Selected Annotated Bibliography: Japanese Learner Language

Sachiko Horii, CARLA, University Minnesota, Dec. 12, 2011

A note on organization:
Within each topic area (e.g. tense-aspect), all research studies focused on that topic are listed in chronological order, beginning with the oldest and ending with the most recent, assuming that each research study entails and builds on the findings of previous studies.

Meta-analysis of Research on Japanese Learner Language

Many of the studies reviewed in the following publication are not in English and/or not available in the U.S.

Mori, Y., & Mori, J. (2011). Review of recent research (2000–2010) on learning and instruction with specific reference to L2 Japanese. Language Teaching, 44, 447-484.

This article offers a comprehensive overview of research on L2 Japanese learning and teaching. The authors reviewed over 200 empirical studies published between 2000 and 2010, covering a wide range of topics including vocabulary and kanji learning, interlanguage grammar development, reading and writing in L2 Japanese, interlanguage pragmatics and language socialization, affective factors in L2 Japanese learning, multilingualism and acquisition of Japanese as a heritage language, as well as instructional approaches and curriculum design in teaching Japanese.

Tense-Aspect

The studies below look at the form '-te –i' that denotes either progressive or resultative meaning. Their findings are mostly consistent: resultative is more difficult to acquire.

許夏珮（1997）「中／上級台湾人日本語学習者による『テイル』の習得に関する横断研究 」『日本語教育』95, 37-48 [tr. Sheu, S. (1997). A cross-sectional study of the acquisition of –te iru by intermediate and advanced Taiwanese learners of Japanese, Journal of Japanese Language Teaching, 95, 37-48.]

Sheu administered a picture description task and a grammar test to 60 Japanese L2 learners with Taiwanese L1 (30 living in Japan and 30 living in Taiwan). She discovered that intermediate/advanced level learners find the following order of difficulty in contexts for the acquisition of -te iru: (beginning with the most difficult) past experience > counterfactual > resultative > progressive > habitual > adjectival > idiomatic usage > personal affiliation.

Shirai, Y., & Kurono, A. (1998). The acquisition of tense-aspect marking in Japanese as a second language. Language Learning, 48(2), 245-279.

Shirai and Kurono conducted two experiments. In the first experiment they examined interview data from three Japanese L2 learners (Chinese L1) conversing with a Japanese native speaker and found that the learners used more activity verbs for –te- i- than native speakers did. In the second experiment they administered three multiple-choice grammar tests during the nine-month period to 17 Japanese L2 learners from different Asian countries. The authors discovered that L2 learners found it more difficult to distinguish the correct uses of the resultative –te-i.
(In Japanese, see also 黒野敦子(1995)「初級日本語学習者における「–テイル」の習得について 」『日本語教育 』87, [tr. Kurono, A. (1995). Journal of Japanese Language Teaching, 87].)

許夏珮（2000）「自然発話における日本語学習者による『テイル』の習得研究 ─OPIデータの分析結果から─」『日本語教育』104, 20−29　[tr. Sheu, S. (2000). Acquisition study of –te iru in Natural speech by Japanese learners: Based on an analysis of OPI data. Journal of Japanese Language Teaching, 104, 20-29.]

Sheu examined the use of te- iru in OPI (Oral Proficiency Interview) data of 90 Japanese L2 learners with Chinese, English, and Korean L1 (5 beginning, 10 intermediate, 10 advanced, and 5 superior level learners). She found a developmental sequence across the 3 proficiency levels in contexts for –te iru that was common among different L1 speakers as follows: Progressive > static > repetitive > resultative > change of state > perfective.

Ishida, M. (2004). Effects of recasts on the acquisition of the aspectual form -te i- (ru) by learners of Japanese as a foreign language. Language Learning, 54(2), 311-394.

Ishida examined how recasts may affect learner accuracy in producing imperfective –te-i. Three native speakers of English and one Chinese/English bilingual learning Japanese L2 at a US university participated in eight weekly conversation sessions in which the author provided recasts as well as other types of corrective feedback. The results show that learner accuracy significantly increased as they were provided with recasts. The researcher also found that they produced the form –te-i- with its resultative meaning more accurately than with the progressive meaning. The paper includes a comprehensive summary of previous studies in a table.

塩川絵里子(2007)「日本語学習者によるアスペクト形式 『テイル』の習得―文末と連体修飾説との関係を中心に―」 『日本語教育 』134, 100-109 [tr. Shiokawa, E.(2007). Acquisition of –te iru by L2 learners of Japanese. Journal of Japanese Language Teaching, 134, 100-109].

While previous research mainly focused on the acquisition of –te iru in main clauses, Shiokawa examined how Japanese L2 learners acquire the form in noun-modifying clauses. Commonly, –te iru is associated with activity verbs and –ta with achievement/accomplishment verbs in main clauses, but in noun-modifying clauses, both –te iru and –ta can be associated with achievement/accomplishment verbs. Shiokawa administered a grammaticality judgment test to 63 college students learning Japanese L2 in Japan (21 beginning, 21 intermediate, 21 advanced learners). She discovered that the non-native-like tendency to associate te iru with activity and –ta with achievement/accomplishment was also found in the grammaticality judgments of noun-modifying clauses by beginning and intermediate learners.

Sugaya, N., & Shirai, Y. (2007). The acquisition of progressive and resultative meanings of the imperfective aspect marker by L2 learners of Japanese: Transfer, universals, or multiple factors? Studies in Second Language Acquisition, 29, 1-38.

Analyzing data from an oral picture description task and a grammar acceptability test administered to 80 Japanese L2 learners living in Japan, Sugaya and Shirai examined how the lexical aspect of verbs in learners' first language may influence the acquisition of Japanese L2 –te- i-ru. Results from the acceptability task showed that lower proficiency learners more commonly associated activity verbs with the progressive. In addition, although learners whose L1 had a progressive marker judged progressive meaning more accurately than those whose L1 did not, overall, the progressive seemed easier to judge than the resultative. However, on the oral picture description task the lower proficiency learners behaved differently; they did not associate te-i-ru with activity verbs. It is concluded that task type interacts with transfer and proficiency level in the acquisition of imperfective aspect in L2 Japanese.
(In Japanese, see also: 菅谷奈津恵(2004)「文法テストによる日本語学習者のアスペクト習得研究 」『日本語教育 』123, 56-65. [tr. Sugaya, N. (2004). Journal of Japanese Language Teaching, 123, 56-65].)

Gabriele, A. (2009). Transfer and transition in the SLA of aspect: A bidirectional study of learners of English and Japanese. Studies in Second Language Acquisition, 31, 371-402.

Gabriele investigated the effects of transfer in acquiring aspectual semantics of the present progressive and resultative marker, te-iru, by English learners of Japanese and the present progressive marker, be-ing, by Japanese learners of English. She gave an interpretation task to 101 Japanese learners of English, 23 English native speakers, 31 English learners of Japanese and 33 Japanese native speakers. The participants were asked to listen to a short story with two pictures and then asked to judge whether18 sentences were compatible with the story or not. The researcher asked whether L2 learners can suppress L1 influence by ruling out interpretations available in L1 but not L2. The results show that L2 English learners had more difficulty than L2 Japanese learners in doing this. Learners of both L2s performed accurately on accomplishments; the data on achievements was more mixed, with the interaction between achievements and either progressive be-ing or imperfective te-iru presenting a greater challenge. The study was the first to examine not just acquisition but preemption contexts presenting problems for learners of both L2s.

Relative clauses

The original studies were based on Keenan and Comrie's (1977) noun phrase accessibility hierarchy (NPAH); it has been found that the accuracy order of relative clause types in learner language follows NPAH and is consistently as follows: SU>DO>IO>OBL>GEN>OCOMP among European languages. However, findings on the acquisition orders of relative clauses (RC) vary in JSL.

Kanno, K. (2007). Factors affecting the processing of Japanese relative clauses by L2 learners. Studies in Second Language Acquisition, 29, 197-218.

Kanno collected and examined data of a listening comprehension task from novice Japanese L2 learners of five Asian language groups. She found that L2 learners were more accurate in comprehending subject-gap RCs than object-gap RCs. However, participant proficiency level was so low that the results showed inconsistencies because it was difficult for them to process the test sentences.

Ozeki, H., & Shirai, Y. (2007). Does the noun phrase accessibility hierarchy predict the difficulty order in the acquisition of Japanese relative clauses? Studies in Second Language Acquisition, 29, 169-196.

Ozeki and Shirai pointed to inconsistent findings on the L2 acquisition of Japanese RC. Drawing on Comrie's new typology of modifying clauses, they explained that many Asian languages have noun-modifying clauses called "attributive clauses" and should be differentiated from RC in European languages. The paper consists of two studies. Using different data corpora, the first study examined noun-modifying clauses in the OPI data of 90 Japanese L2 learners with Chinese, English, and Korean L1 (5 beginning, 10 intermediate, 10 advanced, and 5 superior level learners) in comparison with data from 15 Japanese native speakers. Their findings show that intermediate English L1 learners used DO relatives more frequently than OBL and SU, while advanced learners used SU relatives most frequently. In addition, they found that intermediate learners tended to use animate head nouns with SU and inanimate head nouns with DO/OBL while this tendency was not found among advanced learners. The second study examined accuracy for each RC type with animate/inanimate head nouns using data from a sentence-combining test given to 50 Cantonese college students learning Japanese L2 in Hong Kong, with learners at the intermediate and advanced levels. They found the accuracy order to be SU>DO>OBL regardless of animacy. Thus, these two studies show inconsistent results in the acquisition order of Japanese L2 RC, with lower proficiency learners tending to associate SU with animate head nouns and DO/OBL with inanimate head nouns.
(In Japanese, see also: [tr. Ozeki, H. (2005). Does the acquisition of noun-modifying constructions in L2 Japanese follow the noun phrase accessibility hierarchy? Acquisition of Japanese as a Second Language 8, 64–82].)

Yabuki-Soh, N. (2007). Teaching relative clauses in Japanese: Exploring alternative types of instruction and the projection effect. Studies in Second Language Acquisition, 29, 219-252.

Yabuki-Soh examined how three different types of instruction affect learner acquisition of RC in Japanese L2 and compared the results with NPAH. Sixty college students from different L1 backgrounds (English, Chinese, and Korean, a few European and other languages) in three groups received three 50-minute-long sessions of either form-focused, meaning-focused, or form and meaning focused instruction and took comprehension and sentence-combination tests before and after the sessions. The results show that the form-focused group scored higher than the other groups on the tests. No significant difference was found due to the learners' L1 backgrounds. In addition, the results were not consistent with NPAH.

Negation

Kanagy, R. (1994). Developmental sequences in learning Japanese: A look at negation, Issues in Applied Linguistics, 5(2), 255-277.

Kanagy examined structured interview data gathered over one academic year from 12 beginning learners of Japanese L2 at two US universities. She found that although all the negation forms had been introduced when the study began, her L2 learners gradually acquired them over time. They seemed to first acquire nominal and verb negation before adjective negation. The author concluded that L2 learners first tended to exclusively use '-nai' in the unanalyzed manner and then began to use several kinds of negators, still in the unanalyzed manner, before they finally began to use them in the analyzed manner.

Connective expressions

峯布由紀(2007)「認知的な側面からみた第二言語の発達過程について―学習者の使用する接続辞表現の分析結果をもとに―」『日本語教育』134, 90 − 99 [tr. Mine, F. (2007). Cognitive factors on second language development: How learners develop their connective expressions in Japanese. Journal of Japanese Language Teaching, 134, 90-99].

Drawing on Pienemann's (1998)* Processability theory, which explains how language processing skills develop and become automatized, Mine examined connective expressions in Japanese L2 using OPI data from 90 Japanese L2 learners with Chinese, English, and Korean L1 (5 beginning, 10 intermediate, 10 advanced, and 5 superior level learners). She found that learner connective expressions expanded as follows: cause (e.g. –te, -kara ) > adversative (e.g. –kedo, -ga) > hypothetical cause (e.g. –tara, -to, -ba) > hypothetical adversative (concessive) (e.g. – temo) (p. 138). The author broadly argues that it is challenging for novice learners to focus on forms and meaning simultaneously in their language processing.

* Pienemann, M. (1998). Language processing and second language development: Processability theory. Amsterdam: John Benjamins.

Demonstrative

迫田久美子(1993)「話し言葉におけるコ・ソ・アの中間言語研究」『日本語教育』81, 67-80 [tr. Sakoda, K. (1993). A study on the interlanguage of demonstratives in spoken Japanese. Journal of Japanese Language Teaching, 81, 67-80].

Sakoda examined how Japanese demonstratives 'ko-,' 'so-,' and 'a-' are acquired in SLA. She collected interview data from 60 Japanese L2 learners from 22 countries, which consisting of 20 beginning, 20 intermediate, and 20 advanced learners, as well as 10 Japanese native speakers. In the one-hour-long interviews, participants were asked to talk about teachers they could remember from elementary school or junior high school. The results showed that the learners' frequency and accuracy of use of demonstratives increased as they became more proficient. The beginning level learners rarely used demonstratives and even when they did, they used them incorrectly. The intermediate level learners used more demonstratives but errors were still found frequently. The advanced learners were more native-like; that is, they used less 'ko-' and more 'so-.' However, the incorrect use of 'a-' was still found among them. Sakoda also found that learners whose L1s have three demonstratives acquired the Japanese three demonstrative system with more ease than the learners whose L1s have only two demonstratives.

迫田久美子 (1996)「指示詞コ・ソ・アに関する中間言語の形成過程−対話調査による縦断的研究に基づいて−」『日本語教育』89, 64-75 [tr. Sakoda, K. (1996). Interlanguage development of Japanese demonstratives: A longitudinal study based on oral interviews. Journal of Japanese Language Teaching, 89, 64-75].

Following her previous research (Sakoda, 1993), Sakoda further conducted a three-year longitudinal study on the acquisition of Japanese demonstratives. She conducted one-hour-long interviews with three Korean speakers and three Chinese speakers learning Japanese L2, collecting data every four months for the period of three years (eight interviews per participant). She identified the following developmental sequence among all the participants: anaphoric 'ko-' >=anaphoric 'so-' > anaphoric 'a-' > memory and strategic use (p. 188; p. 67). The main types of errors include the use of 'ko-' instead of 'so-' and the use of 'a-' instead of 'so-.' The former type appeared at the beginning level among all the participants and the latter type was found even at the advanced level when errors decreased. Sakoda further identified when erroneous forms disappeared among speakers of two languages: for Chinese speakers, errors disappeared in the following order: Errors with anaphoric 'so-,' errors of memory and strategic > errors with anaphoric 'a-' and anaphoric 'ko-'; for Korean speakers: Errors with anaphoric 'ko-' > errors with anaphoric 'so-' > errors of memory and strategic use > errors with anaphoric 'a-.'

Particles

久保田美子 (1994) 「第2言語としての日本語の縦断的習得研究−格助詞「を」「に」「で」「へ」の習得過程について−」『日本語教育』82, 72-85 [tr. Kubota, Y. (1994). Longitudinal acquisition research on Japanese as a second language: Concerning the acquisition process of particles o, ni, de, and e. Journal of Japanese Language Teaching, 82, 72-85].

Kubota conducted a 22-month study of two Japanese L2 adult learners living in Japan. She collected written data from grammar tests and journals and oral data from conversation tasks and story telling tasks. She examined four particles, 'o (accusative),' 'ni (locative),' 'de (locative),' 'e (directional)' and found that the L2 learners produced 'ni' and 'o' correctly more often than 'de' and 'e,' both in written and oral productions, until the end of the third quarter of the data collection period. In the fourth quarter, their production began to vary from one learner to the other. For instance, Kubota reported that both produced 'e' correctly most often but one participant was less accurate for 'ni' and the other – for 'o.' Particularly, both confused the use of 'de' and 'ni,' which seemed to be the result of the overgeneralization of 'ni' over 'de.' She also found that while the accuracy for 'o' was not influenced by the verb, the accuracy for 'de' and 'e' differed depending on the verb they were used with.

Transitive-intransitive verbs

These studies all focus on the acquisition of intransitive verbs drawing on Universal Grammar. They investigated how the L2 production of two subcategories of the Japanese intransitive verb, unaccusative and unergative verbs, takes place at the Deep and Surface structure levels. Although both subcategories of verb may present the same Surface structure (Subject-Verb structure), at the Deep structure level the subject in the unaccusative and unergative structures are object and subject respectively. According to Hirakawa (2001), typical unaccusative verbs include: otiru "fall," sinu "die," tuku "arrive," wareru "break," yakeru "burn," and typical unergative verbs include: naku "cry," utau "sing," asobu "play," oyogu "swim," hasiru "run" (p. 229).

Hirakawa, M. (2001). L2 acquisition of Japanese unaccusative verbs. Studies in Second Language Acquisition, 23, 221-245.

Hirakawa examined how 25 intermediate – advanced Japanese L2 learners with English L1 syntactically and semantically made the distinction between unaccusative and unergative verbs in Japanese. This distinction appears explicitly in certain constructions of which she focused on two - 'takusan (a lot)' construction and –te –i(ru) construction in grammar comprehension and multiple choice tests. Hirakawa concluded that the L2 learners seemed to syntactically and semantically distinguish between these subcategories.

Oshita, H. (2001). The unaccusative trap in second language acquisition. Studies in Second Language Acquisition, 23, 279-304.

Theoretically synthesizing previous studies on the acquisition of unaccusative and unergative verbs in Chinese, English, and Japanese, Oshita (2001) proposed the Unaccusative Trap Hypothesis that predicts L2 learners' developmental sequence in acquiring unaccusative/unergative verbs. Although his focus is placed more on English L2 acquisition than other L2 acquisition, he claimed that generally, low- and intermediate- level learners do not distinguish between unacusatives and unergatives. Errors tended to be found with unaccusatives first and L2 learners eventually acquired target-like syntactic structures specific to unaccusatives. In the case of Japanese L2, for instance, this appears in places where the nominative particle 'ga' can be dropped with unaccusative verbs, but not with unergative verbs.

Sorace, A., & Shomura, Y. (2001). Lexical constraints on the acquisition of split intransitivity: Evidence from L2 Japanese. Studies in Second Language Acquisition, 23, 247-278.

Sorace and Shomura examined how adult Japanese L2 learners with English L1 distinguish between unaccusative and unergative verbs that are accompanied by numeral quantifiers. While numeral quantifiers need to be adjacent in unergative structures in Japanese, that is not necessary in unaccusative structures. In their study, two groups consisting of 29 post-beginner and 31 intermediate learners living in Japan, participated in a grammatical acceptability task. They found that the L2 learners were more accurate with unergatives than unaccusatives. In their literature review, the authors nicely summarized which grammatical features of Japanese related to unaccusative/unergative verbs have been studied.

中石ゆうこ (2005) 「対のある自動詞／他動詞の第二言語習得研究―"つく―つける"、"きまる―きめる"、"かわる―かえる"の使用状況をもとに―」　『日本語教育』124, 23−32　[tr. Nakaishi, Y. (2005). A second language study on transitive-intransitive verb pairs: An analysis of the uses of tsuku/tsukeru, kimaru/kimeru and kawaru/kaeru. Journal of Japanese Language Teaching, 124, 23-32.]

Nakaishi first examined OPI data of 90 Japanese L2 learners with Chinese, English, and Korean L1 (5 beginning, 10 intermediate, 10 advanced, and 5 superior level learners) and found that erroneous use of transitive/intransitive verb pairs was more frequent at the intermediate level and above, and more frequent among English and Chinese speakers than among Korean speakers. She further administered a sentence completion task to 214 Japanese L2 learners (99 Chinese speakers and 126 Korean speakers). In this task she focused on 3 pairs of transitive and intransitive verbs, tsuku/tsukeru [to turn on], kimaru/kimeru [to decide], and kawaru/kaeru [to change]. While L2 learners used the transitive verbs more correctly overall, accuracy varied depending on the verb. Nakaishi further identified four different patterns of learner use of the verbs: 1) using only transitive verbs regardless of the conjugation forms, 2) using intransitive verbs regardless of the conjugation forms, 3) using either transitive or intransitive verb depending on the conjugation form, 4) using both transitive and intransitive verbs on the same conjugation form. Her data show that while all the selected verbs had the fourth type most frequently, the first type was observed more frequently than the second type in tsuku/tsukeru[to turn on] and kimaru/kimeru[to decide] and the second type was observed more frequently than the first type in kawaru/kaeru.

Verb conjugation

坂本正(1993)「英語話者における『て形』形成規則の習得について」『日本語教育』 80, 125-135 [tr. Sakamoto, T. (1993). Acquisition of Japanese Te-form formation rules by speakers of English. Journal of Japanese Language Teaching. 80, 125-135].

Administering a grammar test on 30 verbs to Japanese L2 learners with English L1 at a university in Japan, Sakamoto examined how the te-form is acquired across different proficiency levels from pre-intermediate to advanced. The results show that most of the conjugation types are acquired at the advanced level, but some conjugation types may be more difficult to acquire. Such examples include: - ku --> -tte (e.g. だく-->*だって), –gu --> -ide (e.g. およぐ-->*およんで、*およぎて ), -ru -->-tte (e.g. なぐる-->*なぐて) (pp. 132-133).The researcher also found that L2 learners across proficiency levels conjugated verbs correctly even when the verb was unfamiliar to them. The accuracy rates were 75-80% among pre-intermediate/intermediate level learners, 85% among pre-advanced level learners, and 93-97% among advanced level learners.

Kawaguchi, S. (2000). Acquisition of Japanese verbal morphology: Applying processability theory to Japanese. Studia Linguistica, 54, 238–248.

This article examined the developmental sequence of Japanese verbal morphology drawing on Pienemann (1998)*'s Processability theory (1998). Data analyzed in this study come from three longitudinal studies and two cross-sectional studies in which a total of 21 university students learning L2 Japanese completed "a unrestricted conversation and various picture tasks" (p. 244). Kawaguchi's empirical analysis supports her hypothesis of four developmental stages of Japanese verbal morphology acquisition:

Stage 1: Use of invariant forms such as the finite form of the non-past polite form (e.g., tabemasu, 'eat')

Stage 2: Use of lexical-semantic morphemes such as direct mapping of the semantics onto the verbal (e.g., tabemasu-ka, '(do you) eat?')

Stage 3: Use of phrasal morphemes that involves information exchange between the verb and the auxiliary within a verb phrase (e.g., tabe-nai 'do not eat')

Stage 4: Use of Inter-phrasal morphemes such as the conjunctive adverbial formation and the passive construction in Japanese (e.g., tabe-nagara 'while eating,' tabe-rare-ta 'was/were eaten').

* Pienemann, M. (1998). Language processing and second language development: Processability theory. Amsterdam: John Benjamins.

Conditional

スニーラット・ニャンジャローンスック（2001）「OPIデータにおける『条件表現』の習得研究 ―中国語，韓国語，英語母語話者の自然発話から―」『日本語教育』111, 26-35 [tr. Neancharoensuk, S. (2001). An acquisition study of conditional sentences based on OPI data: Natural speech of Chinese, Korean, and English learners of Japanese, Journal of Japanese Language Teaching, 111, 26-35].

Neancharoensuk examined a developmental sequence of Japanese conditional sentences, using OPI data of 90 Japanese L2 learners with Chinese, English, and Korean L1 (5 beginning, 10 intermediate, 10 advanced, and 5 superior level learners). Her results show that L2 learners began to acquire conditional sentences only at the intermediate level. She also found that the hypothesized category and the past habitual were easier to acquire than the counterfactual and the non-past habitual.

Numeral Classifiers

Hansen, L. & Y. Chen (2001). What counts in the acquisition and attrition of numeral classifiers? JALT Journal, 23, 90–110.

This article examined the developmental stages of numeral classifier syntax in Japanese L2 and Chinese L2. The data were collected from 204 English speaking learners of L2 Japanese and 192 English speaking learners of L2 Chinese who had lived in Japan or Taiwan for 18, 24, or 36 months. The participants were presented with a set of 24 drawings of objects and asked to tell the number of them. Hansen and Chen found that the learners generally go through the following three stages in learning number classifiers:

Stage 1: No classifier in the obligatory context

Stage 2: An un-marked classifier is inserted between numeral and noun, with gradual acquisition of appropriate semantic categories

Stage 3: Correct classifier suppliance

More interestingly, the authors identified the Numeral Accessibility Hierarchy as follows: Animate human (e.g., nin) >Animate non human (e.g., hiki) > Shape (e.g., hon, mai, ko) > Function.

Learning in Interaction

Iwashita, N. (2001). The effect of learner proficiency on interactional moves and modified output in nonnative–nonnative interaction in Japanese as a foreign language. System, 29, 267–287.

This article examined the impact of learner proficiency on opportunities for modified output in NNS-NNS interaction, on types of interactional moves (confirmation check or clarification request), and on production of interactional moves (p. 270). The data were collected from 24 college students in Australia who were asked to complete communicative tasks. The results show that more interactions occurred among high – low proficiency dyads than among the same proficiency level dyads but did not produce the most modified output. Overall, the author found no significant impact of learner proficiency on opportunities for modified output through interactional moves such as confirmation checks or clarification requests.

Ohta, A. S. (2001). Second language acquisition processes in the classroom: Learning Japanese. Lawrence Erlbaum.

This volume examined SLA processes (e.g., learners' private speech, peer interaction) of seven college students in a first- year or second-year Japanese class at a US university. Of particular interest is Ohta's analysis of the learners' private speech as the seven participants were responding to corrective feedback to others in their classrooms. She individually audio-recorded the utterances of each learner with sensitive microphones and collected data of their private speech utterances – speech uttered to themselves. Her conversation analysis of the private speech data demonstrated that the learners were actively using private speech even when not speaking up, making use of corrective feedback addressed to other learners by privately responding to it. In this way she highlights the importance of "active listeners" (pp. 71-72) in language classrooms.

(See also: Ohta, A. S. (2000). Rethinking recasts: A learner-centered examination of corrective feedback in the Japanese language classroom. In J. K. Hall & L. S. Verplaetse (eds.), Second and foreign language learning through classroom interaction. Mahwah, NJ: Lawrence Erlbaum, 47–71.)

Iwashita, N. (2003). Negative feedback and positive evidence in task-based interaction: Differential effects on L2 development. Studies in Second Language Acquisition, 25, 1–36.

This study examined the effects of negative feedback and positive evidence on the short-term development of two grammatical structures, the Japanese locative-initial construction and a verb morpheme –te (p. 11), in NS-NNS task-based conversation. The data were collected from 55 college students learning Japanese in Australia who were asked to complete two communicative tasks. They were also given a pre-test, an immediate post-test after completing the tasks, and a delayed post-test. The results show that positive evidence had a larger impact on the immediate post-test for learners with an above average score on the pre-test whereas implicit negative feedback such as recasts was beneficial to learners regardless of their scores on the pre-test. However, overall, approximately half of all non-target like utterances were ignored by NSs.

Foster, P., & Ohta, A. S. (2005). Negotiation for meaning and peer assistance in second language classrooms. Applied Linguistics, 26(3), 402.

This article examines instances of negotiation for meaning in two different language learning settings. The data were collected from 20 young adult learners learning English at a college in London and 19 American college students learning Japanese while carrying out two similar information exchange tasks in class. The authors took quantitative and qualitative approaches to data analysis. Their findings show that negotiation for meaning did not occur frequently. However, in completing the tasks, the learners were actively assisting each other through a variety of ways including co-construction of utterances, self-corrections, and peer-corrections.

Egi, T. (2007). Interpreting recasts as linguistic evidence: The roles of linguistic target, length, and degree of change. Studies in Second Language Acquisition, 29, 511-537.

This study examined how 49 learners of Japanese interpreted recasts provided to them while they were completing two information-gap tasks. 31 learners were asked to report their thoughts immediately after each turn when a recast was given and 18 learners were asked to report their thoughts after the tasks were completed. The results show that the learners tended to interpret recasts as responses to content when they were lengthy and deviated from the original utterances. They were more likely to notice recasts when they were short and close to the original utterances. Egi argues that length and number of changes in recasts may also impact learners' interpretations.

(See also: Egi, T. (2007). Recasts, learners' interpretations, and L2 development. In A. Mackey (Ed.), Conversational interaction in second language acquisition (pp. 249– 267). Oxford: Oxford University Press.)

Yoshida, R. (2008). Teachers' choice and learners' preference of corrective feedback types. Language Awareness, 17(1), 78-93.

This article examines teachers' choice and learners' preference for corrective feedback in Japanese classrooms at an Australian university. The data consist of audio-recordings of second-year Japanese classrooms focusing on seven college students studying Japanese and two teachers of Japanese along with stimulated recall interviews. The results show that although self-correction was perceived to be more effective by the teachers and learners, the teachers used recasts most frequently.

Yoshida, R. (2010). How do teachers and learners perceive corrective feedback in the Japanese language classroom? The Modern Language Journal, 94(2), 293-314.

This article continues to examine the perceptions of corrective feedback by teachers and learners. Re-analyzing the data from Yoshida (2008), Yoshida found that the learners did not always notice corrective feedback even when they responded to it, while the teachers tended to perceive that their students, particularly 'strong ones,' noticed and understood their corrective feedback.

Egi, T. (2010). Uptake, modified output, and learner perceptions of recasts: Learner responses as language awareness. The Modern Language Journal, 94(1), 1-21.

This article examined whether and how 24 learners of Japanese perceived recasts provided while they were completing three tasks. After completing the tasks, the learners were asked to view their recast episodes and report their thoughts on them. The results show that those who provided uptake were more likely to perceive recasts as corrective feedback. Moreover, those who modified their original utterances in response to recasts noticed not only recasts but also the gaps in their interlanguage.

Syntactic Complexity

Iwashita, N. (2006). Syntactic complexity measures and their relation to oral proficiency in Japanese as a foreign language. Language Assessment Quarterly, 3, 151–169.

This article examined the relationship between syntactic complexity measures and oral proficiency in Japanese. Data were collected from 33 English-speaking college students learning L2 Japanese, who were asked to complete three different oral narrative story-telling tasks. In her data analysis, Iwashita employed four productive units to examine four syntactic complexity measures applied to learner language produced by two (high and low) proficiency groups. The four productive units she employed include T-unit (i.e., an independent clause and all its dependent clauses), clause, verb phrase, and word. The four syntactic complexity measures examined in her analysis are length (the number of words per T-unit/clause), general complexity (e.g., the number of clauses per T-unit), coordination (the number of independent clauses per T-unit and the proportion of independent clauses to the total number of clauses), and subordination (the numbers of dependent clauses and verb phrases per T-unit and the proportion of dependent clauses to the total number of clauses). The results reveal that the length of T-units and the number of clauses per T-unit were the best predictors of the learners' oral proficiency.