Earticle

This paper examines whether speech errors in Korean are capable of reflecting the age and gender differences of their producers. Following the classification system used for English SOTs, this study collected a total of 207 Korean SOTs and classified them in three different ways. Errors were first grouped according to the speakers' gender and age, and then they were looked at with relation to the stage at which the errors occurred, the form of the errors, as well as the directionality of the errors. According to the data, lexical errors were found to be far more frequent than the other error types. Several differences, though none extremely large, were observed in the SOT errors among participants of different ages as well as between male and female participants. Although the amount of data collected is not large enough to draw any definitive conclusions, these preliminary findings suggest that a difference in language usage by males and females does exist.

.

Taeho Kim. & Han-gyu Lee, 2020. Journal of Cross-cultural Studies xx-xx. A Linguistic Analysis of Children's Speech Errors in Korean Speech Production. The primary purpose of this study was to investigate how Korean children produce speech errors (or slips of the tongue) during speech production and show that their errors are not arbitrary but systematic, which reflects their language knowledge and abilities. To achieve this goal, we collected 122 speech errors from the spontaneous, casual conversations of two groups of children, pre-schoolers and primary schoolers, and classified them according to three dimensions: linguistic components, directionality, and forms of errors. The rates of errors were compared carefully, not only between the two child groups, but also between children and adults. Such comparisons provide a linguistic clue for language production and language development processes. This study will be linguistically significant in that there have been no studies of children’s speech errors in Korean, although there have been many in other languages.

It is highly likely that Korean college students often commit inadvertant speech errors of replacing /t/ or /d/ with palato- alveolar /?/ and /?/ in the consonant clusters of /tr/ and /dr/. The writers presume that this is due to their confusion about the place of articulation of /t/ and /d/ with /?/ and /?/ /in /tr/ or /dr/ sequence, their lip-rounding difference from that of /?/ and /?/, and finally students’ inaccurate knowledge about the phonetic constraint of */?r/ and */?r/ in English syllable structure. Under such hypotheses this study analyzes the responses from an experiment group of fifteen Korean college students based on phonetic theories and probes into possible background reasons for this over-production as a form of speech errors.

Speech error data have been known to be informative of phonological units that are functional during speech processing. Using short-term memory tasks, we gathered speech error data from Japanese speakers to investigate how they combine segments into higher-level suprasegmental frames. The findings suggest that the phonological affiliations of segments are an important factor to Japanese speakers' speech errors. Specifically, onsets and vowels are more likely to be kept intact as a unit than vowels and codas. We suggest that this finding supports the idea that Japanese speakers' more permanent knowledge of the moraic structure contributes to the their short memory recall of segment sequences.

All types of part-of-speech (POS) tagging errors have been equally treated by existing taggers. However, the errors are not equally important, since some errors affect the performance of subsequent natural language processing seriously while others do not. This paper aims to minimize these serious errors while retaining the overall performance of POS tagging. Two gradient loss functions are proposed to reflect the different types of errors. They are designed to assign a larger cost for serious errors and a smaller cost for minor errors. Through a series of experiments, it is shown that the classifier trained with the proposed loss functions not only reduces serious errors but also achieves slightly higher accuracy than ordinary classifiers.

Анализ грамматических ошибок иностранных студентов в целом подтверждает высказанные ранее идеи о том, что при усвоении первого языка дети овладевают целыми синтаксическими блоками, привязанными к определенным контекстам. Из-за этого они не допускают ошибок на поверхностном уровне, однако нет однозначных свидетельств того, что они действительно знакомы с правилами употребления форм, так как для такого утверждения нет достаточного количества контекстов, в которых могли бы возникнуть ошибки. Если же речь идет об усвоении грамматического правила, здесь между изучением первого и второго языков наблюдаются параллели: и дети, и инофоны (а в некоторых случаях и русскоговорящие взрослые в условиях неформального общения) склонны к так называемой сверх-генерализации правил, в результате которой они производят инновации в соответствии с языковыми законами, но вразрез с практикой употребления. Необходимо отметить богатство и разнообразие тех проблемно-разрешительных механизмов, которые носители другого языка используют для того, чтобы наиболее адекватно передать свои идеи при недостатке средств выражения.

Since the Korean pronoun system does not specify gender indication, Korean speakers of English tend to have difficulties when they use English pronouns. This paper explores how absence of obligatory gender marking in Korean affects gender errors in the production of English. Gender pronoun errors made by Korean learners of English were analyzed using recordings of English conversations of seven Korean subjects. Findings of this study revealed that Korean English learners struggled with the use of gender pronouns due to differences in gender pronoun systems between Korean and English. Both gender-neutrality and pro-drop practices do not require Korean speakers to pay close attention to gender factors when using pronouns in their native language, resulting in native language interference. Considering that the Korean language lacks specific gender orientation in its usage of pronouns whereas English requires clear gender specification, findings of this study imply that native language gender concept can influence second language use.

This study investigates resyllabification errors made by Vietnamese learnersof Korean, utilizing an L2 speech corpus known as "A Collection of Foreigners’ KoreanSpeech Data for AI Training." By analyzing the read-aloud and spontaneous speech datafrom 40 Vietnamese L1 speakers in the corpus, we discovered that the error rate was higherin script reading than in spontaneous speech. We propose the rigid syllable boundary inVietnamese L1 phonology, contrasting with the fluid syllable boundary in Korean L2 phonology,results in a negative transfer, causing Vietnamese learners’ difficulty in resyllabification whenspeaking Korean. Furthermore, the orthographic rigidity of syllable boundary of the Koreanwriting system presents negative orthographic input for Vietnamese learners, whose orthography(Roman alphabet) allows for syllabic ambiguity. Consequently, speakers tend to produce moreerrors in script reading with such orthographic input than in spontaneous speech. We concludethe disparity between the phonological fluidity and the orthographic rigidity of Korean syllablescontributes to failure of resyllabification among Vietnamese speakers.

本研究では，日本語音声の中で「撥音」「促音」「撥音と促音」「拗音と直音の拍数同じもの」「拗音と直音の拍数の違うもの」について聴覚的，音声（調音）学的な特徴にその焦点をおき，韓国語を第一言語とする学習者が日本語を話す（読む）ときに起こる日本語の言いあやまりについて調査を行い，その分析を行った。本研究では先行研究で明らかにされていない，次の二点に力点をおき，研究を行った。1. なぜ，言いあやまりが起こるか，その原因を探るために言いあやまりの中でとくに注目されているインターランゲージ(Interlanguage:中間言語)に関する仮説を設定し，その仮説を言いあやまりの調査から検証した。2. 従来の主観だけによろ方法から脱却して，調査資料の検討，あるいは分析結果に基づく研究をした。これらは先行研究の何れにも十分に組み込まれていない研究である。本論文では，このような問題意識に基づいて言いあやまり研究を実践的な（教育的な面）の側面から，より体系的にとらえていくことを目指した。今後の韓国における日本語音声教育の方法改善の上では，本研究の調査分析結果に示されている面がとくに重視されなければならないと思われる。本研究では，研究対象の外におかざるをえなかった，撥音・促音・拗音と直音の知覚心理学特性への反映の問題，撥音・促音・拗音と直音と発話速度との関係などは，今後の課題にしておきたい。

Koo, Hyun-jung. 2002. On Typology of Speech Errors: With Reference to Mass Media Texts. The Sociolinguistic Journal of Korea, 10(2). Speech error per se is of special significance with respect to research of language production and its application to education. However, research on this subject matter has largely been underrepresented. This research is an attempt to classify speech errors based on a collection of instances in mass media that are judged as such by native Korean speakers. Extant research has largely focused on the type of speech error, a production of linguistic form different from one intended by the speaker, who is unconscious of such misproduction at the time of such utterance. This type of speech error is labeled as cognitive speech error in this research. However, of much more importance in actual communication is such speech error as one in which the speaker's utterance is interpreted, according to the speaker-addressee relationship or the particular speech situation, as something other than what the speaker intended, which is labeled as pragmatic speech error. This research focuses on the latter cataloging its subtypes and exploring its significance in interpersonal communication.

.

The way of classifying language performance errors can affect perspectives on identifying and interpreting phonological speech errors. The way also provides a framework of analyzing attested errors. Along with this research background, the purpose of this study was to propose a new way of classifying speech errors. For data collection, this study obtained speech output from various sources such as free talks among people, formal speeches/lectures, and TV programs. When collecting speech output containing speech errors, I examined words if the contextual errors are relevant only to the words themselves, but examined the whole sentence if the contextual errors were beyond the matter of words. In analyzing the output phonologically, I adopted three ways to classify types of speech errors suggested in previous studies. Based on the error types observed in the collected data, maintaining the strong points of the previous classifications, I suggest the way of classifying speech errors into six categories: replacement, inversion, elision, insertion, reordering and confusion. This way is certainly unique since it adds to the previously suggested categories two new categories, reordering and confusion.

본 연구는 다문화가정 유아의 발화 오류에 대해 교사가 제공한 피드백에 대한 유아의 오류 수정과 피드백 흡수 양상을 종단적으로 탐색하였다. 이를 위해 서울, 경기 지역 유치원 3세 학급에 재원 중인 다문화가정 유아 4인, 교사 2인을 연구 대상으로 선정하고, 2022년 10월부터 2025년 1월까지 3년 동안 총 5회에 걸쳐 교사-유아 간 상호작용을 관찰․녹화하여 자료를 수집하였다. 수집된 자료는 전사한 후 유아 발화 오류에 대한 교사의 피드백과 이에 대한 유아의 오류 수정 여부와 피드백 흡수 양상을 빈도 분석 하였다. 연구 결과 첫째, 유아의 발화 오류는 문법오류가 가장 높은 비중을 나타냈다. 둘째, 교사 피드백 중 명시적 수정과 유도 피드백은 수정 완료 비율이 높고 오류 고쳐 되말하기는 낮게 나타났다. 셋째, 유아의 연령에 따라 교사의 피드백 사용과 유아의 오류 수정 양상에 차이가 나타났다. 넷째, 유아의 한국어 구사력 수준에 따라 교사의 피드백 사용과 유아의 오류 수정 양상에 차이가 나타났다. 본 결과는 다문화가정 유아의 언어 발달 지원에 있어 유아의 연령과 한국어 구사력 수준을 고려한 교사의 피드백 제공이 중요함을 시사한다.

The purpose of this study was to longitudinally examine young children’s error correction and feedback uptake in response to teachers’ feedback on utterance errors among children from multicultural families. Participants were four children from multicultural families enrolled in preschool classes for three-year-olds in Seoul and Gyeonggi Province and their two classroom teachers. Data were collected through five rounds of observation and video recording of teacherchild interactions from October 2022 to January 2025. The data were transcribed and analyzed using frequency analysis, focusing on teachers’ feedback types, children’s subsequent error correction, and feedback uptake patterns. The results are as follows. First, grammatical errors accounted for the largest proportion of children’s utterance errors. Second, among the types of teacher feedback, explicit correction and elicitation led to higher rates of successful correction, whereas recasts were associated with lower correction rates. Third, differences were found in teachers’ use of feedback and in children’s error-correction patterns depending on children’s age. Fourth, differences were also observed in teachers’ use of feedback and in children’s errorcorrection patterns according to children’s Korean language proficiency. These findings highlight the importance of teachers’ feedback practices that consider children’s age and Korean language proficiency in supporting the language development of young children from multicultural families.

This paper is intended to explain the cause of exchange speech errors and after-thought utterances by cognitive linguistic viewpoint and to get wide understanding about language production. By automatic syntactics theory can't we explain the two phenomena. But if we admit that language use is cognitive activity by memory and it's activation as in cognitive linguistic viewpoint, we are able to explain the two phenomena with ease and rightly. Language production is largely originated in the activation of embodied memory and assembly of [meaning structure-phonological structure]s. The exchange speech errors happen by the more activated component element inside of familiar language form. And after-thought utterances happen by complementary activation of the inactivated memory after sentence finishing.

본 연구는 내적수정가설(covert repair hypothesis)에서 중요한 가설 중의 하나인 ‘비유창성(disfluencies)은 말오류(speech error)와 동일한 범주에 속하지 않는다’를 검증하는 것을 목적으로 하였다. 이를 위해 16명의 말더듬성인과 연령 및 성별을 일치시킨 정상성인 16명을 대상으로 발음하기 어려운 문장(tongue twister)을 자신의 말속도에 따라 암송하게 하여, 정확도 조건(low accuracy(LA)와 high accuracy(HA))에 따른 말오류, 자기수정 및 비유창성 빈도의 관계를 분석하였다. 연구결과, 정상집단 및 말더듬집단 모두 정확도가 강조됨에 따라 말오류 빈도는 감소하는 경향을 보였으나, 반대로 비유창성 및 자기수정 빈도는 증가하는 경향을 보였다. 특히, 말오류 빈도는 LA 조건에 비해 HA 조건에서 유의하게 감소하였다. 그러나 자기수정과 비유창성 빈도는 정확도 조건에 따라 유의한 차이가 없었다. 각 정확도 조건에서 자기수정 및 비유창성 빈도의 집단간 차이를 분석한 결과 LA 조건에서는 집단간에 유의한 차이가 없었으나, HA 조건에서 유의한 차이가 있었다.

The present study aimed to examine one of the assumptions of the covert repair hy- pothesis by examining whether speech errors and disfluencies belong to the same category. For this purpose, 16 adults who stutter and the 16 age and sex-matched normally fluent adults were asked to recite aloud the tongue twister sentences under 2 different conditions. In the low accuracy condition(LA), the subjects were instructed to recite the sentences fast. However, in high accuracy condition (HA), they were instructed to read correctly as well as fast. The results were as follows: (1) the speech error frequency increased significantly when accuracy was emphasized, but the frequency of disfluencies and self-repairs remained the same, (2) a significant group difference was found in the frequencies of disfluencies and self-repairs in the HA condition. Both groups showed that speech errors tended to decrease with accuracy emphasis, but disfluencies and self-repairs tended to increase.