I. Foundations of Translation (2,500 Words → 3,200 Words)
1.1 Defining Translation
- Beyond Lexical Transfer:
- Problem: 牛饮 (niú yǐn) ≠ “cow drink”
- Solution: Dynamic equivalence (“gulp down”) + cultural note (Chinese farming imagery)
- Text as Living Organism: Contextual dependency of terms (e.g., 意思 varies in “没意思” vs. “意思意思”)
1.2 Theoretical Frameworks
| Theory | Application to C-E Translation | Case Study |
|---|---|---|
| Nida’s Dynamic Equivalence | 对牛弹琴 → “Cast pearls before swine” (function > form) | Buddhist sutra localization |
| Reiss’ Text Typology | Technical manual: Omit 之乎者也 → Use ISO syntax | GB/T 19001 translation |
| Venuti’s Resistance | Retain 胡同 as hutong + italicization | Cultural preservation in Mo Yan’s novels |
1.3 Linguistic Contrasts
- Syntax:
- Chinese topic-prominent (主题突出): 这件事,我们没办法 → “As for this matter, we can do nothing”
- English subject-prominent: Requires dummy subjects (“It is…”)
- Morphology: Lack of conjugation in Chinese → Time markers must be added (“曾经去过” → “had visited”)
II. Analyzing Source Text (2,000 Words → 2,800 Words)
2.1 Contextual Triangulation
2.2 Rhetorical Device Decoding
- Parallelism (排比):
- Original: 山不在高,有仙则名。水不在深,有龙则灵。
- Translation: “Mountains need not be high; they’re famous with immortals. Waters need not be deep; they’re sacred with dragons.” (Retains rhythm via semicolons)
- Classical Chinese (古文):
- 子曰 → “The Master said” (Confucian texts) vs. “Confucius remarked” (modern readership)
III. Core Translation Techniques (3,500 Words → 4,500 Words)
3.1 Technique Matrix with 20+ Examples
| Technique | Chinese Source | Literal Translation | Adapted Translation | Rationale |
|---|---|---|---|---|
| Amplification | 请慢走 | “Please walk slowly” | “Please watch your step” | Adds safety connotation |
| Transposition | 他唱歌唱得好 | “He sings songs well” | “He is a good singer” | Converts verb→noun structure |
| Division | 她吓得心砰砰直跳 | “She scared heart pound jump” | “Her heart pounded with fear” | Splits S-V-O into clause |
3.2 Nuance-Driven Solutions
- Untranslatables:
- 缘分 (yuánfèn) → “serendipitous affinity” (philosophical) / “karmic connection” (religious context)
- Measure Words:
- 一匹马 (yī pǐ mǎ) → “a horse” (omits 匹) BUT retain in linguistics papers: “pi classifier for animals”
IV. Genre-Specific Strategies (6,000 Words → 7,000 Words)
4.1 Literary Translation Deep Dive
- Poetry Constraints:
Challenge Solution Example Tonal patterns Replace with stress-meter (e.g., iambic) 床前明月光 → “Before my bed, moonlight glows” Classical allusions Footnote + simplified metaphor 夸父追日 → “Kuafu chases the sun1”
4.2 Technical Translation Protocol
Step 1: Extract terminology → Create TM (Translation Memory)
Step 2: Validate against ISO/GB standards
Step 3: Post-edit with QA tools (e.g., Xbench flagging 的/地/de errors)
Step 4: Consistency check: 螺丝 vs "screw"/"bolt" per industry
4.3 Media Localization: Trending Terms
- 内卷 (nèijuǎn) → “involution” (academic) / “rat race” (social media)
- 社死 (shè sǐ) → “social doom” (literal) / “digital embarrassment” (domesticated)
V. Linguistic Pitfalls (3,000 Words → 3,500 Words)
5.1 Structural Minefields
- Zero-Subject Sentences:
- 下雨了 → Add dummy subject: “It’s raining”
- 改革开放以来 → Add agent: “China has… since reform and opening-up”
- Directional Complements:
- 想起来 (“recall”) vs. 坐起来 (“sit up”) → Contextual disambiguation
5.2 False Friends Database
| Chinese | Naive Translation | Accurate Equivalent | Domain |
|---|---|---|---|
| 红茶 | “red tea” | “black tea” | Food & Beverage |
| 干货 | “dry goods” | “substantive content” | E-commerce |
VI. Technology Section (1,500 Words → 2,200 Words)
6.1 CAT Tool Workflow
6.2 NMT Failure Analysis
- Polysemy Errors:
- Input: “打文件” (dǎ wénjiàn)
- NMT Output: “hit documents” (Correct: “type documents”)
- Fix: Forced glossary entries: 打 → “type” when collocated with 文件
VII. Ethics & Practice (1,500 Words → 2,000 Words)
7.1 Geopolitical Sensitivity Protocol
IF source contains "台湾" THEN
USE "Taiwan, China" (UN style)
AVOID "Republic of China"
ADD footnote explaining PRC's stance IF audience is academic
END IF
7.2 Copyright Frameworks
- Derivative Works: Translating 金庸’s novels → Require Shaw Brothers’ permission
- Moral Rights: Translator’s right to be credited (Berne Convention Article 6bis)
Expansion Toolkit
To achieve 20,000+ words:
-
Embedded Exercises:
- Translate: “老张拍桌子瞪眼睛,吓得小王一愣一愣的”
- Hints: Onomatopoeia (拍), idiom (瞪眼睛), reduplication (一愣一愣)
-
Annotated Parallel Corpus:
Chinese English (Literal) English (Adapted) Commentary 风里来雨里去 “Come in wind, go in rain” “Braving all weathers” Idiom compression -
Glossary of 100+ “Untranslatables” with risk ratings:
- 客气 (kèqi): ★★★☆ (Medium risk) → “polite deference”/“ceremonial humility”
-
CAT Tool Screenshots: Trados Studio interface highlighting termbase conflicts.
This structure provides depth without redundancy, using:
- Multimodal learning (diagrams/tables/code)
- Industry-aligned protocols (ISO, UN terminology)
- Controversy navigation (political/ethical dilemmas)
Chinses!
中文版参考:
###I.翻译基础(2500字)→ 3,200字)
####1.1定义翻译
-超越词汇转换:
-问题:牛饮 (niúy"n)≠“牛饮”
-解决方案:动态对等(“狼吞虎咽”)+文化注释(中国农业意象)
-文本作为生物体:术语的上下文依赖性(例如。,意思 “无心”与“无心”的含义有所不同。)
####1.2理论框架
| 理论 | 在汉英翻译中的应用 | 案例研究 |
|---|---|---|
| 奈达的动态对等理论对牛弹琴 → “猪前铸珠”(功能>形式) | 佛经本土化 | |
| Reiss的文本类型学 | 技术手册:省略之乎者也 → 使用ISO语法 | GB/T 19001翻译 |
| Venuti的抵抗 | 保留胡同 如胡同+斜体 | 莫言小说中的文化保护 |
| ####1.3语言对比 | ||
| -语法: | ||
| -中文话题突出:“对于这件事,我们无能为力” | ||
| -英语科目突出:需要虚拟科目(“It is…”) | ||
| -形态学:中文中缺少共轭→ 必须添加时间标记(“曾经去过” → “已访问”) |
二、分析源文本**(2000字→ 2,800字)
####2.1上下文三角测量
图表LR
A[源文本]-->B{词汇分析}
A-->C[实用目的]
A->D[文化潜台词]
B-->E[词义消歧]
C->F[寄存器映射]
D->G[历史典故]
####2.2修辞手段解码
-平行度:
-原件:山不在高,有仙则名。水不在深,有龙则灵。
-翻译:“山不必高,以神仙而闻名。水不必深,以龙而神圣。”(通过分号保持节奏)
-文言文:
- 子曰 → “大师说”(儒家经典)与“孔子说”(现代读者)
###III.核心翻译技巧(3500字)→ 4,500字)
####3.1技术矩阵,包含20+个示例
| 技巧 | 中文来源 | 直译 | 改编翻译 | 基本原理 |
|---|---|---|---|---|
| 放大 | 请慢走 | “请慢慢走” | “请小心脚下” | 增加安全内涵 |
| 换位 | 他唱歌唱得好 | “他唱歌唱得很好” | “他是个好歌手” | 转换动词→名词结构 |
| 部门 | 她吓得心砰砰直跳 | “她吓得心砰砰直跳” | “她的心因恐惧而砰砰直响” | 将S-V-O拆分为子句 |
| ####3.2 Nuance驱动的解决方案 | ||||
| -无法翻译: |
- 缘分 (尤南芬)→ “偶然的亲和力”(哲学)/“业力联系”(宗教语境)
-测量单词: - 一匹马 (yīpǐm.)→ “马”(省略匹) 但在语言学论文中保留:“动物的pi分类器”
###IV.体裁策略(6000字)→ 7,000字)
####4.1文学翻译深度学习
-诗歌限制:
| 挑战 | 解决方案 | 示例 |
|---|---|---|
| 音调模式 | 替换为应力计(例如,抑扬格) | 床前明月光 → “在我的床前,月光闪闪发光” |
| 经典典故 | 脚注+简化隐喻 | 夸父追日 → “夸父追逐太阳1” |
| ####4.2技术翻译协议 |
第一步:提取术语→ 创建TM(翻译记忆库)
步骤2:根据ISO/GB标准进行验证
步骤3:使用QA工具进行后期编辑(例如,Xbench标记的/地/消除错误)
步骤4:一致性检查:螺丝 与每个行业的“螺钉”/“螺栓”相比
####4.3媒体本地化:流行术语
- 内卷 (nèijuǎn)→ “内卷化”(学术)/“老鼠赛跑”(社交媒体)
- 社死 (嘘")→ “社会末日”(字面意思)/“数字尴尬”(驯化)
###V.语言陷阱(3000字→ 3,500字)
####5.1结构性雷区
-零主题句:
- 下雨了 → 添加虚拟主题:“下雨了”
- 改革开放以来 → 添加代理:“中国自改革开放以来…”
-方向性补充: - 想起来 (“召回”)vs。坐起来 (“坐起来”)→ 上下文消歧
####5.2虚假好友数据库
|中文|直译|准确对等|域名|
|-------------|------------------------|-------------------------|-----------------|
| 红茶 | 红茶|红茶|食品饮料|
| 干货 | “干货”|“实质性内容”|电子商务|
###VI.技术部分(1500字)→ 2,200字)
####6.1 CAT工具工作流程
图TB
A[中文来源]-->B{细分}
B-->C[TM查找]
C-->D[术语库匹配]
D->E[机器翻译]
E->F[后期编辑]
F-->G[QA检查:标签验证]
####6.2 NMT故障分析
-多义错误:
-输入:“文件”
-NMT输出:“点击文档”(正确:“键入文档”)
-修复:强制术语表条目:打 → 与搭配时使用“type”文件
七、道德与实践**(1500字→ 2,000字)
####7.1地缘政治敏感性协议
如果源包含“台湾”,则
使用“中国台湾”(联合国风格)
避免“中华民国”
如果听众是学者,请添加脚注解释中国的立场
结束IF
####7.2版权框架
-衍生作品:翻译金庸’s小说→ 需要邵氏兄弟的许可
-精神权利:译者的署名权(伯尔尼公约第6条之二)
###扩展工具包
达到20000+字:
1.嵌入式练习:
-翻译:“张晓明,王小君一感兴趣。”
-提示:拟声词、成语、重叠
2.带注释的平行语料库:
| 中文 | 英文(直译) | 英语(改编) | *评论 |
|---|---|---|---|
| 风里来雨里去 | “风雨无阻” | “不畏风雨” | 成语压缩 |
| 3.100多个“不可翻译”术语表,包括风险评级: |
- 客气 (凯奇):★★★☆ (中等风险)→ “礼貌的尊重”/“礼仪上的谦逊”
4.CAT工具截图:Trados Studio界面突出显示术语库冲突。
这种结构提供深度而没有冗余,使用:
-多模式学习(图表/表格/代码)
-行业一致协议(ISO、联合国术语)
-争议导航(政治/伦理困境)
1566

被折叠的 条评论
为什么被折叠?



