [+^1 BUI Ul] [ ts ui da ] — > [£uida] or [suida] [ dze^o ] 60 "zai" (fortune) [dtai] — > [djai] or (5*i] ’ ’zuibun" (considerably) (cfeuiibuil?] — > [ctyuibuiW] or (5uiibuj&] Korean has no [ + ], a voiceless bilabial fricative. Korean speaking students may produce [p], a lax voiceless bilabial stop, or [h], a lax voiceless glottal fricative instead because [$] and [p] share the same place of articulation, and [$] and [h] share the same manner of articulation. word (meaning) correct anticipated pronunciation pronunciation "furansugo" (French) [guiransuirjo] — > [pui^ansuitjo] or [hiulfansiuijo] "saifu” (wallet) [saijnu] — > [saijxu] or [ saihui 1 1 — o 1 "fujisan" (Mt. Fuji) [$uid^isa(9] — > [£uidjisa(0] or [huid^isalS] [z], a voiced alveolar fricative does not occur in Korean. Instead of [z], [^] might be employed. [ft, a voiced palatal affricate, sounds close to [z]. Both are voiced, their points of articulation are on the roof of the mouth, and their manners of articulations produce friction. word (meaning) correct anticipated pronunciation pronunciation "suzushii" (cool) rsmzm^ii 1 (fsuidztu^ii 1)— >[sui^ui^ii] 61 "shizuka" (quiet) [^izuikal (f ^idzuikal) — >[^ij[uika] "shizen” (nature) [Jizefl)] ( [J idzefg]) — >[$iie|0] [^], a voiceless pre-palatal fricative, and [^], a voiced pre-palatal fricative, do not occur in Korean. The closest sounds to them are (s), a voiceless alveolar fricative, and [^f], a lax voiced palatal affricate respectively. The following mispronunciations are expected. word (meaning) correct anticipated pronunciation pronunciation ” jisho” (dictionary) [cfaijo] — > [dsiso] "tabemashoo" (letMs eat) [tabema^oo] — > [tabemasoo] "shijimi" (a kind of (Jizimi] shell) "saji " (spoon) (sazi] — > [sazi] In Korean there is no [x], voiceless pre-velar fricative. This consonant may be replaced by [h], a (lax) voiceless fricative. The following examples illustrate the expected mispronunciations. word (meaning) correct anticipated pronunc iation pronunc iat ion "hito" (person) (xjto] — > [Hito] "nihiki" (two animals) (nixjki) — > [ nihiki ] "mahi" (paralysis) [max^] — > [mah£] 62 The incomplete velar nasal (syllabic nasal) [ft] is a new sound in Korean. Korean speaking students may apply [ml, [n] or [tj] instead since they share the same manner of articulation, word (meaning) M hon" (book) "asagohan" (breakfast) 'sensei" (teacher) correct anticipated pronunciation pronunciation [hoj£] — > [hom], [ hon ] or [hotj] [asaijohatp] — > [as’ H^johajji], [asarjohag] cr [asagohay ] [sebsee] — > (se^see]7~ [segsee] or [segsee] Level 5 In Japanese, /h/, voiceless glottal fricative, becomes [$] when followed by [u j], and [x] when followed by [i]. It may be confusing for Korean speaking students when and which sound to pronounce. They may pronounce [h] when they encounter Japanese [$] and [x). The following examples illustrate the expected mispronunciations. word (meaning) correct > anticipated pronunciation pronunciation 1 1 furui" (old) "naifu" (knife) "himitsu" (secret) [juiruii] — > [huiruii] [ nai£i£ ] — > [ naiht^i ] [ximitsi^i ] — > [ himitsi^i) 63 "koohii" (coffee) [kooxii] — > [koohii] In addition to the predicted problems in Level 4 and 5, the following problems may occur when Korean speaking students speak Japanese. The first prediction is the problems caused by the influence of Korean voicing assimilation, a rule that is absent in Japanese. The voiceless stops and affricates [p, t, k, become voiced [b, d, g, 5l respectively between voiced sounds in Korean. If these voiceless consonants occur intervocalically in Japanese, they might be pronounced as voiced by Korean speakers. The following examples illustrate. word (meaning) correct anticipated pronunciation pronunciation "ippon" (one stick) [ippo^)] — > [ibbo£> ] "watakushi" (I) [watakiji$i ] — > [wadagui^ i ] "kokuban" (blackboard) [kokuiba£)] — > [koc[Luba|c)] Moreover, based on the Korean voicing assimilation rule, it can be predicted that initial voiced sounds in Japanese may be pronounced voiceless by Korean speaking students, because voiceless stops and affricates are voiced only between voiced sounds. The following examples illustrate this possibility. 64 word "banana" "gakkoo" "jisho" (meaning) (banana) (school) (dictionary) correct _> anticipated pronunciation pronunciation [banana] [£akkoo] — > [panana] — > [ka£koo] [^ijo] The second prediction is the problems caused by Korean [h] deletion. Korean [h] is deleted when it occurs between vowels. When this deletion is applied to Japanese, the following unacceptable forms may occur: word (meaning) correct anticipated pronunc i at ion pronunciat ion "nihongo" (Japanese) [nihorjTjo] — > [ni0oijijo] "ohayoo" (good morning)[ohayoo] — > [o0ayoo] "asagohan" (breakfast) [asaijohafD] — > (asarjoOaK)] The third prediction is the problems based on the duration of long consonants. In Korean, long consonants have a much shorter duration than Japanese long consonants. Therefore, Korean speaking students may pronounce them as too short. The following kinds of problems are anticipated. word (meaning) correct anticipated pronunciation pronunciation (more) [mofrto] — > [moto] (slow) [yujfrkuiri] — > [yuikui^i] "motto" "yukkuri” "rippa" (good looking) [£ippa] — > [Sripa] 65 Chapter 5 Error Analyses of Japanese Pronunciation by English speaking and English-Korean Speaking Students In this chapter, I report an experiment I designed and carried out to show what sort of unacceptable pronounciations students actually had when speaking Japanese. The error analysis was necessary because the students seemed to make errors that the contrastive analyses did not predict. The experiment involved native Japanese speakers listening to what students said in Japanese and judged whether the students' pronunciation was acceptable or not. The procedures and the results of the experiment are described below. Experiment 1. Material Twenty seven expressions, statements or simple conversations, were used as the material. The expressions were given in English and the Japanese equivalent was elicited from the students. The expressions were taken from the materials that students had studied in the first eight weeks of a Japanese I class at USC, in Fall 1994. Every 66 line of the expressions contained at least one item which could possibly cause pronunciation difficulty as predicted by the contrastive analyses. The material was presented in the form of exercise sheets as shown in Table 1. The expressions in parentheses are the expected Japanese translations. They were not printed on the sheets. (They are provided here in Romaji for the reader's convenience.) Table 1. Exercise: Say the following expressions in Japanese. 1. What language dictionary is this? (kore wa nanigo no jisho desu ka?) 2. It's a Chinese dictionary. (sore wa chuugokugo no jisho desu.) 3. Do you understand French? (anata wa furansugo ga wakarimasu ka?) 4. No, I don’t. (iie, wakarimasen.) 5. Then, let's speak in English. (ja, eigo de hanashimashoo.) 6. Excuse me, please speak more slowly. (sumimasen, motto yukkuri itte kudasai.) 67 7. Which do you like better, sukiyaki or hamburger? (sukiyaki to hambaagaa to dochira no hoo ga suki desu ka?) 8. I like them both very much. (ryoohoo tamo daisuki desu.) 9. How * s your school? (gakkoo wa doo desu ka?) 10. It is very interesting. (totemo omoshiroi desu.) 11. What a big desk it isl (zuibun ookii tsukue desu net) 12. Is Japanese difficult? (nihongo wa muzukashii desu ka?) 13. no, it is very easy. (iie, totemo yasashii desu.) 14. 'empitsu' is usually written in hiragana. ('empitsu' wa futsuu hiragana de kakimasu.) 15. Let's read those characters together. (sono ji o issho ni yomimashoo.) 16. Good morning. It's beautiful today, isn't itl (ohayoo gozaimasu. kyoo wa ii otenki desu nel) 17. I eat Japanese food for breakfast. (watakushi wa aBagohan ni nihon-shoku o tabemasu.) 68 18. I study a little every night. (watakushi wa maiban sukoshi benkyoo shimaau.) 19. What kind of teacher is Professor Suzuki? (Suzuki-sensei wa donna sensei desu ka?) 20. Professor Suzuki is a very good teacher. (totemo ii sensei desu.) 21. What's this? (kore wa nan desu ka?) 22. It’s an eraser. (sore wa keshigomu desu.) 23. My watch is very small. (watakushi no tokei wa totemo chiisai desu.) 24. Do you know Mr. Tanaka's telephone number? (anata wa Tanaka-san no denwa-bangoo o shitte imasu ka?) 25. Just a moment. Mr. Tanaka's telephone number is 502-2307. (hai. chotto matte kudaaai. Tanaka-san no denwa- bangoo wa go-zero-ni no ni-san-zero-nana desu.) 26. Thank you very much. (doomo arigatoo gozaimaBu.) 27. Not at all. You're welcome. Good-bye. (iie. doo itashimashite. sayoonara.) 69 2. Subjects and recording procedures Ten students participated in this experiment: five monolingual English-speaking students and five bilingual English-Korean speaking students. All participants were enrolled in Japanese I class. This experiment was conducted as a review exercise. First, they were asked to silently compose the Japanese equivalent of each English sentence, but not to write anything on the sheets. This eliminated the possibility of their uBe of Romanized letters or their own native language to transcribe sounds. Second, each sentence translation was reviewed in a repetition-drill: the students were asked to speak the Japanese equivalents of the English expressions. The instructor confirmed their grammatical and lexical correctness, providing the correct expressions as needed. This drill was necessary to avoid entirely different responses from students. Then, they individually looked at the exercise sheets and recorded their Japanese expressions. No time pace was set to avoid unnecessary pressure. This experiment was conducted in the Language Lab room at the USC's Taper Hall of Humanities. The equipment used was Sony, Boothrecorder ER-840. A cassette-tape, Gemini MP60 (speed of 2 in. per second), was provided to each student by the instructor. 70 3. Listening Analysis by Judges Ten native speakers of Japanese, from now on called 'judges', listened to the recordings. The followings are the brief descriptions of the judges and their backgrounds. Judge 1 is a 33-year-old female native speaker of Japanese. She was educated through college in Japan, and has been in the United States for 5 years. Judge 2 is a 34-year-old female native speaker of Japanese. She was educated through junior college in Japan, and has been in the United States for 4 years. Judge 3 is a 48-year-old female native speaker of Japanese. She was educated through junior college in Japan, and has been in the United States for 4 years. Judge 4 is a 39-year-old female native speaker of Japanese. She was educated through high school in Japan, and has been in the United States for 4 years. Judge 5 is a 22-year-old male native speaker of Japanese. He was educated through high school in Japan, and has been in the United States for 5 months. Judge 6 is a 25-year-old female native speaker of Japanese. She was educated through junior college in Japan, and has been in the United States for 3 months. Judge 7 is a 23-year-old male native speaker of 71 Japanese. He was educated through high school in Japan, and has been in the United States for 1 year. Judge 8 is a 28-year-old female native speaker of Japanese. She was educated through college in Japan, and has been in the United States for 1 year and 6 months. Judge 9 is a 21-year-old female native speaker of Japanese. She was educated through high school in Japan, and has been in the United States for 6 months. Judge 10 is a 24-year-old male native speaker of Japanese. He was educated through high school in Japan, and has been in the United States for 1 month. These people are all monolingual Japanese speakers and represent a common young generation of Japanese. They had no linguistic training or no known hearing difficulties. They were selected because they were available and willing to act as the judges when requested. The judges listened to the tapes and examined whether each student's pronunciation was acceptable or not to their ears. They were asked not to pay attention to wrong vocabulary or grammar. They listened to the ten cassette-tapes, one by one, on a cassette player, a JVC RC-X3. Every judge had adequate opportunity to listen to and check each speaker's utterances. They were given a 72 check sheet (Table 2) in which the expected Japanese utterances were provided. They were asked to mark on the sheet when they heard unacceptable pronunciations, and to transcribe the incorrect sound they heard, either in Roman alphabet or in the Kana syllabaries. Table 2. Check Sheet 1. "kore wa nanigo no jisho desu ka?” 2. "sore wa chuugokugo no jisho desu." 3. "anata wa furansugo ga wakarimasu ka?" 4. "iie, wakarimasen." 5. "ja, eigo de hanashimashoo." 6. "sumimasen, motto yukkuri itte kudasai." 7. "sukiyaki to hambaagaa to dochira no hoo ga suki desu ka?” 8. "ryoohoo tomo daisuki desu." 9. "gakkoo wa doo desu ka?" 10. "totemo omoshiroi desu." 11. "zuibun ookii tsukue desu nel" 12. "nihongo wa muzukashii desu ka?" 13. "iie, totemo yasashii desu." 14. "empitsu wa futsuu hiragana de kakimasu." 15. "sono ji o issho ni yomimashoo." 16. "ohayoo gozaimasu. kyoo wa ii otenki desu nel" 17. "watakushi wa asagohan ni nihonshoku o tabema su." 18. "watakushi wa maiban sukoshi benkyoo shimasu." 19. "Suzuki-sensei wa donna sensei deBU ka?" 20. "totemo ii sensei desu." 21. "kore wa nan desu ka?" 22. "sore wa keshigomu desu." 23. "watakushi no tokei wa totemo chiisai desu." 24. "anata wa Tanaka-san no denwa-bangoo o shitte imasu ka?" 25. "hai. chotto matte kudasai. Tanaka-san no denwa-bangoo wa go-zero-ni no ni-Ban-zero-nana (502-2307) desu. 26. "doomo arigatoo gozaimasu." 27. "iie. doo itashimashite. sayoonara." Marking of the judges looks like the example presented below. "gakkoo wa doo desu ka?" (How's your school?) k o l > j > This example indicates that a judge noticed that a 74 speaker pronounced a voiceless [k] for the described voiced [g], the long vowel [oo] was too short and sounded like a single vowel [o], and the voiceless alveolar fricative [s] sounded more like ' . To transcribe the vowel length and the sound deletion the following markings were used. [VJ short vowel (V stands for any vowel) [W] long vowel (V+J -vowel shorter than [WJ but longer than (V] [0]-----vowel missing Each judge's markings were later translated into phonetic alphabets on the summary check sheet (Table 3). In the table below only the first sentence shows the spaces for judge's markings. In the actual summary check sheets the same format was used for all the sentences, 2 through 27. Phonetic transcriptions were provided below the expected Japanese utterances. 75 1. Table 3 Summary check sheet 'kore wa nanigo no jisho desu ka?" ko¥e wa nanirjo no dji^o desv^ ka] judge 1 2 3 4 5 6 7 8 9 1 0 2. "sore wa chuugokugo no jisho desu." [sore wa t^unugokuiijo no ^^5° desi^ ] 3. "anata wa furansugo ga [anata wa ^uiransuirjo rja wakarimasu ka?" waka^imasui ka] 4. "lie, [iie wakarimasen." wa ka£ imasetd ] 5. "ja, eigo de hanashimashoo." [d^aa eer|o de hana|iina|oo] 6. "sumimaaen, motto yukkuri itte kudasai." v. [sui mimase(0 motto yuikkuiri itte kuidasai] 7. "sukiyaki to hambaagaa to dochira no hoo ga [suikiyaki to hambaagaa to dot^ilfa no hoo rja suki desu ka?" suiki desui ka] e > o 76 8. " ryoohoo tomo daisuki desu.” [Ifyoohoo tomo daist^ki desiyj 9. "gakkoo wa doo desu ka?” [gakkoo wa doo desi^ ka] 10. "totemo omoshiroi desu.” [totemo omojiroi destu] 11. "zuibun ookii tsukue desu nel" [dzmibuitd ookii tsmkrne desui nee] • © • 12. "nihongo wa muzukashii desu ka?" [nihoyrjo wa mmzmka^ii desm ka] 13. "iie, totemo yaaashii desu." [iie totemo yasa£ii desm) 14. "empitsu wa futsuu hiragana de kakimasu.” v [empitsm wa ipmtsmui hirarjana de kakimasiu ] 15. "sono ji o issho ni yomimashoo." [sono dji o yomima^oo] 16. "ohayoo gozaimasu. kyoo wa ii otenki desu nel" [ohayoo gozaimasm kyoo wa ii otenki deem nee ] 17. "watakushi wa asagohan ni nihonshoku o [watakm£i wa asagohan ni nihoR^okm o tabemasu." tabemasm] o 77 18. 19. 20. 21. 22. 23. 24. 25. "watakushi wa maiban sukoshi benkyoo shimasu." [watakt^i£i wa maibaQ si^ko^i benkyoo £imasm] "Suzuki-sensei wa donna sensei desu ka?" [ suizmki-Befesee wa donna selQsee desm ka ] "totemo ii sensei desu." [totemo ii seOsee desm] • • "kore wa nan desu ka?" [kolf'e wa nan desm ka] • » "sore wa keshigomu desu." [solfe wa ke^igomm desm] "watakushi no tokei wa totemo chiisai desu." [wataki^i^i no tokee wa totemo t£iisai desm] "anata wa Tanaka-san no denwa-bangoo o [anata wa tanaka-san no de£wa-baygoo o shitte imasu ka?" ^itte imasm ka] "hai. chotto matte kudasai. Tanaka-san no [hai tjotto matte kmdasai tanaka-san no] denwa- bangoo wa go-zero-ni no ni-san-zero-nana detdwa - bapgoo wa go-dzeFo-ni no ni-sa|&-<feeFo-nana desu." desm ] f t 78 26. "doomo arigatoo gozaimasu." y [doomo anrjatoo gozaintasi^] 27. "iie. doo itashimashite. sayoonara." [iie doo ita^ima^ite sayoonara] The following is an example of how a summary check sheet was used by the ten judges to record the results of the perception analysis. Example:"sore wa chuugokugo no jisho desu." [so^e wa tfuiuigokiiiT|o no d|i^o desi^i ] ju d g e 1 m 2 oo 3 4 Ul 5 Ul o+ 6 0 7 8 o o 9 Ul 10 s This example shows that judge 6 did not hear [r] in [solfe]. The long vowel [uivu] in [ tjunugokuitjo] was perceived as a short [m] by judges 1, 4, 5 and 9, and the final [o] in the same word as [oo] by judges 2 and 8, and [o+] by judge 5. Also [d^] in [djijo] was heard as [t^] by judge 3, and as [£] by judge 10. 79 Every judge was capable of distinguishing acceptable pronunciations from unacceptable pronunciations. They mostly agreed as to the student's pronunciation whether it was acceptable or not. Next, the number of phonemes in each expression, the number of errors (unacceptable pronunciation) and the number of errors per expression were tabulated, and presented in Table 4. 80 Table 4 Number of errors and the percentage of errors by each student Expression The number of phonemes in the expression Total 1 numberof phonemes heard by the 10 judges [T h e n u m b er o f e rro rs m a rk ed by th e 10 ju d g e s ( # ... stu d e n t n u m b er) #1 #2 # 3 # 4 # 5 # 6 # 7 #8 # 9 #10 1 24 240 2 0 0 0 2 9 3 2 0 2 2 25 250 1 1 4 1 0 9 6 3 0 2 3 30 300 0 0 1 0 1 0 4 1 0 4 4 14 140 0 0 0 0 0 0 2 0 0 0 5 20 200 1 0 1 1 0 9 3 2 0 1 6 32 320 | 8 2 6 0 7 17 1 4 3 2 7 44 440 1 14 18 10 2 1 18 14 19 22 7 8 22 220 1 5 3 12 0 0 3 0 1 7 7 9 17 170 | 10 11 9 0 0 3 9 5 5 14 10 18 180 1 0 6 0 2 6 3 3 4 5 2 1 1 23 230 | 5 6 4 0 2 14 14 13 12 6 12 24 240 | 8 0 2 1 1 2 9 11 6 15 13 20 200 | 3 0 0 0 0 5 9 2 11 0 14 31 310 | 4 0 1 1 0 18 14 2 1 1 3 15 22 220 | 1 4 0 1 6 14 0 7 1 3 16 36 360 § 3 5 7 0 2 6 11 6 13 0 17 38 380 j 3 0 0 1 0 10 19 13 4 6 18 35 350 | 1 1 3 1 2 0 13 2 3 2 19 31 310 § 6 5 6 2 5 7 1 10 4 9 20 18 180 | 0 2 3 0 7 0 1 0 1 0 21 15 150 1 0 0 0 0 0 0 0 0 0 1 22 18 180 | 0 0 0 0 0 0 0 1 0 2 23 33 330 | 7 2 3 2 2 15 7 9 4 0 24 42 420 j 8 6 2 0 0 0 6 6 2 11 25 71 710 | 6 7 1 4 0 14 4 1 5 10 26 22 220 J 3 0 6 0 11 7 4 2 2 4 27 26 260 | 5 10 6 0 9 6 0 2 6 1 T o t a l 751 7510 j 104 89 87 19 64 189 157 128 126 114 The percentage of errors marked (%) 1.38 1.19 1.16 0.25 0.85 2.52 2.10 1.70 1.72 1.52 81 Expression 1 had 24 phonemes. Because the ten judges listened to these 24 phonemes, there were a total of 240 phonemes heard. In Expression 1, student 1 (#1) pronounced two unacceptable phonemes. Two out of 240 were heard as errors. One out of 250 in Expression 2, zero out of 300 in Expression 3, zero out of 140 in Expression 4 and five out of 260 in Expression 27 were counted as errors. Finally, all phonemes and errors were totaled and the percentage of errors was calculated. Since there were 751 phonemes in the whole expressions, the total of phonemes heard by the ten judges were 7510 (751 phonemes x 10 judges). 104 errors were noted by the judges. The ratio of errors and the sounds listened to, or the percentage unacceptability by the judges in other words, were 104/7510 = 1.38%. The same processes were employed to calculate each student's number of errors and the ratio of errors and the sounds listened to. 82 Table 4 Each student's summary check sheet Bapri 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. Student 1 ■ ■ion jl ko^awananiijonoodji^ odas^ikal ■ o£avai§m uj gokmtjopodj l^oda Big ] anatawa+uifagsanjogawakarisesijika I liawakafiaaa^ ^aaaarjodahana^taa^oo ] ■umi^asajQao ttoyuifrJcuiEii^takuida»ai ] smkiyakitohaabaagaatodot^iranohoorja ■uikidasi^ka ] ^ ^yoohootomodaistpkidesip ] g akkoowadoodasmka ] Ay 'p 8 totamooao^ i£oida»iy ] (feuiibuiya ookiilsuikuiadaauinaaj * "3" i i . nihoyrjowamm zuika^iidaaiuka] iiatotaeoyas^iidasm ] a«p jtsmva^iu-tsmuihif arjanadakakiaaaui ] ■ono^ioiy oniynai i aafed ohayoogozaiaasuikyoot^ai-ioterjkidasuinaa ] wataJuu^i waasagohanninlhol^okuiotabaaasui ] watakvji^ i warna iba^ ■ iu ko^ibayky oo Cim* * ^ ] suizuik^s^Qsaawado^nass^saadasiuka ] totaaoiiaabsaedasui 1 a • ko^avana^dasi^ka] ao^ewaka^igoaiudasty ] vatakui$inotokaawatotaaot£ii saidasig} anatawatanakasaqnodabwabaygooo^ i^taiaaaiyka ] hai t^o t^.o«atta kuida ■ aitanakaa annodaftwa baygoowagcKke^oninonisat^karonanadaBiu ] dooaoa¥igatoogocaiaaaig ] iiedooita^ iaaCita»ayoona£ a ] f ° pronounced total phonaaas notad arrors by 10 judges 24 25 30 14 20 32 44 22 17 IB 23 24 20 31 22 36 38 35 31 18 15 18 33 42 71 22 26 751 2 1 0 0 1 8 14 5 10 0 5 8 6 0 0 0 7 8 6 3 5 104 83 Studwit 2 raaaion [ b>{M *nani3oao<9iSo<lMVU| aotawtJuiuigflkunjonodjiJodaaig ] V / ^ tn ittw i^ m rtn iu w jo ijiw tlL triM a y k a } <^u»tij O(tohtn»^lat^oo j ■oaiJMit)) ■ottoyuikkmr 1 l^jilou^ntl ] auikivalcitoha»byq^ytodolj iranohoorja s<£ik±daa\iika] ^oonootgpodaiaiykidaaig ] g afrktyadoodaamka ] to^aaoomo^ iioidaat^ ] ckuiibuite ooklitavpkuiadaaumaai nihoipowaauizuika^iidaai^k^) iietotaaoyaaajjiidaaui j afp£.lauiwa4igilBuiiiihi£aTjaiiadakak±aaaip ] aonodjioi^oniyoaiaaJooJ ohavcyiozaijaauikvoowaiiota^kidaauinaa ] watakmji waaaagohapninihol^okuiotabaataaiu ] watakiy^ iwamaiba{0Bi^k^.^Lba^kyoc£i»aa»£ ] suijuik jBajOaaavadonnaeafceaadasmka ] totspoiiaaAaaadaatu ] x * o ko^ewanandaavyka] aorawaka^igoamdaai^ ] watak^inotakaavatotanot^iiaaidaaui ] anatawatanakaaanDoda^vabaijqcgo^ ittalaaat^ka hait^ottoaa^takuidaaaitanakaaannod^pwa baijq^ovaqodea^oniiionlaafoAaronanadaaiy ] dooaoa^igatoogozaiaaa^ ] iiadooi t^i*af itaaavyona^a ] pronouncad total phonaaaa notad arrora by 10 judgaa 0 10 12 Student 3 Uprtaalon 1. [ ko)f«¥>nuili|onodji^ odai^lu) 24 2. f aorewatj^ ujmaokuitjonod^l^bdeaiy l 25 4 3. [anatawa+uiEanaanjogaeakaEieaay*aj 30 1 4. [ iiewmknSieaee*] 14 0 5. [ djaaoeijodahana^jjM^oo l 20 1 6. [ tnwii ■neofe eo^oymfcfcui^lltf kuidaaal ] 32 6 7. (aipklvaldtohMbiyaa^todot^iMoliooija at£kldaai£ika] 44 8. (£yoohootoaoda±at^kideatp ] 22 9. [gafrkoowadoodeaipka ] 17 9 10. [ totenooaoJJ i^oideaiy ] 18 0 11. r tfcrnibudflookillgmkiydeauinoei 23 4 12. r nlhojTjo%f^Muizuika^i^.daaiykai 24 2 13. [ iietotanoyaaa£iideatg J 20 0 14. [e^^isuiva+ijilBuiuihliaTjanadekakimaBui ] 31 1 15. [ aonodjdoi^ joniyomiea^oaj 22 0 16. [ohayoogozalaaatnJcyoowa±i.ote!£ki.deBuinee] 36 7 17. [wataktu^l waaaagohagninlhoia^okujotabeinaatgi ] 38 0 18. [watakty^i.waaaibaJ£aigkc£±benkyoo£Laaam] 35 3 19. [auizuik^.a eewadonnaaelgBmdeauika ] 31 6 - 1 7 / / 20. [totaeoiiaalOaeadMm] 18 3 $~T 21. t korewanandeai^ka] 15 0 22. [ ao¥ewak^igoeuideaig ] 18 0 23. [watak^i£ ±notokeewatotaaoi£^.Baideai£] 33 3 24. [aoatawatanakaaaanoda^waba^gooo$ ittoleaaraka ] 42 2 25. [ hait£ o^t.oaa^takuidaaa±tanakaaarino<4a£wa bajjgoowagcxka^oiiijioniaanckaroiianadaa^ ] 71 1 * 26. fdyaBoa^lTjatooqozaimaaiu ] 22 6 27. [ lladooita^lJM^if aavc^pa^a ] 26 6 / pronounced ‘ total phoneaea 751 noted errora by 10 judgea 87 as Student 4 Kxpraaaion koifawananlijonodji^ odMiylu) 24 ao^aiMt^yipigokunjonodf i^odaaty 1 2 5 iiawaka£iaaa«M 14 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 2 2 . 23. 24. 25. 26. 27. 20 ^aaaayxiahana^laa^oo 1 ainaiaaaabaoftoyuikkuiriiytakuidaaal] 32 siuklyakitohaabaagaatodot^i^anoboorja awk^deatyka] 44 ryoohootaaodalatykidaaiu ] 22 gakkoowadoodaatyka] 17 to£aoooao^ iroidaaty ] 18 c f e luibuiVp ookii tavykuiedaaviinea] 23 nihorjTiowaaiiizvuka^iidaaiyka] 24 i±atotaaoyaaq£iidaaiy J 20 amp j^uiwa^mtauiuihiifaTjanadakakijiaaiy 31 aonod^ipi^oniyoaiBa^oc) 22 ohayoogo2aiauaiykyoowai±otaijkidaauinaa ] 36 watakiy^ i waaaag ohanninlhoQ ^okuiot a haaut a tu ] 38 watakiu^ iwama ibal^avyko^lbaTjkyoo^iaaam ] auizuik^aafoaaawado^naaajgaaadaamka ] totanoiiaajysaadaaui ] 18 ko^awanaydaamlu) 15 aoj^awaka^igoauidaaiy ] ‘ 18 watakiu ^ inotokaawatotaaot^liaaidaa'y ] 33 anatawatanakaaaynod«)0wabaTjgooo^ittaiJMSwka ] 42 hai^ottoaa^takudaaaitanakaaayD O dajgw a baiigoowagodEafoainonlaafQdEaronanadaaty ] 71 « * * Z dooaoarinatoogoiaimaaiy ] 22 iiadooita^iWjitaaayoona^a] 26 pronouncad total phonaaaa 751 notad arrors by 10 judgaa 1 35 31 Student 5 Bxpraaaion 1. 2. 3. 4 . 5. 6. 7. 8 . 9 . 10. 11. 12. 13. 1 4 . 15. 1 6 . 17. IB. 19. 20. 21. 22. 23 . 2 4 . 25 . 26 . 2 7 . korawanani-gono<^i£ odaai^ikaj ao^awat^inuigokuiTjoncxtjifcxlaavy ] anatawa+iu^aQBunjoijavakaf iaasipka ] iiawakafiaaaa^] djanaarjodahana^JjuiJoo ] sxumleasefa ■o^oymkkmril^tekmdaaai ] suiklyakit.ohaabaagaatodot£ iranohooga atgiki.de a i^ka ] * ryoohootoaodai.aigiki.de8ip ] gritkoowadoodesvpka} cfcuiibuitp ookiltaiykuiedaauineei nihoyjowanuix uikaf lldeaiy kal i.iatoteaoyaaa^iidaaigi ] efpitauiwa+mtauiuiti±£aijanadekaJciaaa^i ] sonod^io i^on iyondjaa^oc) ohavooqozalaaaaikvoowaiiotaTjkldaamo— ] vataktu^ i w a a a a gobanninihoto£okuiotabeaaatp] wataktgi C iwaaaiba^atyko^iJjeTjkyoo^iaaatg ] auizinkiaabaeewadojinaae^aeedeaigika ] tojj.eaoii.se^aaedeaui ] korawanandeauika] ao£awaka£igoauidaatgi ] watakigif in^tokaowatotaaotf ij.aaideaui J ana-tawatanakaaannoda|0wabai^gooo£ i^^a Laaa igika hait^o^'toaattakuidaaaitanakaaapnodabwa baygoowagcxkefronlnoniaaftdf ronanadeaty ] d<gaaoari.qatooaozaiaaam J iiedooltaClaaCiteaavtynara ] 1 pronouncad total phonaaaa notad arrora by 10 judgas ilttakmdasai] Studant 6 Bxpraaaion {ko^awananiTjonod^i^odnat^ka} [ uataw0uiia|iuii|oi|mkariM>iylca ] [ liemkalljMse'ei [ ^aajarpdahana^iaa^oo] 6. [ imaiaaa*^ w t ^ y m M n 7. (auikly akltoha& aa g aatodot£ iJfanohoo^t atgkidaai^ka ^ ^ ■2'^F 8. [ ryoohootoaiodaiaipkidaaip ] 9. [gakkoowadoodaatuka] ' ' * n » 10. [ t otamoomoA Iroidaaui ] T T 7 1 1 11. [ckuilbuijft ooklitau^kiuadaauinafl) l o * " 5 " . 12. [ nihoyrjowanuzmka^iidaagikat 13. [ llatotaaoyaa ^ddaa^] 14. [«pp|t»mwaf iu»uiuihirayanad«kaki»aavy ] 15. [ aono^iol^onlyoadJM^ooj 16. [ o^ayoogo^almaavykyoowaiicrtaijkidafuinaa ] 17. [watakui^i waaaagohanninokmotahaaaaui ] 18. (watakui^ ivaaa±ba{3a wko^iba^kyoo^iaaaiu ] 19. [auizuik^aaftaeawadognaaatoaaadaaigrica] 20. [ totaaoilaatQaaadaa^i ] 21. [ kofavanandaaiuka) 22. [ aotavaka^igoamdaaiu J 23. [watakui^lnotokaawatotaad^ liaaidaaqt] 7 / TT . 24 . [ anatawatanakaaannodajQwabangooo^i^taijMaiyka ] 25. [ hait^ottoaattakuidaa altanakaaaonodaj^wa baijaoc>waao<fca£oninon^aal^chatonanadaam ] 26. [dooaoajfiTjatQpgo^aimaaiji ] 27. [ iladooitajima^itaaavtyna^a ] pronouncad total phonaaaa notad arrora by 10 judgaa Studant 7 Bxprasaion ( t m i o n 1. [ koravana-nitjonod^ odaai^ka] 24 3 2. [ aoravat^ m m gokunjonod^odaa m ] 1 4 \ 25 < 3. [ anatawa4nira n a unpijaailra fl— aiyka ] [ ^.awakarimaama] 30 4 4. 14 2 5. [ <^aaaarp^dahana^±aa^£Ll 20 3 6. (auuaiaaaarc jpttoymkkuifiittakuidaaai] 32 1 7. [ ai^kiy akitohaafe U aaa todot$ lranohootja auikidaai^ka J ^ ^ 2 44 14 8. {fyoohootoaoda±a\£ikidasiu ] 22 0 9. radcJcoywadoadaaiyka ] 17 9 10. (to^uooao Ji^oidaaiji ] 18 3 11. (^uiibuft oo^i-lt^iykuiadaauinaa) 23 14 12. [ nlhoyrjowaiu^uik^i^Ldaaiykail 24 9 13. r iiat.ota«ovaaa^i^.daa^ ] 20 9 14. f ayp ui waa ty y m n hiira-q anada kakiiaa a ty ] 31 14 15. [ aonodjioij^oniyomijiajJoci 22 0 16. (o^ayoogo^aiaaai^ikyoowaiioteTjki.daamDaa ] 36 11 17. [wa^a&iujjl waaaago^annini^pt^okuiotabmasui] 38 19 IB. [ wa^a^iu^ iwa*albata»i^ko(lbe^ky^y*aam ] 35 13 19. [ avnz\uk±aa|0aa«wado£naaa(3aaadaa»^ka ] * ^ * 31 1 20. [ totamoiiaaJOaaadasiji ] [ korawanagdaaiuka] 18 1 21. 15 0 22. [ sofcavaka^igomuidaa^ ] 18 0 23. [wa£a^uiCi_notokaarvatota»ot£iiaaidaagi ] 33 7 24. (anatawatanakaaannodaf»wa£pijgQSp£ i^teiaaaqika1 «2 6 25. [ hait^ottoaat'takvudaaaitanakaaannod^wa ^a-y goowagock a£on inonis al&fc a£onanada a ip ] 71 4 26. f daoai?li]at:oooo^alaaa^ ] 22 4 27. [ iiadooita^ima^taaayoonara ] 26 0 pronouncad total phonaaaa notad arrora by 10 judgaa 751 157 Studant. 8 Sxpraaaioc 1. [ koJfmnuiijono^ odaiifk^ £ 2. [ aormt^uHogoloiiijoiio^^odii^ ] 3. [ aniuw4ittlfania)i|S^awtkA^iau<ulu ] 4. (lia^akafiaaaafrl 5. [ t1jaappT|nriyiana$1aajrtn ] 6. [ ainalaaaafp ao^oym^kmf ti.ttaktudaaai ] 7. [ »iy kivaki tohayby a uto dot-^ljfanohooTja aipkidaatgti^J ‘ 8. [ ryoohootoaodaiatgkidaatgi ] 9. [gzkkoowadoodaaiuka] 7 " ~ v * 10. [ totemoonoj iroidaatg ] 7 11. [cfcuiibuiteookiitauikuiadaaumaai '71“ * 12. [ nUtonriowaanuzuikariidaaiulua *J T T 7 V 7 13. [ ilatotaaoyaaaViidasip ] 14. [anpilauiwa+wlsuiuihi&rganadakakiBaaui] 15. [ aonodjioioo oniycwiea^od) iif T 16. [ohayoogozaiaaawkyoowaiiotagkldaauinaa] 17. [ wataki^i vaaa ag ofaan n^n Uidft^okuiot a haaa aiy ] 18. [watak^±waaaiba^atpka£j£aTjkyoc£iaaaui] 19.. [ aiuzmk-Haaftaaawadoyiaaafea— daaipka ] 20. (fcotoaoiiaa&aaadaai^ ] 21. [ kolrawanagdaaigkaf 22. (»o£awaka^iao«iidaaiy) 23. [ vat akmjj Inot j^aawato^aaot^iisaidaau] ] 24. (anatavatanakaaannodafevabayiooo(ittalmaaiykaJ 25. { hait£o-£to*a^takuidaaaitanakaaa^noda|^va ba^goowagotkafoninonlaa&fco^onanadaaip ] 26. rdooaoa^ijatopoozaimaay ] 27. [ iiadooitajJinaJitaaayjjppaxa ] pronouncad total phonaaaa notad arrora by 10 judgaa 6. 1 1. 12. Student 9 Expression ( kofewananitjonodjiJ odesiuka) [ sotmtj aimgokm y>no<^ ljodt»y J [ anatawa+uiraysuiTjogawakay 1 ■asiyka ] [llevakafiaasaM] [ dyeaeerjodehan^ies^oo ] [ suiaieaseto aottoyrnkkiiiJtiittelcuidasai ] 7 T * 7. [ aiyhiyakityEia^by gyt^d qt^Lranohogqa st^ikidesiy ka ] 9. [ ^v<x? h ootoaodaisiykidssvy ] 9. [j} rttkgpwadoodesiyka ] 10. [^s£eaooBo£iroidest£l [cfeuiibuiK) ^ok y t^mkuiedesuineei [ ni^QTjtjowain guikaf lidaavykal 13. [ liatotaaovasa^lldesiii ] 14 - [eepitsmwafiyj^ uiyihl_?a-qanadekalr1 ■asui ] [ sonodjioi^oniyoe^pu^ocij [ ohayoogpzaieasiykyoowaiiotegkidesuinee ] 17. [ wa£akw£ i waasag oftannini^pQ^ okuiotabemaa^ ] 18. [wa^akuif iwa»aib«<gsuiko^Lbe^ky22.^LBaem ] 19. [suijmk^sslyseewado^iai^aseedesmka] 20. [ tq£eeoii»alQsssdea<y ] 21. [ kof ewanagdesiy ka] 22. [ sof ewake^igoauidastp ] 23. [ watafop^ ino jpkeewato£eeot^ ±isaides\ji ] 24. [ anatavatanakasafnodslBvaba^g^pp^ i^tsiaasvpka 25. [ hait^ottoaa^-takuidasaitanakasaipodfl^wa ba^a<^wa<KX^aEoninon±satj>^eronanadastp ] 26. rdooaoafriqatooootaleasui ] 27. [ iiedooitaCiaa^ite savynaira ] ^ pronounced total phoneaes noted errors by 10 judges 1 5 . 1 6 . e. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. ( Stuctent 10 ixpraaaion [ koEawananlgonodj i^odwiyluj [ ao£awatjuitngokuw|onod^l$odaay ] (anatawa+ui^anaui^ogawaka^inaaiuka ] [ llewBkai Irnaeem ‘ 3 [ d^aaaa Tjodahana^ ina^oo] [ emmlaasew noj^oyuikkmfilttokmdaaai ] •i^klyakitohyb^g^todot^ifanobooya aipkidaaiyka ] (fyoohootoaodaiawkidaaui ] T 3 I [gjfrkoowadoodaagika ] S $ 3 I [ to£a®oomo^if, oiduayj ] [ jfciu Ibua yp ooki ±ia t j i kuiadaaui naa) [ nlhorpjowam^iulca^^.daa\ykai (±iatotaBoyaaa£±ldaaip ] [empj.lsxuwa4mtamuihlf arjanadekakiaasiy ] [ aono^lol^5 oniyo—4 ~~S~1 [ ohayoogozaiaaaijikyoowaiiotaTjkldaauuiaa ] [ vatakin^ i waaa ag ohapninl^pfl ^ okuiotaha»a a iu ] (watakm $ iwa»aiJ>al8^mko§ibaijkyoo$ianaiji ] [ am^,uik|aateaaawadonnaaaiaaaadaa^ka] [ totaaoiiaaftaaadaatp ] [ ko£awanandaatpk4f [ aofrawaka^lqoauidaa [watakuij inotokaavatotaaotjiiaaidaaiji ] (wnatiiwwtanakwannortattwhaijqqflo^ ly>ia>aiaki ] *2 (hait^ottanattakuidaaaitanakaaaynodafawa baTjaoovaqo^aronlnoDlaaOcjta^onanaflaaiy j 71 [ dooaoa£irjatoogo£aimaai£i ] 22 [ iladoolta^lna^taaavoonaEa ] 2 6 pronouncad total phonaaaa notad arrora by 10 jodgaa 24 25 30 14 20 32 44 22 17 18 23 24 20 31 22 36 38 35 31 18 15 18 33 751 0 6 2 9 0 1 2 0 11 10 4 1 114 92 / By checking each student's summary check sheet, the followings observations can be made. The errors ranges from 18 to 189, which means some students are excellent in speaking Japanese, others are not so good. There are particular sounds that some students have troubles. A careful comparison of the ten students’ summary check sheets pointed to some patterns of errors. First, I studied the vowel errors. The errors in vowels are summarized in Table 5. Table 5 Errors in vowels student # vowel changed to different vowel quality duration errors lengthened short vowel shortened long vowel 1 1 10 21 2 2 6 8 3 1 6 20 4 0 2 3 5 2 8 5 6 3 4 13 7 1 1 15 8 0 25 14 9 1 9 11 10 0 4 18 93 Although the errors in vowel quality varied depending on the students, they were small in number. The duration errors, however, were far greater than that of the vowel quality. As predicted by the contrastive analysis of both Japanese and English, and Japanese and Korean, Japanese five vowels pose no particular problems, but, the duration of vowels do cause problems. Next, the errors in consonants were compared. Because there were many types of errors in consonants, they were categorized into groups which were introduced in contrastive analysis. Table 6 shows the types of consonant errors. Table 6 Errors in consonants student # single and geminate consonant distinction shortend or deleted syllabic nasal deleted [h] voiceless/ voiced distinction others 1 6 3 0 1 7 2 4 0 0 2 4 3 4 3 0 0 3 4 2 3 0 2 1 5 5 1 0 2 0 6 5 1 3 12 12 7 2 1 3 15 9 8 8 0 1 7 2 9 6 0 3 5 8 10 5 1 4 11 9 94 All ten students show some degree of difficulty in single and geminate consonant distinction and syllabic nasal. Besides these errors, there are two other types of noticeable errors; [h] deletion and voiceless/voiced distinction. Student #5, 6, 7, 8, 9, and 10 deleted [h], while #1, 2, 3, 4 and 5 did not. Student #6, 7, 8, 9 and 10 made much more voiceless/voiced distinction errors than student #1, 2, 3, 4, 5 did. As mentioned in chapter 3, the deletion of [h] is one of the unique features of Korean language and so is the voiceless/voiced control. Students #1, 2, 3, 4 and 5 speaks English as their first language. Students #6, 7, 8, 9 and 10 speaks Korean. The certain features of the first language seem to influence the students 1 target language pronunciation. The students' errors are analyzed in two groups: errors by the students #1, 2, 3, 4 and 5, and, those by #6, 7, 8, 9 and 10. In order to focus on the students' habits of errors, expressions which contain zero or one mark by a judge were dismissed, because a sound that only one judge marked as unacceptable While the others did not could be accidental. Expressions 3, 4, 17, 21 and 22 spoken by students #1, 2, 3, 4 and 5 were dismissed. Expressions 21 and 22 spoken by 95 students #6, 7, 8, 9 and 10, were also dismissed. In the remaining expressions, words which contained no unacceptable pronunciations were dismissed. Next, words containing the sounds which had been marked unacceptable were chosen and the unacceptable soundB were underlined. The following list shows the errors by the students #1, 2, 3, 4 and 5, English speakers. Expressions 1. jisho 2. chuugokugo 5. hanashimashoo." 6. motto, yukkuri 7. hambaagaa, hoo, 8. ryoohoo, tomo 9. gakkoo, doo 10. totemo 11. zuibun, ookii, tsukue 12. wa, muzukashii 13. yasashii 14. empitsu 15. issho 96 16. ohayoo, kyoo 18. wa, sukoshi, benkyoo 19. Suzuki, sensei, wa, donna 20. totemo, sensei 23. chii sai 24. bangoo, shitte 25. kudasai, denwa, bangoo 26. doorao, arigatoo 27. doo, sayoonara The following list, shows the errors made by the students #6, 7, 8, 9, and 10, English-Korean speakers. Expressions 1. jisho 2. chuugokugo, jisho 3. furansugo, ka 4. iie 5. ja, eigo, hanashimashoo 6. motto, yukkuri, itte 7. to, hambaagaa, to, dochira, hoo 8. ryoohoo 9. gakkoo, doo 10. totemo 11. zuibun, oolcii, tBukue 12. nihongo, wa, muzukashii 13. iie, totemo, yaaashii 14. empitsu, futsuu 15. iBsho, yomimaahoo 16. ohayoo, gozaimasu, otenki 17. watakushi, asagohan, ni hon sh oku, o 18. watakushi, sukostvi, benkyoo 19. Suzuki, donna 2 0. totemo 23. watakushi, tokei, totemo, chiisai 24. bangoo, shitte 25. chotto, matte, kudasai, bangoo, zero, 26. doomo, arigatoo, gozaimaBu 27. doo, sayoonara zero 98 In order to examine the degree of difficulty of pronunciation, the percentage of unacceptability of each problematic phoneme needs to be calculated. For each sound, the number of markings by the judges on a phoneme per total number of chances to be judged is the percentage of unacceptability. If there are a total of 5£ markings on a phoneme in an expression, which are totally acceptable to all judges, the percentage of unacceptability is 100 % (50 markings out of 50 chances to be judged). If there are a total of 2b_ markings on a phoneme in an expression, the percentage of unacceptability is 50% (25 markings out of 50 chances to be judged). In case there are 12_ markings on one phoneme, the percentage of unacceptability is 24% (12 markings out of 50 chances to be judged). In this step, the words which contain unacceptable pronunciation are categorized according to the types and the percentage of unacceptability and sounds heard is computed. The followings are unacceptable pronunciations by students #1, 2, 3, 4 and 5 (English speaking students). The ordering is from more frequent to less frequent. Vowels The followings short vowels were perceived as too long to be a short vowel. 99 sound expression # student’s pronunciation percentage of error [i] 7 tsiykii] or [s\yki+] 10% 15 or [i+|Jo] 6% to] 8 (toomo] or [to+rooj 36% 1 [djijoo] or (dji£o+] 4% 2 [ tjmvugokmijoo ] or [ t^uiuigokuiTjo+ ] 4% 6 [moo^to] or [mo+tto] 4% 19 [doonna] or [do+nna] 4% The! followings long vowels were perceived as too short to be a long vowel ■ sound expression # student’s pronunciation percentage of error [ii] 23 [t^isai] or [tji+saij 24% 12 [muizuika^i] or [muizuika£ i+ ] 8% 13 [yasaji] or [yasaji>] 4% [aa] 7 [haigbagaa] or (hamba+gaa ] 38% 7 [ hambaaga] or [ haqibaaga+ ] 34% [OO] 27 V V [sayonara] or [sayo+nara] 54% 16 [ohayo] or [ohayo* ] 28% 26 [domo] or [do+mo] 28% 9 [ gajfko ] or [ gaJjko+ ] 26% 24 [ baijgo ] or [ barjgo+ ] 16% 100 25 [baijgo] or [ baggo+ ] 10% 26 [afirjato] or [a£itjato+] 20% 8 [fyohoo] or [ryo+hoo] 16% 8 [£yooho] or [?yooho+} 14% 7 [ho] or [ho+ ] 8% 11 [ooki] or [ooki+] 6% 9 [do] or [do+] 4% 16 [kyo] or [kyo+] 4% [ mui] 2 [t^vngokuiTjo] or [tfui+gokuiTjo] 4% The following vowel was perceived as a different vowel according to vowel quality. sound expression student's percentage # pronunciation of error [in] 11 [ tsi^kee] 10% [ tsuikie] 2% Consonants The following single consonants were perceived as geminate consonants. sound expression student's percentage # pronunciation of error It] 10 *Itofrteroo] 28% 20 *[to^temo] 18% 101 (First instance of 'totemo' appeared in #10 and the second in #20.) IS] 23 [tjiijsai] 4% The following consonants. long consonants were perceived as single sound expression student’s percentage # pronunciation of error [tt] 6 [moto] 22% 24 [Jite] 44 [kk] 9 [gakoo] 18% 6 [yvukui^i] 12% tSS i f t 15 |i£o] 12% The following short or deleted. syllabic nasals were perceived as too sound expression student‘s percentage # pronunciation of error [n] 19 [do0na] 14% [ft] • ♦19 [se0see] 22% ♦20 [seOsee] 4% ♦(There were two instances of 'sensei'. One appeared in #19, and the other appeared in #20) 25 [de0wa] 4% 102 The followings consonants were perceived as being incorrect consonants. sound expression student’s percentage # pronunciation of error [d] 25 [kuirasai] 4% [ts] 11 [si^kuie] 8% 14 [empjtui ] 8% [<fe] 25 [ selfo ] 8% [z] 19 [suisuiki ] — o' 10% [suisuiki] 4% 12 [muitj uikajii] 4% [muasuika^ii] 2% The unacceptable pronunciations of students #1, 2, 3, 4 and 5 are summarized in what follow. The percentage of the error occurrence is provided in the parentheses. Vowels [i] and [o] were perceived as lengthened (4% to 36%). [ii], [aa], [oo] and [uiui] were perceived as shortened (4% to 54%). [ui] was perceived as [e] or [i] (4% and 10% respectively). 103 Consonants [tj and [s] ,were perceived as [£t] and [80] respectively (4% to 28%). [tt], [£k] and [^] were perceived as [t], [k] and (4% to 22%). [y] and [E)] were perceived as deleted or shortened. (4% to 22%). [d] was perceived as [r] (4%). [z] was perceived as [s], [^] or (tf] (2% to 10%). [ts] was perceived as [s] or [t\f] (8% for both cases). [ cfe ] was perceived as [s] (8% ). The followings are unacceptable pronunciations by students #6, 7, 8, 9 and 10 (Korean speaking students). The ordering iB from more frequent to less frequent. Vowels The following short vowels were perceived as too long to be a short vowel. sound [i] [«] expression # 17 18 student's pronunciation [ niihojja^okua ] or [ ni+hoft^okui ] [siykojii ] or [smkoJi+] [ ] or [*a+] percentage of error 10% 4% 4% 104 [O] 2 [ t^mmgokmyoo ] or [tJuiujgokuirjo+ ] 10% 17 [oo] or [o+] 8% 12 [ nihoyjoo ] or (nihoipjo+ ] 6% Thei followings long vowels were perceived as too short to be a long vowel * sound expression student's percentage # pronunciation of error [ii] 13 [yasa^ij or [yasa^ it] 32% 12 [mujzuika^ ij or [muizuika£i+J 26% 13 [ie] or [i+e] 4% tee] 5 [erjo] or [e+rjo] 6% [aa] 7 [ hajjibagaa ] or [ hamba+qaa ] 34% 7 [hambaaga] or [hambaaqa+] 28% [oo] 27 v V [sayonara] or [sayo+nara] 22% 9 [gakko] or [gakko+] 18% 11 [okii] or [o+kii] 16% 25 [ ba^go ] or [ barjgo+] 16% 7 [ h2.1 or I hot l 12% 15 [ycmima^o] or [yomimajo+j 12% 5 [hanajima^o] or [hana^imaJ*o+] 10% 8 (£yohoo] or [ryo+hoo] 10% 8 [ryooho] or [£yooho+] 4% 26 v V [aritjato] or [arirjatot] 10% 105 26 [dorao] or [do+mo] 8% 27 [do] or [dot] 6% 16 [ohayo] or [ohayo+] 4% 18 [bei^kyo] or [bei^kyo+] 4% [uiui] 14 [fiutsiu] or [#uatsui+] 12% a o ----- 2 [ tjuigokunjo] or [tJua>gokuaijo] 8% The following according to vowel sound expression # [a] 7 [o] 17 vowels were perceived quality. student's pronunciation [hajbaagaa] [ nihobjm km ] as different vowels percentage of error 18% 8% Consonants The following single consonants were perceived as geminate consonants. sound expression student's percentage # pronunciation of error [t] *10 [ tottemo) 32% *13 [tottemo] 10% 20 [tottemo] 2% (* The first instance of 'totemo' appeared in #10, the second in #13 and the third in #20.) 106 [k] 11 (ookkii) 22% The following geminate consonants were perceived as single consonants. sound expression student * s percentage # pronunciation of error [tt] 6 (moto] 22% 24 [Jite] 20% 25 Itjoto] 12% 25 [mate] 10% [kk] 6 [yuikui^i] 22% 9 [gakoo] 10% [tf] s 15 Ujo] 26% The syllabic nasal [n] was perceived as too short or deleted. sound expression student's percentage # pronunciation of error Eg] 19 [do0na] 20% [h] was perceived as deleted. sound expression student's percentage # pronunciation of error [h] 8 [£yoo0oo] 16% 12 [ni0ojjijo) 14% 107 17 [ni0oft£okui] 17 [asaTjo0af2] 16 [o0ayoo] 10% 10% 8% The following consonants were perceived as being incorrect consonants. sound expression student’s percentage # pronunciation of error [b] 7 [ hantpaagaa ] 10% *25 [£aijgoo] e 10% *25 [£apgoo] 8% (* In #25 the word "bangoo" appeared twice.) m *i8 [wadakw^i) 16% *17 [wadaki^i] 14% (* The first instance of 'watakushi' appeared in #17, an the second in #18.) 23 [dokee] 14% 16 [oderjki ] • 4% [k] *17 [wata£ui^i] 14% *18 [watacpu^i] 12% *23 t watagui^ i ] 12% (*The first instance of 'watakushi' appeared in # 17, the second in #18, and the third in #23.) [g] 9 [kakkoo] 38% 108 (ts] 14 [«®pit£uil 42% 14 [+uat£uiui ] 38% 11 [tfiykuie] 22% [si^kuie] 14% [dz] 11 [ djui ibuil^ ] 32% [ tsui ibuife] 8% *25 [djero] 16% *25 v (djero] 16% (* In #25, 'zero' appeared twice.) [dj] *1 28% *2 20% (*The first instance of 'jisho' appeared in #1, and the second in #2. ) [<M 3 [pvu^ansuigo] 8% [s] 18 l I ' V k o iil 4% [z] 12 [ mui d£Ui ka^ ii ] 22% [muit^uika^ii] 6% [muitsuaka^ii] 4% [mm sui ka^ii ] 2% *16 [ god^ aimasiy ] 32% [ got^ aimasiy ] 2% *26 [ god^ aimasm J 22% (* There were two instances of 'gozaimasu'. One was in i and the other was in #26.) 109 19 [suid^uiki ] [suitjuik^] [Siujmkj] 26% 4% 2% iSl 17 [ nihoQcfeokvu ] 12% [nihoftsokuj ] 4% t — The unacceptable pronunciations of students #6, 7, 8, 9 and 10 are summarized below. The percentage of the error occurrence is provided in the parentheses. Vowels [i]f [a] and [o] were perceived as lengthened (4% to 10%) . [ii], [ee], [aa], [oo] and [uiui] were perceived as shortened (4% to 34%). [a] was perceived as [*] (4% to 18%). [o] was perceived as [ui] (8%). Consonants [t] and [k] were perceived as [tt] and [kk] respectively (42% to 32%). [tt], [kk] and [££] were perceived as [t], [k] and [X ] respectively (10% to 26%). [n] was perceived as deleted or shortened (20%). 110 [h] was perceived as deleted (8% to 16%). [b], [g] and [d^] were perceived as ip], [k] and [tj] respectively (8% to 38%). [t], (k] were perceived as [d] and [g], respectively (4% to 16%) . [tB] was perceived as [t^] (14% to 42%). [dz] was perceived as [d^] or [ts] (8% to 32%). [ < ( > ] was perceived [s] was perceived [z] was perceived 32%). ] was perceived respectively). Ill as [p] (8%). as [J] (4%). as [dj], [t£j, [ts], [s] or [§] (2% to as [s] or [cte) (4% and 12% As the final stage of the investigation, the students' predicted difficulties and how their actual errors were compared to see if prediction was correct. Every error by the students fits into one of the following categories. 1) Errors made as predicted: 2) Errors which occurred differently from the prediction: 3) Errors which were predicted but did not occur: Items in category 1) and 2) need review and further analysis about how the sound is perceived and modified. Items in category 3) can be dismissed since they were perceived as acceptable pronunciations. First, predicted difficulties of students #1, 2, 3, 4 and 5, English speakers, and how the actual errors were compared. In the following comparison, prediction by the contrastive analysis is followed by the arrow and the actual errors perceived by the ten judges. Vowels 1) Errors made as predicted: Long vowels, [ii], [ee], [aa], [oo] and [uiui], were 112 predicted to be pronounced too short to be a long vowel. — > Except for [ee], long vowels, [ii], [aa], [oo] and [uiui] were perceived as too short to be a long vowel. In English, tense vowels are longer than lax vowels, but the duration of the longer tense vowel is not as long as the corresponding Japanese long vowel. A Japanese long vowel is approximately 2.5 to 3.0 times as long as a short vowel. English speakers who tend to shorten long vowels seem to carry the durational habit of an English tense vowel when pronouncing Japanese long vowels. 2) Errors which occurred differently from the prediction: [i] and [o] were not predicted to be mispronounced. — > [i] and [o] were perceived as too long to be a short vowel. It is plausible that the students pronounced them as diphthongs [iy] and [ow] as they are diphthongized in English. [ui] was not predicted to be mispronounced. — > [ ui ] was perceived as [i] or [e]. 113 There seems to be only one reason why [ ui ] was perceived as Ii 1 or [e]. [ui] and [i] share the similar quality of the tongue height and they are both unrounded vowels, [ui] and [e] are also unrounded vowels. 3) Errors which were predicted but did not occur: No instances were perceived. Consonants 1) Errors which were perceived as predicted: [ts] was predicted to be pronounced as [t], [t£] or [s]. — > [ts] was perceived as [t$] or [s], but not as [t]. [tsJ is an alveolar affricate. It seems that if the students are overconsciouB about the point of articulation at alveolar, they may pronounce [s] instead, and if the students pay more attention to the affricate quality, they may pronounce [t$] instead. Geminate consonants [pp], [tt], [kk], [ss], and [ ] were predicted to be pronounced as single consonants. -> [tt], [kk] and [ ], were perceived as single 114 consonants, It], [k] and [\] respectively ([pp] and [ss] « • were not included in the material). Geminate consonants do not occur within a word in English. The students may pronounce them as single consonants because they are not familiar with the duration of the geminate consonants. 2) Errors which occurred differently from the prediction: [dz] was predicted to be [d], [dj] or [z]. — > [dz] was perceived as fa]. In English, there is no [dz], a voiced alveolar affricate. If the students are not aware that it is a voiced sound and pay attention to its alveolar quality , this error may happen. [y] and [£] were not predicted to cause mispronunciation. — > [n] and [G] were perceived as deleted or shortened. In Japanese, [g] and [ | Q ] are syllabic nasals, but in English, there is no syllabic nasal which has the same length as Japanese syllabic nasals. 115 Single consonants were not predicted to cauBe mispronunciation. — > the single consonants [t] and [s] were perceived as the geminate consonants l^t] and [ss] respectively* This type of error seems to happen when the students are not used to the distinction between the geminate and single consonants. 3) Errors which were predicted but did not occur: [ < t » ] was predicted to be pronounced as [f] or [hj. — > No instances were perceived. [x] was predicted to be pronounced as [h]. — > No instance were perceived. [fa] was predicted to be pronounced as [m], [n] or [tj]. * • — > No instances were perceived. [r] was predxcted to be pronounced as [1] or [r]. — > No instances were perceived. Next, predicted difficulties of students #6, 7, 8, 9 and 116 10 (English-Korean speakers) and the errors were compared. Vowels 1) Errors which were perceived as predicted: The long vowels [ii], [ee], [aa], [oo] and [mm] were predicted to be pronounced as too Bhort to be a long vowel. — > The long vowels, [ii], [ee], [aa], [oo] and [mm] were perceived as shortened. In Korean, although there is a distinction between long vowels and short vowels, it is not exactly the same as that in Japanese. Korean long vowels are shorter than Japanese vowels. Therefore, Japanese long vowels pronounced by English-Korean speakers tend to be too short to be long vowels. 2) Errors which occurred differently from the prediction: Short vowels were not predicted to be pronounced as too long to be a short vowel. — > Short vowels [i], [a] and [o] are perceived as lengthened. 117 If English-Korean speakers transfer their knowledge of their other base language, English, and consider these short vowels to be like diphthongs [iy], [aw] and [ow] respectively, it may explain why they are perceived by the judges as lengthened. [a] was not predicted to be mispronounced. --> [a] was perceived as [e]. The vowel [a] is perceived as [e] only in the word 'hambaagaa' (hamburger). In English,the the word is pronounced [hwnbdr:gBr]. It seems that [e] is pronounced instead of [e] rather than [a] just because the students use their knowledge of English here rather than Korean. Both [a] and [e] are front vowels. [o] was not predicted to be mispronounced. — > [o] was perceived as [uij. Both [o] and [uj] are back vowels. 3) Errors which were predicted but did not occurt No instances were perceived. 116 Consonants 1) Errors which were perceived as predicted! [p]r [t], [k] and [£] were predicted to be voiced intervocalically. — > [t] and [k], except for [p] and [X], were perceived as voiced initially and intervocalically. English-Korean speakers seem to carry over the voicing ruleB of Korean when speaking Japanese. [b], [<*]# [g] and [?f] were predicted to be voiceless word-initially. — > [b], [g] and [J] were perceived as voiceless ^ord-initially or word-medially, but [d] was not. v In Korean, [p], [t], [k] and [c] become voiced when they occur intervocalically, while remaining voiceless unless they occur between voiced sounds. It may be reasonable to analyze that [b] and [d], the voiced counterparts of [p] and [t], become voiceless if they do not occur between voiced sound. [ts] was predicted to be pronounced as [Sf] or [s]. — > [ ts ] was perceived as [ t$ ](*■[$]), but not as [ s ]. 119 If English-Korean speakers pay more attention to the affricate quality than the alveolar one, they may pronounce [t^] instead of [ts]. (Korean (£] and Japanese [t^] have similar qualities. Both are affricates, and the point of articulation of [£] is palatal and that of [t^] is pre-palatal. Therefore, f£] could be perceived as [tj] by the judges.) [ dz ] was predicted to be pronounced as [jf] or [<»;)• — > [ dz ] was perceived as [ d^ ] (= [ ]). If English-Korean speakers pay more attention to the affricate quality than alveolar one, they may pronounce [dj] instead of [dz]. (Korean [] and Japanese [d^] share similar qualities. Both are affricates and the point of articulation of [^] is palatal and that of [d£] is pre-palatal. Therefore, [^] could be perceived as [d^ ] by the judges.) [4>] was predicted to be pronounced as [p] or [h]. — > [4>] was perceived as [p], but not as [h]. It seems that English-Korean speakers tend to pronounce [p] 120 because [+] and [p] share the same point of articulation, bilabial. [^] was predicted to be pronounced as [s]. — > [^] was perceived as [s]. Since there is no [^1 in Korean, [s] may be substituted for [£] because both sounds share the same manner of articulation, which is fricative, and similar point of articulation, which is alveolar and pre-palatal. [h] was predicted to be deleted. — > [h] was perceived as deleted. The deletion of [h] is one of the unique features of Korean. English-Korean speakers tend to apply this deletion to Japanese. The geminate consonants [pp], [tt), [kk], [ss] and [^$] were predicted to be pronounced as single consonants. — > [tt], [kk] and were perceived as single consonants • • • ([pp] and [ss] were not included in the material). • • In Korean there is no geminate consonants which have the 121 same quality as in Japanese. Therefore, Japanese geminate consonants may be pronounced too short or as single consonants. 2) Errors which occurred differently from the prediction: [dz] was not predicted to be mispronounced. — > [dz] was perceived as [ts]. If [dz] is considered to be very similar or the same as []f] by English-Korean speakers, [dz] may become devoiced [ts] because of the same phonetic rule that governs [b], [g] and [dj]. [s] was not predicted to be mispronounced. — > [s] was perceived as [^ ] . This error seems to be a random error, because [^] does not occur in Korean. [^] was not predicted to be mispronounced. — > [^] perceived as [dz]. This error seems to be a random error, because [dz] does not occur in Korean. 122 [ z ] was predicted to be pronounced as [ ]. — > [z] was perceived as [ts], [t^ ], [dj] or [s]. In Korean there is no [z], the voiced alveolar fricative. The closest Korean sounds to [z] are [^] and [s]. If English-Korean speakers pronounced [z] as [3f], the judges could have perceived as [<^J. Moreover, if English-Korean speakers apply the devoicing rule to [d^], it can be understood why the sounds [ts], [t^] and [s] were heard by the judges, [ts] and [t^] are considered to be the voiceless counterparts of [d^], and [s] is that of [z]. [9] was not predicted to be mispronounced. — > [£] was perceived as deleted or shortened. In Japanese [y] is a syllabic nasal, but Korean has no syllabic nasal having the same quality as [y]. Japanese [y] may be pronounced as a mere nasal [n] by Korean speakers. Single consonants were not predicted to be mispronounced. — > Single consonants [t] and [k] were perceived as geminate consonants, [£t] and [kk] respectively. If the students are not used to the distinction between the 123 geminate consonants and single consonants, this type of error can happen. 3) Errors which were predicted but did not occur: No instances were perceived. 124 Chapter 6 Conclusion This thesis studied the pronunciation problems that English speaking and English-Korean speaking students encounter when speaking Japanese. In Chapter 1, the sound system of Japanese as the target language, in Chapter 2, the sound system of English and in Chapter 3, the sound system of Korean as the base languages were presented. Vowels and consonants were described phonemically as well as phonetically. The unique features of each language, such as sound distributions and , allophones were explained also. This was the foundation of the analyses presented in this study. In Chapter 4, contrastive analyses of sound systems of Japanese and English, and Japanese and Korean were studied. In each analysis, the sound systems of the target language and the base language were contrasted, and the predictions of difficulties were formulated through the application of a hierarchy of difficulty. In the prediction, examples of predicted difficulties by English speaking and English-Korean speaking students were presented. In Chapter 5, error analyses of Japanese pronunciation 125 by English speaking and English-Korean speaking students were presented. In this chapter, an experiment was conducted to collect pronunciation errors that the students actually made when speaking Japanese. Ten native Japanese speakers listened to each student's utterances and judged which sound was acceptable or not. After that, the number of errors and the percentage of error occurrences in each expression, and sounds of the target language which involved the students' errors were summarized. Two groups of the students emerged after the students' errors were categorized according to the types: a group of English speakers and a group of English-Korean speakers. In each group, the percentage of error occurrences was calculated and the students' errors were presented. As the last stage of this section, the students' predicted difficulties and how their actual errors were perceived by the judges were compared to confirm which prediction was correct. Errors that judges perceived were analyzed. The followings are questions that were raised when this thesis was started. 1) Is the contrastive analysis necessary? Is the error analysis sufficient without contrastive analysis? 126 2) Does the types and number of errors differ among the students? 3) Do English-speaking and English-Korean speaking students make same types of errors? 4) What are the frequent errors of English and English-Korean speaking students? 5) Do the students' errors occur randomly? 6) What is the factor that causes errors? The followings are what I found through the study. They will answer the questions raised in the introduction. 1) Contrastive analysis is necessary to provide the foundation for the comparison between the target language and the base language. Without the contrastive analysis, the differences between the two languages can not be learned. The error analysis is not sufficient without the contrastive analysis. Data on the students' errors obtained in the error analysis can be random errors if there were no base provided by the contrastive analysis. Also, the contrastive analysis has to be accompanied by the error analysis, because without the error analysis, the students' actual errors are never learned. As studied in Chapter 5, 127 there were errors which occurred differently from the prediction and those which were predicted but did not occur. This confirms that both analyses need each other. For example, in case of errors by English speaking students, although vowels [i], [o] and [ui] were not predicted to be mispronounced, the students made errors. Also, consonants V [$], [x], [r] were predicted to be mispronounced but they were not. As for English-Korean speaking students, short vowels were not predicted to be mispronounced, however, they resulted in errors. Consonants [dz], [s], [^] and [z] were also mispronounced while no difficulties were predicted. Also, syllabic nasal [n] and single consonants were not predicted to be mispronounced, but they caused errors. 2) The types and number of the errors vary among the students. This indicates that some students speak with Japanese-like pronunciation but others speak with a heavy foreign accent. 3) Both English speaking and English-Korean speaking students tend to make same errors on sounds which have length distinctions. Long and short vowels, geminate and single consonants, and syllabic nasals are problematic 128 sounds for both types of students because these sounds do not occur in English and Korean. However, English-Korean speakers make more pronunciation errors than English speakers. Also, their errors vary more probably because they have varieties of sound assimilation in Korean. 4) Followings are examples of frequent errors. Errors by English speaking students Short vowels were pronounced too long and heard as long vowels by the judges. Long vowel were pronounced too short and heard as short consonants by the judges. Single consonant was pronounced too long and heard as geminate consonants by the judgeB. Geminate consonants were pronounced too short and heard as single consonants by the judgeB. Syllabic nasals were pronounced too short or deleted, therefore, it was not heard by the judges. Errors by English-Korean speaking students Short vowels were pronounced too long and heard as long vowels by the judges. Long vowels were pronounced too short and heard as 129 short vowels by the judges. Single consonants were pronounced too long and heard as geminate consonants by the judges. Geminate consonants were pronounced too short and heard as single consonants by the judges. Syllabic nasal was pronounced too short or deleted, and heard as deleted by the judges. [h] was pronounced deleted ,therefore, it was not heard by the judges. Voiced consonants, [b], [d] and [g] were pronounced as voiceless consonants and heard as [p], [t] and [k] respectively by the judges. Voiceless consonants, [t] and [k] were pronounced as voiced consonants and heard as [d] and [g] respectively by the judges. Alveolar affricates, [ts] and (&] were pre-palatalized and heard as [t^] and [dj] respectively by the judges. 5) Some of the students' errors could be random errors but most of them are not. The students tend to pick up the most similar sound to the target language in the sound systems of their base languages. The most similar sound means what has the same or similar manner or point of 130 articulation. 6) Most errors reflect the student's base language. If a student is an English speaker, he or she first interprets the Japanese sounds in the English sound system. While interpreting, the Japanese sound become transformed into English style sounds. In case of a student whose first language is Korean, he or she seems to first filter the Japanese sounds through Korean sound system. During the filtering, Japanese sounds are influenced by the unique sound rules and come out as Korean style sounds. How to correct the students’ pronunciation will need further work, however, it is necessary to know that teaching of pronunciation should not be done uniformly; there must be one way to teach to English speaking students, another way to English-Korean speaking students, and even other ways to other language speaking students. 131 Bibliography Akmajian, A, et al. Linguistics: An Introduction to Language and Communication. 2nd ed. Cambridge, Hass.: HIT Press, 1984. Block, B. "Studies in Colloquial Japanese IV: Phonemics" Language 26, 1950. Brown, D. H. Principles of Language Learning and Teaching. 2nd ed. NJ: Prentice-Hall, Inc, 1987. Han, H. S. Acoustic phonetics of Korean. Los Angeles: University of California, Los Angeles, 1962a. Han, H. S. Japanese Phonology: An Analysis Based upon Bound Spectgrams. Tokyo: Kenkyusha, 1962b. Han, H. S. "The Timing Control of Geminate and Single Stop Consonants in Japanese: A Challenge for Nonnative Speakers" Phonetics 49. Ed. Kohler. K. Switzerland, 1992. Jorden, E. H. Japanese Language Instruction in the United States: Resources, Practice, and Investment Strategy. Hassachusetts: The National Foreign Language Center, 1991. Kim, Kon-On Temporal Structure of Spoken Korean; An Acoustic Phonetic Study. Los Angeles: University of Southern California Dissertation, 1974. 132 Kim, Nam-Kil "Korean" The Major Languages of East and South-East Asia. Ed. Cotnrie. B. London: Routeledge, 1990. Ladefoged, P. A Course in Phonetics.: Harcourt Brace Jovanovich, 1975. McManis, C, et al. Language Files. 5th ed. Ohio: Ohio State University Press, 1991. Martin, S. E. "Korean Phonemics” Language. New York: American Council of Learned Societies, 1958. Prator, C. H. Hierarchy of Difficulty. Unpublished classroom lecture, University of California, Los Angeles, 1967. Prator, C. and Robinette, B. A Manual of American English Pronunciation. 3rd ed. New York: Holt, Rinehart and Winston, 1973. Sato, Yumiko. "The Duration of Syllable-Final Nasals and the Mora Hypothesis in Japanese" Phonetics 50. Ed. Kohler. K. Switzerland, 1992. Trager, G. and Smith, H. L. An Outline of English Structure. Oklahoma: Battenberg press, 1951. Vance, T. J. An Introduction to Japanese Phonology. Albany: State University of New York Press, 1987. Whitman, R. Contrastive analysis: Problems and Procedures. Language Learning 20, 1970. 133 INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UM3 films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note w iD indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs inchided in the original manuscript have been reproduced xerographically in this copy. Higher quality 6“ x 9* black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. A Bel! & Howell Information Company 300 North Z eeb Road Ann Arbor, M i 46106*1346 USA 313/761-4700 800/521-0600 UMI Number: 1376521 UMI Microfora 1376521 Copyright 1995, by UMI Coapany. All rights reserved. This aicrofora edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The Japanese demonstratives ko, so and a
PDF
Locatives in Chinese and Japanese: distribution, case assignment and c-command
PDF
Predictors of pretend play in Korean-American and Anglo-American preschool children
PDF
An investigation of the pseudohomophone effect: where and when it occurs
PDF
A thesis
PDF
The dream becomes a reality (?): nation building and the continued struggle of the women of the Eritrian People's Liberation Front
PDF
Perceptions and realities: women, equality and discriminatory barriers in Japanese society and workplace
PDF
The relationship between alcoholism and crime: autonomic and neurpsychological factors
PDF
Meta-analysis on the misattribution of arousal
PDF
Work for the masters degree
PDF
On the limits of not being scripted: video-making and discursive prositioning in coastal southeast Sulawesi
PDF
All in fun: a translation with an introduction
PDF
The use of occupational therapists or interdisciplinary teams in the evaluation of assistive technology needs of children with severe physical disabilities in Orange County schools
PDF
Structural equation modeling in educational psychology
PDF
Fine motor skills of two- to three-year-old drug exposed children
PDF
A qualitative study on the relationship of future orientation and daily occupations of adolescents in a psychiatric setting
PDF
Dye laser characterization of two films of baceriorhodopsin
PDF
Children's immediate reactions to interparental conflict
PDF
The house that Jack built
PDF
Superinsulation applied to manufactured housing in hot, arid climates
Asset Metadata
Creator
Takemori, Kanzo
(author)
Core Title
Japanese pronunciation by English speaking students and English-Korean speaking students: contrastive and error analysis
School
Graduate School
Degree
Master of Arts
Degree Program
East Asian Linguistics
Degree Conferral Date
1995-08
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
education, language and literature,language, linguistics,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Han, Meiko S. (
committee chair
), [illegible] (
committee member
), Kim, Nam-Kil (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c18-5386
Unique identifier
UC11357629
Identifier
1376521.pdf (filename),usctheses-c18-5386 (legacy record id)
Legacy Identifier
1376521-0.pdf
Dmrecord
5386
Document Type
Thesis
Rights
Takemori, Kanzo
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
education, language and literature
language, linguistics