Praktiken der Personenreferenz in chinesischen und deutschen Chat-Interaktionen: Die kommunikative Konstruktion von Kulturalität

Zusammenfassung Dieser sprach- und kulturkontrastiv ausgerichtete Beitrag untersucht kommunikative Praktiken der Personenreferenz in chinesischen und deutschen Chat-Dialogen (SMS-, WeChat- und WhatsApp-Nachrichten). Im Gegensatz zur gängigen Präferenz des Gebrauchs minimaler Referenzformen (u. a. mittels deiktischer Pronomen wie „ich“ und „du“ bzw. „wo“ [我] und „ni“ [你] sowie zero pronouns im Chinesischen) verwenden chinesische wie auch deutsche TeilnehmerInnen in diesen translokalen, computer-vermittelten Interaktionen oftmals Verwandtschaftsnamen zur Referenz auf sich selbst sowie auf ihre RezipientInnen. Auf der Basis einer interaktional ausgerichteten Studie sollen am Beispiel von Geschwisterrollennamen (wie „Schwester“ und „Bruder“ bzw. „Jiejie“ [姐姐 ältere Schwester], „Meimei“ [妹妹 jüngere Schwester] und „Gege“ [哥哥 älterer Bruder], „Didi“[弟弟 jüngerer Bruder]) sowohl Parallelen als auch Unterschiede in der Verwendung dieser kinship terms zur Selbst- und Fremdreferenz in den chinesischen und deutschen Daten aufgezeigt werden. Hierbei wird erkenntlich, dass Praktiken der Personenreferenz eng mit sprach- und kulturspezifischen Traditionen sozialer Beziehungsformationen verwoben sind.

Referential Choice in the Written Narratives of Indonesian Adults

This study aims to describe the referential choice of two protagonist animate characters in a silent six-minute film entitled The Pear Story (Chafe, 1980). A total of 80 undergraduate and graduate Indonesian students were asked to watch the film and then retell the story by writing a narrative about the film in Indonesian. Findings indicate that when the protagonist animate referents are mentioned for the first time, a classifier seorang ‘a person’ is always used before the NP. When they are reactivated, they are mostly expressed by zero, pronouns dia or ia ‘he’, clitic –nya ‘his’ or ‘him’, NPs with determiners ini ‘this’, itu ‘that’, tersebut ‘aforementioned’, tadi ‘mentioned before’, NPs with relative clauses and NPs with definite articles si or sang ‘the’ which are often used in fables or tales. This study also demonstrates that the choice of anaphoric expressions of the protagonists is determined by factors such as referential distance and referential interference. This study confirms other cross-linguistic studies about referential choice – that there is a correlation between salience and referring expressions in discourse. AbstrakPenelitian ini bertujuan mendeskripsikan penggunaan acuan untuk dua tokoh utama dalam film bisu berdurasi enam menit yang berjudul The Pear Story (Chafe, 1980). Sebanyak 80 mahasiswa Indonesia tingkat Strata 1 dan Pascasarjana di sebuah universitas swasta diminta untuk menonton film tersebut dan kemudian menuliskan narasi mengenai film itu dalam bahasa Indonesia. Data menunjukkan bahwa jika referen kedua tokoh utama tersebut disebutkan untuk pertama kalinya, bentuk leksikal seorang selalu muncul sebelum frasa nomina (FN). Jika referennya diaktifkan kembali, maka yang digunakan adalah zero, pronomina dia atau ia, klitik –nya, FN + ini/itu/tersebut/tadi, FN + klausa relatif, dan si/sang + FN. Penelitian ini juga menunjukkan bahwa pilihan bentuk anaforik untuk kedua tokoh utama itu ditentukan oleh faktor jarak antara referen dan anteseden dan juga oleh adanya interferensi referen. Hasil penelitian ini sejalan dengan penelitian lintas bahasa mengenai pilihan acuan, yaitu bahwa ada korelasi antara referen utama dan bentuk acuannya.

Hierarchical Attention Network with Pairwise Loss for Chinese Zero Pronoun Resolution

Recent neural network methods for Chinese zero pronoun resolution didn't take bidirectional attention between zero pronouns and candidate antecedents into consideration, and simply treated the task as a classification task, ignoring the relationship between different candidates of a zero pronoun. To solve these problems, we propose a Hierarchical Attention Network with Pairwise Loss (HAN-PL), for Chinese zero pronoun resolution. In the proposed HAN-PL, we design a two-layer attention model to generate more powerful representations for zero pronouns and candidate antecedents. Furthermore, we propose a novel pairwise loss by introducing the correct-antecedent similarity constraint and the pairwise-margin loss, making the learned model more discriminative. Extensive experiments have been conducted on OntoNotes 5.0 dataset, and our model achieves state-of-the-art performance in the task of Chinese zero pronoun resolution.

The universal topic prominence stage hypothesis and L1 transfer

This paper tests the universal Topic Prominence (TP) stage hypothesis and L1 transfer hypothesis by analyzing attested TP features, namely zero pronouns, bare (i.e., article-less) nouns, and post-nominal topic marking found in Korean written narratives collected from L2 Korean (TP language) speakers whose L1 is English (SP language) and Mandarin Chinese (TP language), respectively. The results showed that L2 Korean speakers used zero pronouns as frequently as L1 Korean speakers did from the early stage of their acquisition, whereas the use of the topic marking lagged behind, regardless of their L1 backgrounds. We explain the results by appealing to the universal acquisition path based on pragmatic-lexical-morphosyntactic cline: while the concept of the topic is (arguably) universal, its morphological marking is influenced by L1 as well as learning experience.

A Deep Neural Network for Chinese Zero Pronoun Resolution

Existing approaches for Chinese zero pronoun resolution overlook semantic information. This is because zero pronouns have no descriptive information, which results in difficulty in explicitly capturing their semantic similarities with antecedents. Moreover, when dealing with candidate antecedents, traditional systems simply take advantage of the local information of a single candidate antecedent while failing to consider the underlying information provided by the other candidates from a global perspective. To address these weaknesses, we propose a novel zero pronoun-specific neural network, which is capable of representing zero pronouns by utilizing the contextual information at the semantic level. In addition, when dealing with candidate antecedents, a two-level candidate encoder is employed to explicitly capture both the local and global information of candidate antecedents. We conduct experiments on the Chinese portion of the OntoNotes 5.0 corpus. Experimental results show that our approach substantially outperforms the state-of-the-art method in various experimental settings.