Chapter 4 NTI Identification and Rewrite/Pre-editing Rule Formulation
4.2 Formulation of NTIs and rewrite rules
meta-97
language for categorizing JA-based CLs and, by extension, NTIs since NTIs are the lexical, syntactic, or semantics features that trigger specific CL rules.
Moreover, O’Brien’s (2003) CL rule classification framework is useful for this research in understanding what kind of linguistic phenomena can be governed by a CL rule and can be related to a translation error. The explanation of each sub-category is helpful as it can provide guidance for designing the rewrite/pre-editing rules after an NTI is identified. (See Appendix 1 for O’Brien’s CL rule categories and explanation in detail.)
98
Table 4.3 NTIs allocated based on O’Brien’s CL rule classification
Category Sub-Category NTI
Lexical Vocabulary Usage Noun (verbal): 支援, 提案, 反映, 発信, 追求, 活用, 提供, 活動, 配信, 伝達, 評価, 対応, 貢献, 抽出, 導入, 決定, 徹底, 啓蒙, 活性化, 合意, 連携, 締結
Verb: 活かす/生かす, 努める, 取り込む, 加わる, 考える, 取り入れる, 求められる, 言わ
れる, 問われる
Noun: 社内, 国内, 戦前, 想い, 考え, 考え方, 拠点, 現場, 実績, 地域, 分野, 事務局, 管 理職社員, 生活者, 有力企業, 前置機関, 当社グループ, 相手本位, 職方, 見守り, キャリ ア意識
Noun/Verb: 取り組み/取り組む
Adjective: 独自, 高度, 適切, 最適, 適正, 確実
Loan word: オフィス, センター, ダイバーシティ, イメージ, モビリティ, ニーズ, トー
タル, キーワード, マテリアリティ, レビュー, システム, モデル, ロールモデル Abbreviation/Acronym Usage Inappropriate style is used to indicate an acronym
Prefix/Suffix Usage Use suffix “系” Use suffix “化” Use suffix “性”
Pronoun Usage これら
その/それ
Conjunction Usage その結果
Date Format Only use DD to indicate a date omitting MM
Note: Sub-categories that are not allocated with NTIs are: Comparatives and Superlatives, Word Division, Synonymy, Verb Form Usage, Spelling, Anaphoric Reference, Quantifier Usage, Negation, Relative Pronoun Usage, Numbering, Dictionary Usage, Polysemy, Clarity, Word Combination
Syntactic Subject-Verb Agreement Subject and verb do not agree
The agent of 講演 is an organization
Ellipsis [N1に対するN2] is abbreviated to [N1のN2]
[N1に関するN2] is abbreviated to [N1のN2]
[N1についてのN2] is abbreviated to [N1のN2]
[N1というN2] is abbreviated to [N1のN2] or [N1+N2]
[N1をVinf+N2] is abbreviated to [N1のN2]
99
[N1によるN2] is abbreviated to [N1のN2]
[Vsahen-passive+N] is abbreviated to [Vsahen-stem+N]
[V1sahen-stemをV2sahen-stemする] is abbreviated into [V1sahen-stem+V2sahen-stem]
[N1をVsahenするN2] is abbreviated to [N1のVsahen-stem+N2]
[N+『』/「」] is used to indicate reference Noun Cluster Size/Structure (N1+)N2+Vsahen-stem+N3
Coordination N1やN2
N1やN2、N3など N1とN2、およびN3 N1やN2などのN3 N1やN2、N3
N in [N1、N2、N3、Nn] do not belong to the same category Coordination structure unclear
Punctuation ※ is used to indicate a superscription or note A centered dot (・) is used to connect two nouns
／ is used to indicate a slash
Note: Sub-categories that are not allocated with NTIs are: Modifier Usage, Adjective Functionality, Adverb Functionality, Article Usage, Pronoun Usage, Preposition Usage, Participle Usage, Tense, Number, Voice, Mood, Case, Person, Modals, Apposition, Queries, Parallelism, Repetition, Lists, Segment Independence
Textual Information Load Modifier is too long (Limit: 20) Sentence is too long (Limit: 120)
Phrase beforeとして/においてis too long (Limit: 20) Phrase beforeを目的とするis too long (Limit: 20) Information Structure A definition is included in the modifier
Specificity of Information A statement is not logical Subject is unclear
The relation between phrases is unclear
Textual Devices Play on words
Verb Form Usage Use [Vinf+N] to indicate potential or present progressive
Note: Sub-categories that are not allocated with NTIs are: Text Structure, Layout, Sentence Length, Paragraph Structure, Paragraph Length, Keyword Usage, Word counting, Capitalization, Use of Parentheses, Pragmatic, Text Type Structure, Text Type Labeling, Text Purpose
100
However, 29 of the NTIs (as shown in Table 4.4), including JA-specific syntactic and textual features, required the creation of new subcategories (6 syntactic and 2 textual) to accommodate them within O’Brien’s typology, given that O’Brien’s typology was based on CL rules for authoring texts in EN.
Table 4.4 NTIs not accommodated by O’Brien’s typology
Category Added sub-category NTI
Syntactic Particle usage End a sentence with a particle The meaning of “で” is unclear Syntactic Compound particle usage
ながらの/からの、において、となる/による、
では、のような、などのほか、に対して、に 向けて、N1に向けたN2
Syntactic Collocation Awkward collocation between verb and object Awkward collocation between adjective and noun Syntactic Auxiliary verb usage たい、ように、このように
Syntactic Adnominal adjective usage こうした
Syntactic Sentence patterns
(V1sahen-stem+)V2sahen-stemを実現する Vsahen-stemを検討する
Vsahen-stemが進められる Vsahen-stemを進める
N1やN2+を/について+V1・V2 NのV1sahen-stem・V2sahen-stem
…化をVsahen Adj(na)stemをＶ
…し、…するため ため、…ように
N1、N2、N3の3 + Counter
[N1+を+Vinf·past+N2] is used to indicate a means Textual Style Use English as a title
Textual Political correctness Biased expression is used
In principle, rewrite rules for lexical NTIs are formulated by offering a substitute expression to be used in a restricted context. For example, if “考え” (idea, thought, opinion, notion, view, plan, or intention) is used in a sentence with the grammatical Subject/Actor being a company, replace it with “予定” if it refers to a plan, and replace it with “理念” or “方針” if it refers to a concept or policy. Rewrite rules for syntactic NTIs, on the other hand, specify an alternative structure. For example, the instruction
101
for “Vsahen-stemを進める” is to replace it with “Vsahen-stem+する.” For textual NTIs, the rewrite rules provide suggestions to avoid problematic features. However, there are NTIs for which specific rules or suggestions are difficult to give. Examples are: “The relation between phrases is unclear” and “Subject and verb do not agree.” These belong to the violation of general writing guidelines. Therefore, this study will only provide general guidelines for such NTIs.
As shown in Appendix 2, which presents the list of NTIs with the number of occurrences and the translation error(s) associated with each NTI, 90 out of 142 NTIs are shown as occurring only once. The reasons why NTIs that appeared only once were identified as such are: (1) the present author’s experience as a JA to ZH-CN translator and linguistic solution consultant allows her to easily recognize those problematic features that tend to cause translation issues or require more time in finding an appropriate translation; (2) the occurrences in Appendix 2 were based on the bilingual corpus built in Chapter 3 the JA word count of which is only 24,001. The frequency/impact of each NTI is to be established in the 500,000-character corpus created in Chapter 5.
Taking examples from the category Lexical, the JA word “地域” occurred nine times inducing seven Misrepresentation] errors, one [TT-Accuracy-Calque], and one [TT-Fluency-Awkward/Unidiomatic] error. This word is used to designate a region, an area, or a local community. It is often used in the CSR reports to mean the neighborhood of the company that has published the report. However, the ZH-CN word “地域” basically refers to a region or an area, and does not include the concept of local community or neighborhood of a company. Consequently, accuracy errors are likely to occur if it means “a local community” in the JA ST. JA and ZH-CN share a significant number of lexical items that use the same kanji (Chinese ideograph) but differ in meaning or usage, which is observed to be one of the main causes of translation errors.
102
Taking other examples from the category Syntactic, a selection of NTIs that fall under “Ellipsis” is shown in Table 4.5. The common factor is that these structures are all abbreviated to “N1 no N2.”
Table 4.5 Example of NTIs and the rewrite rule (syntactic) NTI Type and occurrences of associated
translation errors
Rewrite rule
[N1に対するN2] is abbreviated to [N1のN2]
TT-Accuracy-Calque*1
If “N1 no N2”
does not indicate possession, add compound particles to make the relation of N1 and N2 clear.
[N1に関するN2] is abbreviated to [N1のN2]
TT-Fluency-Awkward/Unidiomatic*5 TT-Accuracy-Calque*2
TT-Fluency-Grammar*1 [N1についてのN2] is
abbreviated to [N1のN2]
TT-Fluency-Grammar*1 [N1というN2] is abbreviated to
[N1のN2] or [N1+N2]
TT-Accuracy-Calque*1 [N1をVinf+N2] is abbreviated
to [N1のN2]
TT-Accuracy-Calque*1 [N1によるN2] is abbreviated to
[N1のN2]
TT-Accuracy-Calque*1
A noun phrase with the structure “N1 no N2” can have multiple interpretations.
For example, “村上の本” can be “book(s) Murakami wrote,” “book(s) Murakami owned,” or “book(s) about Murakami.” In the first two interpretations, N1 (=Murakami) is interpreted as the subject and N2 (=book or books) as the direct object.
However, in the third interpretation, neither N1 nor N2 receives a case interpretation.
According to Makino and Tsutsui (2018, p. 29), the term “case” here refers to the grammatical property of a given noun, which determines its role in the sentence in connection with the predicate. It can function as a subject, direct object, place, time, means, etc. “Being in receipt of case interpretation” means a noun is interpreted to have one of those properties.
103
As described by Makino and Tsutsui (2018, pp. 29-34), the relationship between N1 and N2 varies: (1) N1 can be the qualifier of N2; (2) N2 can be an attribute of N1;
(3) N2 can represent something related to N1; (4) N1 can specify a time frame for N;
and so forth. In short, “N1 no N2” is an economic expression for Japanese native speakers in the sense that there is no need to make explicit the relationship between N1 and N2. From the bilingual corpus collected in Chapter 3, it is observable that Japanese writers tend to use this structure extensively for all these relationships. The assumption is that it takes less effort for writing, and it can easily fit into the generally lengthy sentences. Sometimes, it is useful to make the meaning opaque on purpose.
“N1的N2” is a structure in ZH-CN, which indicates possession (similar to “N1’s N2” in EN and“N1 no N2” in JA ). It has a much narrower range of meanings than “N1 no N2,” yet translators tend to use “N1 de N2” to translate “N1 no N2” indiscriminately.
As noted in Table 4.5, the structure “N1 no N2” tends to trigger the accuracy error
“Calque” and fluency errors such as “Grammar” and “Awkward/Unidiomatic.” The assumption is that translators cannot understand the precise relationship between N1 and N2 from the simplified structure, leaving them no choice but to translate the ST word for word. Depending on what N1 and N2 are, the degree of the translation error varies.
The NTI “Modifier is too long,” which falls in the category Textual, has 17 occurrences and tends to trigger [TT-Accuracy-Misrepresentation] errors. A text with long modifiers is found to be difficult to understand for assessors. Table 4.6 shows an example of a sentence with such an NTI and its official ZH-CN and EN translations for reference. What modifies the noun “職場風土” (workplace culture) is highlighted in grey. The official ZH-CN and EN translations have respected the sentence structure of the JA ST as much as possible. The error labeled was “TT-Fluency-Awkward/Unidiomatic”: while the text is understandable, it is clearly awkward in both its official ZH-CN and EN translations.
104
Table 4.6 Example of the NTI “Modifier is too long”
JA 人権尊重はもとより、国籍、人種、性別、宗教、年齢、学歴、キャリア、障碍 の有無など、さまざまな背景からなる個性や価値観、ライフスタイルを互い に尊重し合う職場風土の醸成に努めています。
ZH-CN 除了尊重人权之外，公司还致力于营造使不同国籍、人种、性别、宗教、
年龄、学历、职历、有无残疾等各种背景中形成的个性、价值观和生活 方式都能相互尊重的职场文化。
EN The XXX Group is working to develop a work climate that respects diverse individual characteristics, concepts of value and lifestyles of different backgrounds, such as respect for human rights, nationality, race, gender, religion, age, academic background, career, and the presence or absence of disabilities.
As an example of “Compound particle usage,” “N1に向けたN2” occurred twice and each occurrence was labeled [TT-Accuracy-Misrepresentation]. “に向けた” is a compound particle that indicates the destination/direction of movement, the target/direction of an action, or the goal/purpose of an action. “向” in ZH-CN means
“toward,” indicating the destination/direction of movement. [TT-Accuracy-Misrepresentation] occurred when “に向けた” was used to indicate the goal/purpose of an action in the ST. The interference of kanji is again observed.
The rewrite/pre-editing rules and examples presented in Appendix 2 are drafted in JA as they are intended for Japanese native speakers who can apply this NTI set. For the sake of readers who do not understand JA, explanations of the rules, examples of JA features, and pre-editing in EN for an illustrative sub-set of NTIs are given here.
105
NTI#1: 支援 (assist/support)
Rule
JA: 誰（何）が誰（何）をどのように（経済的に? 技術的に? 物理的 に?）「支援」するのかが不明確になる場合は、他の表現に置き換える。
EN: Replace the word “支援” (assist/support) with an alternative to explicitly indicate who or what “assists” whom or what and how (for example, financially, technically, or physically).
Example of JA text with the NTI
当社は、XXXコンテストを第1回から支援しています。
EN gloss We have supported the contest since its first year.
Example of
rewriting 当社は、XXXコンテストに第1回から協賛しています。
EN gloss We have co-sponsored the contest from its first year.
NTI#2: 提案 (suggest/suggestion, propose/proposal)
Rule
JA: 誰に対して何を「提案」するのかが不明確になる場合は、他の表 現に置き換えて「誰に対して何を」の部分を明確にする。
EN: Add information to make it clear what the “提案” (suggestion proposal) is, and to whom it was made.
Example of JA text with the NTI
XXXは、様々な地域課題の解決策を提案しています。
EN gloss XXX has been suggesting various solutions regarding environmental problems.
Example of rewriting
XXXは、様々な地域課題の解決策を各自治体に対して提案していま す。
EN gloss XXX has been suggesting various solutions regarding environmental problems to municipalities.
106
NTI#3: 反映 (reflect/reflection)
Rule
JA:中国語にも「反映」という表現はあるが、日本語の「反映」とは意味 と用法が異なり、誤訳につながりやすいため、できるかぎり他の表現に 置き換える。
EN: Avoid using the word “反映” (reflect/reflection) because it also exists in ZH-CN with a different meaning.
Example of JA text with the NTI
XXXのサービスへ女性の価値観を反映する。
EN gloss Reflect women’s sense of values in XXX’s services.
Example of
rewriting 女性の価値観を考慮してXXXのサービスを改善する。
EN gloss Improve XXX’s services in consideration of women’s sense of values.
NTI#6: 活用 (exploit/make the most of)
Rule
JA:中国語にも「活用」という表現はあるが、日本語の「活用」とは意味 と用法が異なり、誤訳につながりやすいため、類似する他の表現に置 き換える。文脈に応じて「有効利用」、「利用」などに置き換えてくださ い。
EN: Avoid using the word “活用” (exploit/ make the most) because it also exists in ZH-CN with a different meaning. Replace it with “有効利用” (make the most of, take advantage of) or “利用” (use, utilize) based on the context.
Example of JA text with the NTI
水素エネルギーを活用して持続可能な社会の実現をめざす
EN gloss Exploit hydrogen energy to realize a sustainable society.
Example of
rewriting 水素エネルギーを利用して持続可能な社会の実現をめざす
EN gloss Using hydrogen energy to realize a sustainable society.
107
NTI#7: 提供 (provide/deliver)
Rule
JA:「提供」とは具体的にどういう行為なのを明確に表現できる言葉に置き換 える。また、誰に対して「提供」するのかが不明確になる場合は、「誰に対し て」の部分を明確にする。
EN: Replace the word “提 供” (provide/deliver) with a more specific verb to indicate what the behavior is, and make clear to whom something is provided.
Example of JA text with the NTI
XXXでは、これからの時代の医療を支える新しいソリューションを提 供しています。
EN gloss XXX provides new solutions that support the future of healthcare.
Example of rewriting
XXXでは、これからの時代の医療を支える新しいソリューションを販 売しています。
XXXでは、YYYに向けてこれからの時代の医療を支える新しいソリ ューションを提供しています。
EN gloss
XXX sells new solutions that support the future of healthcare XXX provides YYY with new solutions that support the future of healthcare
108
NTI#10: 伝達 (communicate/communication)
Rule JA: 中国語にも「伝達」という表現はあるが、人間どうしの情報のやり
取りに限定している傾向がある。それに合わせて「伝達」を使用し、人 間どうしの情報のやり取りでない場合は、他の表現に置き換える。
EN: The word “伝達” (communicate/communication) exists in ZH-CN specific to human-to-human communication. Replace it with other expressions if it human-to-human.
Example of JA text with the NTI
XXXとは、財務情報の効率的な伝達のために開発されたコンピュー タ言語です。
EN gloss XXX is a computer programming language developed for efficient communication of financial information.
Example of rewriting
XXXとは、財務情報の効率的な交換のために開発されたコンピュー タ言語です。
EN gloss XXX is a computer programming language developed for the efficient exchange of financial information.
109
NTI#14: 抽出 (extract/extraction)
Rule JA:中国語にも「抽出」という表現はあるが、「データの抽出」といったコ
ンピュータに関する文脈で使用されるケースが大半であり、それ以外 の文脈で「抽出」を使用すると誤訳につながりやすいため、他の表現 に置き換える。
EN: The word “抽出” (extract/extraction) also exists in ZH-CN, but in general, it is only used in the context of computers (e.g., data extracting data) and does not apply to human actions. Restrict the use of “抽出” to “(data) extraction,” or replace it with other verbs if otherwise.
Example of JA text with the NTI
社会課題を幅広く抽出するため、ステークホルダー別に意見を収集す る。
EN gloss To extract a broad spectrum of social issues, collect feedback from each stakeholder.
Example of rewriting
社会課題を幅広く認識するため、ステークホルダー別に意見を収集す る。
EN gloss To discover a broad spectrum of social issues, collect feedback from each stakeholder.