1    Use of Translation Corpora as a New Method in Chinese Language Research and Its Pedagogical Implications: The Case of Viewpoint in Narratives* Wei-lun Lu Masaryk University Annotation This article is a preliminary attempt to apply the use of translation corpora as a new method in Chinese language research. We look into how the use of a Chinese literary masterpiece and its English translation can help us identify interesting differences between the two languages. The discussion is focused on how the author and the translator present viewpoint structure in the narrative. Use of deictic verbs and stylistic strategies such as zero-anaphora and (free) indirect speech are discussed. At the end of the article we propose a pedagogical implication. Keywords: cognitive linguistics, deixis, stylistics, translation, viewpoint 1. Parallel Texts, Linguistics Research and Teaching Language corpora have been a useful resource extensively used in research in general linguistics and language teaching.1 Various studies have shown the usefulness of language corpora in the above fields. However, the use of translation corpora is a practice that has not yet been systematically adopted in the field of Chinese Linguistics and teaching Chinese as a second language.2 The current paper intends to occupy the niche by presenting a case study and will discuss the pedagogical implications of the use of translation corpora in Chinese Linguistics and teaching. Parallel texts, as a type of corpora, are a collection of texts put alongside their translations. For the interest of our current readers, researchers of the Chinese language may put together a collection of world masterpieces written in Chinese (e.g. A True Story of Ah-Q, Dreams of Red Chamber, etc.) and find their translations in English, Japanese, Czech, German, etc. which will form a set of parallel texts available for the purpose of language research and teaching. Parallel texts are usually aligned sentence by sentence.                                                              *  Part of this study was presented at The Eighth Czech and Slovak Sinological Conference in Olomouc, 7–8  November 2014. I thank conference participants and two reviewers for valuable comments, with the typical  disclaimer that applies. The research was supported by the project “Employment of Best Young Scientists for  International Cooperation Empowerment” (CZ.1.07/2.3.00/30.0037) co‐financed from European Social Fund and  the state budget of the Czech Republic.  1  Typical examples include Chen (2010); Chung and Huang (2010); Lu and Su (2009); Lu (Accepted).   2  A couple of nice exceptions are Xiao and Wei (2014); Xu and Li (2014).  2    We argue that such methodology has a great advantage, as aligning the texts sentence by sentence allows one to see how the same situation is coded in different languages, and a comparison between the source language and the target language(s) may reveal the distinctiveness of the source language. In our current case, comparing Chinese with the other languages (i.e. English, Japanese, Czech, German, etc.) will reveal how exactly Chinese is different from the other languages. Furthermore, we propose that such comparison may have important pedagogical implications. It has been extensively reported that second language learners are constantly bothered by the inconsistencies between their first language and the second language that they are trying to acquire, and that such inconsistencies are an important cause of learner’s erroneous production patterns. Therefore, we believe not only that parallel texts may be used as an important research material for linguistics research, but also that distinctive patterns of Chinese identified by a parallel-text-based study may serve as an important basis for the design of teaching materials. Below, we present a case study of deictic verbs in Chinese-English parallel texts and will discuss the pedagogical implication of the research. 2. Material Choice, Corpus Design and Scope The material chosen for the present study is a famous novel written by Mo Yan, the Nobel Laureate in 2012, entitled 酒国 Jiu Guo (The Republic of Wine).3 The novel is chosen for the fame of its author and we believe the work will become increasingly important given Mo Yan’s role in the field of contemporary Chinese literature.4 Only the first chapter of both the Chinese original and the English translation is included in the scope of study as a sample. We focus on deictic verbs in both the Chinese original and the English translation, as deictic verbs are probably one of the most fundamental words in language. Their semantics is rather basic and frequency is relatively high, so they are suitable candidates for linguistics research. Learners of a second language may also find deictic verbs among the first words that they need to know. We will focus on how deictic verbs and their interactions with other figures of speech allow the speaker to express viewpoint in authentic written discourse. Deictic verbs are motion verbs that denote the direction of motion with reference to the speaker’s position (the so-called origo). Motion towards the speaker in Chinese is coded by lai, whereas away from the speaker by qu. In English motion towards the speaker is coded by come (or its variants comes, came, coming), whereas away from the speaker by go (or its variants goes, went, going, gone). When we did a corpus search, we included all variants of the lexemes.                                                              3  Mo (2012).  4  The English translation is done by Prof. Howard Goldblatt, a Sinologist based in the United States. He is famous for having translated several important contemporary Chinese classics, including The Taste of Apples by Huang Chunming and The Execution of Mayor Yin by Chen Ruoxi. Therefore, given the translator’s extensive experience, we believe that the quality of the translation is more than acceptable so is suitable for the purpose of linguistics research.  3    However, note that deictic verbs, given their basic status in language use and their high frequency, also show a tendency of being involved in idioms and appearing in grammaticized forms, where the four lexemes may lose their physical and directional nature. Typical examples in Chinese include guan-qilai ‘close-QILAI’, kan-qilai ‘see-QILAI’, where lai cannot be replaced with qu, and ting-xiaqu ‘listen-XIAQU’, du-xiaqu ‘read-XIAQU’, where qu cannot be replaced with lai. Typical English examples include the progressive aspectual expression be going to, where go cannot be replaced with come to express a different viewpoint and come down to earth ‘be realistic’, where come cannot be replaced with go. We exclude such idiomatic expressions from our scope, since our focus is viewpoint expression in discourse. 3. Representative Cases Identified in the Parallel Texts Below are some representative examples that we find in the first chapter of the Chinese and English versions of The Republic of Wine. We discuss the examples in two groups: Section 3.1 introduces examples where only the Chinese version uses a deictic verb to express the narrator’s viewpoint but such a stylistic effect is entirely absent in the English version. Section 3.2 includes examples where the Chinese version uses a deictic verb for the viewpointing effect (Dancygier 2012) while the English version does not. But not only that, the examples also involve another stylistic means to achieve the effect. 3.1 Viewpoint expressed only in the Chinese original In this category, we introduce examples where only the Chinese version expresses the viewpoint of the narrator with a deictic verb while the English version does not. Example (1) and (2), both taken from the narration of the novel, are typical.5 (1a) 看门人 把 手 拿下来, 放 在 kanmen-ren ba shou na-xia-lai fang zai gate keeper PRT hand take-down-come put LOC 面前 看。 mian-qian kan face-front see                                                              5  Examples marked with (a) are all passages from the Chinese original. The first line presents the passage in  Chinese characters, and the second line is the transliteration of the passage, and the third line is the word‐by‐word  translation of the original text. For the glosses of grammatical markers, we follow the convention in linguistic  typology, The Leipzig Glossing Rules developed by the Max Plank Institute of Evolutionary Anthropology. Examples  marked with (b) are passages from the English translation. Examples marked with (c) are constructed for the  purpose of explaining the viewpointing effect in the original (a) or (b) passages.  4    “The gatekeeper put his hands down (in the direction of the speaker) and looked at them in front of his face.” (1b) The gatekeeper lowered his hands and examined them. (1c) 看门人 把 手 拿下去, 放 在 kanmen-ren ba shou na-xia-qu fang zai gate keeper PRT hand take-down-go put LOC 面前 看。 mian-qian kan face-front see “The gatekeeper put his hands down (away from the speaker) and looked at them in front of his face.” (constructed) Example (1a) is a typical example of the narrator’s language that uses lai to construct a viewpoint that is identical to the protagonist’s (Ding Gou’er’s). This point becomes straightforward when we compare (1a) with its constructed counterpart (1c), which is created by replacing lai with qu. The viewpoint of the speaker in (1c) is identical probably to the gatekeeper’s. On the other hand, we cannot find a similar viewpoint expressed in its English translation, with no deictic verb found in (1b), which is a mere description from the perspective of a neutral and objective observer. The English version therefore shows much less involvement of the narrator in the scene than the Chinese original. (2a) 丁钩儿 接过 酒瓶子, 晃晃, dinggouer jie-guo jiuping-zi huang-huang Ding Gou’er take-over wine bottle shake-RED 蝎子 在 参须 间 游泳, 怪 味道 xiezi zai sen-xu jian youyong guai weidao scorpion LOC ginseng root LOC swim strange odor 从 瓶口 冲出来。 cong ping-kou chong-chu-lai LOC bottle mouth rush-out-come “Ding Gou’er took over the bottle, shook it, (and found) the scorpions swimming among the ginseng roots, with a strange odor coming (towards the speaker) out from the mouth of the bottle.” 5    (2b) He shook the bottle, and the scorpions swam in the ginseng-enhanced liquid. A strange odor emanated from the bottle. (2c) 丁钩儿 接过 酒瓶子, 晃晃, dinggouer jie-guo jiuping-zi huang-huang Ding Gou’er take-over wine bottle shake-RED 蝎子 在 参须 间 游泳, 怪 味道 xiezi zai sen-xu jian youyong guai weidao scorpion LOC ginseng root LOC swim strange odor 从 瓶口 冲出去。 cong ping-kou chong-chu-qu LOC bottle-mouth rush-out-go “Ding Gou’er took over the bottle, shook it, (and found) the scorpions swimming among the ginseng roots, with a strange odor going (away from the speaker) out from the mouth of the bottle.” (constructed) Example (2) is rather similar to (1) in terms of the viewpoint alignment in the Chinese original and the English translation. In (2a), the viewpoint of the narrator is with the protagonist, which is outside of the bottle. But if we replace lai with qu, as is shown in the constructed (2c), the viewpoint of the speaker is no longer outside of the bottle with the protagonist but instead inside of the bottle. On the other hand, the English translation does not show such viewpoint management by use of a deictic verb. Rather like (1b), (2b) is also an objective description of the source of the odor using a viewpoint-neutral motion verb emanated, which only means ‘to exit’. 3.2 Interaction of deictic verbs and other figures of speech The phenomena that we show below are more complicated than what we have seen in the previous section. Here we will discuss the interaction of a deictic verb and other figures of speech in the same passage. (3) is an interesting case where the deictic verb lai creates a special literary effect with zero anaphora. (3a) 他 感到 乏味、 无趣, 便 把 她 ta gan-dao fawei wuqu bian ba ta he feel-PFV bland uninteresting then PRT she 6    推开。 她 却 像 一只 凶猛的 tui-kai ta que xiang yi-zhi xiongmeng-de push-aside she nevertheless like one-CL fierce-LINK 小豹子 一样, 不断地 扑上来… xiao baozi yiyang buduandi pu-shang-lai leopard cub same relentlessly spring-up-come “He felt uninterested and then pushed her away. But she was like a fierce leopard cub and kept coming up (against the speaker) relentlessly.” (3b) But, like a plucky fighting cock, she sprang back at him hard, catching him off guard and making resistance all but impossible. In (3a), the viewpoint of the narrator is obviously with the protagonist, which is shown by the use of lai. Note also that zero-anaphora is at play, with the goal of the female character’s motion, which is the protagonist, left unspecified. This naturally gives the association of the narrator’s viewpoint with the protagonist’s. But on the other hand, the viewpoint of the narration in (3b) is more neutral. In the English text, there is no use of a deictic verb, and the goal of the female character’s motion is specified by a pronoun him, which objectively refers to the protagonist from outside of the scene. Here we see clearly that the narrator’s viewpoint is not with the protagonist (deeply involved in the scene) but instead adopts a distal perspective. Excerpts (4) are a similar set of instances that involves the interplay of deixis and zero- anaphora. (4a) 检察长 拉开   抽屉, 把  一封信  jiancha-zhang la-kai chouti ba yi-feng-xin Head of Inspectors pull-open drawer PRT one-CL-letter 拿出来, 先 瞄了两眼, 才 na-chu-lai xian miao-le-liangyan cai take-out-come first glance-PFV-a couple of times PRT 递给 丁钩儿。 di-gei dinggouer hand-to Ding Gou’er “The Head of Inspectors pulled the drawer open, took out (towards the speaker) a letter, first glanced it and handed it to Ding Gou’er.” (4b) He opened a drawer and took out a letter, glanced at it, then handed it over. 7    In (4a), the viewpoint of the narrator is with the protagonist. This is evidenced by the use of lai in the narrative. However, what is interesting is that Ding Gou’er is explicitly mentioned in a later part of the narrative so is an objective part of the scene. Note that the explicit mention of Ding Gou’er is important in understanding the viewpoint structure in this chunk of narrative. In the first part of the narrative, the narrator’s viewpoint coincides with the protagonist’s (i.e. outside of the drawer and away from the Head of Inspectors), whereas in the second half where Ding Gou’er is explicitly mentioned, the viewpoint is no longer with the protagonist, as the protagonist is explicitly mentioned as an object of conception. This stretch of discourse thus presents a mixture of narrative viewpoints, commonly known as free indirect discourse.6 On the other hand, the English translation (4b) does not exhibit such a mixture of viewpoints like its Chinese counterpart does. In (4b), only zero-anaphora is used as a stylistic strategy that induces the reader to identify the viewpoint of the narrator with the protagonist’s. Consider a constructed example (4c), where the goal of the motion of the letter, Ding Gou’er, is explicitly mentioned. (4c) He opened a drawer and took out a letter, glanced at it, then handed it over to Ding Gou’er. (constructed) This constructed stretch of discourse looks like an objective neutral description of the scene, with the protagonist mentioned as an object of conceptualization instead of as the origo of the narrative. The viewpoint of the narrator in (4c) is thus clearly not with the protagonist. Therefore, by comparing the original Chinese text (4a) with its English translation (4b), we show that lai is an important lexical means that the narrator uses to create a stylistic effect of mixing the narrator’s viewpoint with the protagonist’s, which is missing in the English text. By comparing (4b) with its constructed counterpart (4c), we show that zero-anaphora is a stylistic means employed by the English translator to induce the reader to identify the narrator’s viewpoint with the protagonist, which is not used in the Chinese text (4a).7 Examples (5a) and (5b) are even more complicated, as they involve the interplay of deixis and the protagonist’s perceptual content and thought represented by the narrator. (5a) 丁钩儿 推开 小 门 时, dinggouer tui-kai xiao men shi Ding Gou’er push-open small door when                                                              6  For a discussion of (free) indirect discourse and direct discourse, see Short (1996), 298–307.  7  Note that zero‐anaphora is an important grammatical phenomenon extensively discussed throughout literature  (e.g. McEnery 2000; Sanford et.al 1994, among numerous others). The point of the analysis here is simply to show  how Chinese and English uses zero‐anaphora in very different ways.  8    那条狗 猛 扑上来。 狗 的 布满 na-tiao-gou meng pu-shang-lai gou de buman that-CL-dog fierce spring-up-come dog LINK full of 汗珠 的 湿 鼻子 几乎 碰到 他 hanzhu de shi bizi jihu peng-dao ta sweat LINK wet nose almost touch-PFV he 的 手背。 准确地说 触到了 他 de shoubei zhunque-di-shuo chu-dao-le ta LINK back of hand precisely speaking touch-PFV-PFV he 的 手背, 他 感到了 它 的 de shoubei ta gan-dao-le ta de LINK back of hand he feel-PFV-PFV it LINK 鼻子 上 的 温度。 狗 鼻子 凉森森的, bizi shang de wendu gou bizi liangsensen-de nose on LINK temperature dog nose cool-LINK 使 他 想到了 紫色 的 乌贼鱼 和 shi ta xiang-dao-le zise de wuzeiyu he make he think-PFV-PFV purple LINK cuttlefish CONJ 荔枝 的 皮肤。 lizhi de pifu lychee LINK skin “When Ding Gou’er pushed the small door open, that dog came (towards him) fiercely. The sweaty wet nose almost touched the back of his hand. More precisely, it did touch his hand, as he felt the temperature of its nose. The dog’s nose is cool, which reminded him of the skin of a purple cuttlefish and a lychee nut.” (5b) Ding Gou’er pushed on the small gate, bringing the dog quickly to its feet. Its damp, sweaty nose was but a fraction of an inch from the back of his hand. In fact, it probably touched his hand, since he felt a coolness that reminded him of a purple cuttlefish or a lychee nut. An immediately noticeable difference between (5a) and (5b) is the absence of a deictic verb in (5b), which again makes use of deictic verbs a distinctive feature of Chinese narratives. However, what is furthermore interesting in the passages is the co-contribution by the deictic verb and the reported perceptual content and thought of the protagonist to the stylistic effect. In 9    the Chinese text, the stylistic effect of blending the protagonist’s viewpoint in the narrator’s language is achieved by embedding not only the deictic verb lai but also the protagonist’s perceptual content in the narration. The direct access of the narrator to the protagonist’s feeling and thought, which is impossible in real-life scenarios, creates a mixture of viewpoint representation in this stretch of discourse. However, the mixture is more vivid in the middle of (5a), where Ding Gou’er’s feeling is described by the narrator with reference to Ding Gou’er as ta ‘3rd person sg’, which represents the narrator’s perspective. The association of the coolness of the dog’s nose and a cuttlefish and lychee nut at the end of the passage is also a thought that only the protagonist, not any observer, would have access to. But now the fact that this is reported by the narrator also proves the mixture of viewpoints that we have argued. The English text employs a similar writing technique but does not use a deictic verb. The last pair of examples, (6a) and (6b), also illustrates the use of deictic verbs as a main feature in the Chinese narrative. However, there is also more to it in this stretch of discourse. (6a) 守门人 脸 上 干巴巴的, 好像 shoumen-ren lian shang ganbaba-de haoxiang gate keeper face on dry-PRT like 烤焦 的 馒头。 丁钩儿 不 kao-jiao de mantou dinggouer bu bake-burned LINK steamed bun Ding Gou’er NEG 想 继续 吓唬 他, 说 我 不是 什么 xiang jixu xiahu ta shuo wo bushi sheme want continue scare he say I NEG what 矿长, 放开胆子 烤 吧! 我 是 来 kuagzhang fangkaidanzi kao ba wo shi lai director feel free to heat PRT I PRT come 办事 的。 banshi de work PRT “The gatekeeper’s face was as dry as a burned steamed bun. Ding Gou’er did not want to scare him any more so said I am not the Director, so feel free to enjoy the warmth! I came to work.” (6b) The gatekeeper had a dried-out face, like an overcooked bun. Deciding he’d frightened the man enough, Ding Gou’er confessed that he was not the new Director, and 10    that the man was free to heat the place up as much as he liked, since Ding Gou’er had work to do. A comparison of (6a) and (6b) also shows the use of a deictic verb to be an exclusive feature of the Chinese narrative, which is absent in the English translation. In addition to the main difference, the passage also shows another striking stylistic diversion. Ding Gou’er’s speech is presented in the Chinese original with a first person point of view, indicated by the use of pronoun wo ‘I’, which is typical of direct speech and reflects the protagonist’s perspective in the narration. On the other hand, Ding Gou’er’s speech in the English translation is not presented in the form of either direct speech or free indirect speech, but rather indirect speech that shows the narrator’s distanced viewpoint, evidenced by the use of third person pronoun he. This passage thus demonstrates not only the use of deictic verbs as an important feature in Chinese narratives but also a difference in the mode of representing speech in the Chinese passage and its English translation. 4. Concluding Remarks With the use of a representative Chinese literary piece and its English translation, we have shown the usefulness of translation corpora in Chinese language research, as a comparison of texts between Chinese and another language can show various interesting differences between Chinese and the other language involved. First of all, the examples that we have presented consistently show that the use of deictic verbs seems to be a feature of Chinese narratives, which is absent, at least in all those corresponding English passages. Secondly, the use of a translation corpus allows us to see how the Chinese language uses very different stylistic means from English in creating viewpoint structure in narratives. In addition to the use of deictic verbs, different use of zero-anaphora in (3) and (4), and the different modes of speech representation discussed in (6) are all illustrations of how a typical Chinese narrative is stylistically different from an English one. However, note that the less frequent use of deictic verbs in the English text does not mean that the English language is not equipped with corresponding stylistic strategies for the purpose of viewpoint construction. Zero-anaphora in (4b), for instance, is a strategy that is used only by the English translator but not by the Chinese author in that particular stretch of discourse. An important generalization that can be made across the examples is the lack of neat correspondences between Chinese and the other language, which is exactly the potential pedagogical value of using a translation corpus for Chinese teaching. I propose that teachers of Chinese should include the use of translation corpora as a small part of their curriculum, with the purpose of showing students the impossibility of trying to find translation equivalents across languages. By pointing out how languages “misbehave”, the teacher may emphasize the importance of using original (Chinese) discourse as a study material and the inadequacy of the traditional dictionary-plus-grammar-book model (Taylor 2012:8) of language learning. In other words, the use of translation corpora should not be extensively used as a main part of the 11    curriculum but as supplementary material that helps motivate students and adjust their learning mode. It is in this sense that the use of translation corpora has important pedagogical implications at least for the material design of teaching Chinese as second language. Furthermore, given the potential of the use of translation corpora in contrastive linguistics research, we believe that differences that are identified in Chinese-English (or even ChineseCzech, Chinese-Japanese, etc.) translation corpora can be used in teaching material design, which should be brought to the attention of language teachers of Chinese with the students’ native language involved as the target language in comparison. Finally, although we have presented a selection of examples as evidence, we are fully aware that more research needs to be done with the proposed methodology so that a generalization of what is specific to the Chinese language can be made. That said, we believe that such an initiative has shown the potential of using translation corpora in Chinese language research, especially the distinctiveness of the Chinese language from a contrastive linguistic perspective. References Chen, Alvin Cheng-Hsien. 2010. “A conceptual understanding of bodily orientation in Mandarin: A quantitative corpus perspective”. Corpus Linguistics and Linguistic Theory 6 (1), 1–28. Chung, Siaw-Fong and Chu-Ren Huang. 2010. “Using collocations to establish the source domains of conceptual metaphors”. Journal of Chinese Linguistics, 38 (2), 183–223. Dancygier, Barbara. 2012. The Language of Stories: A Cognitive Approach. Cambridge: Cambridge University Press. Goldblatt, Howard. 2012. The Republic of Wine: A Novel. Delaware: Arcade Publishing. Lu, Wei-lun. Accepted. “A Cognitive Linguistic approach to teaching spatial particles: From contrastive constructional analyses to material design”. In Kyoko Masuda and Carlee Arnett (eds.) Cognitive Linguistics and Sociocultural Theory. Berlin: de Gruyter. Lu, Wei-lun and I-wen Su. 2009. “Speech in interaction: Mandarin particle le as a marker of intersubjectivity”. Zeitschrift für Interkulturellen Fremdsprachenunterricht 14.155-68. McEnery, Tony. 2000. Corpus-based and Computational Approaches to Discourse Anaphora. Amsterdam: John Benjamins 12    Mo, Yan. 2012. Jiu guo (The Republic of Wine). Shanghai: Shanghai wenyi chubanshe. Sanford, A., L. Moxey and K. Patterson 1994. Psychological studies of quantifiers. Journal of Semantics 11, 153–170. Short, Mick. 1996. Exploring the Language of Poems, Plays and Prose. London: Longman. Taylor, John R. The Mental Corpus: How Language Is Represented in the Mind. Oxford: Oxford University Press. Xiao, Richard and Naixing Wei. 2014. “Translation and contrastive linguistic studies at the interface of English and Chinese: Significance and implications”. Corpus Linguistics and Linguistic Theory 10 (1), 1–10. Xu, Jiajin and Xiaochen Li 2014. “Structural and semantic non-correspondences between Chinese splittable compounds and their English translations: A Chinese-English parallel corpus-based study”. Corpus Linguistics and Linguistic Theory 10 (1), 79–101. Internet resource consulted Leipzig Glossing Rules (http://www.eva.mpg.de/lingua/resources/glossing-rules.php) Contact address of the author Dr. Wei-lun Lu (weilunlu@gmail.com) Faculty of Arts of Masaryk University, Arna Novaka 1, 60200 Brno, Czech Republic