English Language Teaching and Linguistics Studies ISSN 2640-9836 (Print) ISSN 2640-9844 (Online) Vol. 7, No. 4, 2025 www.scholink.org/ojs/index.php/eltls 51 Original Paper ACorpus-based Study on the Translation of Shallow Needling Related Chapters in Huangdi Neijing Zhiqiong Ye1*, Liangbing Wu2, Yingzhu Chen2, & Quanrui Jiang2 1 Guangzhou Songtan Polytechnic College, Guangzhou 511370, China 2 Guangxi University of Chinese Medicine, Guangxi 530020, China * Zhiqiong Ye, Guangzhou Songtan Polytechnic College, Guangzhou, China Received: May 29, 2025 Accepted: July 29, 2025 Online Published: August 13, 2025 doi:10.22158/eltls.v7n4p51 URL: http://dx.doi.org/10.22158/eltls.v7n4p51 Abstract Objectives This study aims to analyze chapters related to shallow needling in Li Zhaoguo's and Wu Liansheng & Wu Qi's English versions, identify the reasons for differences and draw a certain reference for its translation in the future. Methods In this study, the two translations and Chinese original version were constructed into a corpus. WordSmith 4.0, AntConc 3.4.4w, and SPSS 26 were employed for qualitative and quantitative analysis at the lexical, syntactic, and discourse levels of translations related to shallow needling. Results 1) Similarities: There is no significant difference in vocabulary density and quantity between the two translations (P>0.05), and both emphasize the use of conjunctions to express semantic relations. 2) Differences: Wu’s translation has longer word lengths compared to Li’s (P<0.01). Li often uses transliteration with literal translation and annotation, while Wu tends to apply free translation. The differences in sentence length are relatively small in Li's translation (P<0.05), and sentences are more balanced. The two translations have significant differences in using “when” (P<0.05). These differences are closely related to the translators' purposes, strategies, methods, and personal expression habits. Conclusion For the international dissemination of shallow needling, vocabulary translation can draw inspiration from Li's methods and strategies to convey more cultural knowledge related to shallow needling. Sentence translation can benefit from Wu's explicit representation of underlying medical information, facilitating the dissemination of medical knowledge. In terms of discourse, attention should be paid to conveying semantic relationships within passages. Keywords Corpus, Huangdi Neijing, Shallow needling, Translation study www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 52 1. Introduction The depth of needle insertion is one of the key factors determining the clinical efficacy of acupuncture treatment (Wu et al., 2024). Shallow needling refers to a technique in which the needle is inserted shallowly into acupoints to stimulate qi, dredge meridians, regulate the functions of the internal organs and balance yin and yang, thereby treating diseases (Li et al., 2022). Studies have shown that shallow needling has a wide range of applications and provides significant therapeutic effects for various conditions (Gong & Wang, 2023; Li et al., 2022; Cong et al., 2022; Li et al., 2021). Furthermore, this technique is almost painless and thus more acceptable to patients abroad. Many overseas acupuncturists also focus on the clinical application and research of shallow needling. For instance, the “Sugar Acupuncture — Comfort-Oriented Acupuncture” method in the United States is a developed form of this approach (Wang, 2022). As early as two thousand years ago, Chinese physicians placed great emphasis on the depth of needle insertion. While both deep and shallow needling were discussed, shallow needling was predominant (Mo & Xu, 2019). The Huangdi Neijing (hereafter referred to as Neijing) contains extensive discussions on shallow needling, including relevant theories, instruments, techniques, and clinical applications, particularly in chapters such as Pibu Lun (皮部论), Jiuzhen Shier Yuan (九针十二原), Guan Zhen (官针), Xiao Zhen Jie (小针解), and Zhong Shi (终始) (Yu et al., 2021). These chapters provide significant guidance for modern clinical acupuncture (Yu et al., 2021) and hold great value for the international dissemination of the shallow needling technique. This study builds a bilingual corpus based on the original text of Neijing and its two complete English translations. Through a combination of qualitative and quantitative analysis, it examines lexical, syntactic, and discursive features of shallow needling-related chapters in the two translations. The objective is to uncover the reasons behind translation differences and offer insights for the English translation and overseas dissemination of shallow needling. 2. Corpus and Translation Studies on Neijing The term “Corpus Translation Studies (CTS)” was first proposed by Baker (1993), who pioneered the application of corpora in translation studies. In China, Yang Huizhong (1993) was among the earliest scholars to explore the relationship between corpus linguistics and machine translation, nearly in parallel with the international development of CTS (Yang & Bai, 2010). According to the literature reviewed by the author, the earliest application of corpora to Neijing began around 2008, mainly focused on the construction of corpus frameworks. In 2015, Zhu Jianfei (2015) built a bilingual corpus of Neijing and used it to explore translation strategies for classical Chinese medicine texts. Based on a parallel corpus, Ye Xiao et al. (2015) conducted a comparative study of pulse terminology translations. Liu Chunmei (2016) illustrated how to construct a bilingual parallel corpus using selected content from Neijing. Two years later, He Nanna et al. (2018) applied corpus analysis to the translation of neurological disease names in Neijing, comparing them with standardized medical terminologies. In www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 53 recent years, studies combining corpus methods and the classic have increased. Pan Hai’ou (2022) constructed a corpus for the chapter Yinyang Yingxiang Dalun (阴 阳 应象 大 论 ) and analyzed vocabulary and translation strategies across versions by Ilza Veith, Maoshing Ni, and Yang Mingshan. Lu Dechao et al. (2024) used corpus tools to examine the translation of qi in Wiseman’s translation of Suwen (素问 ). However, little research has specifically addressed the English translation of shallow needling in Neijing. Some existing work only briefly mentions related techniques such as bloodletting. In addition, the combination of corpus methodology with translation studies of shallow needling remains relatively scarce. This study aims to fill that gap by conducting a qualitative and quantitative investigation of shallow needling related chapters in two English versions of Neijing, offering a more objective and comprehensive view of translation features and their implications for the international communication of this acupuncture technique. 3. Research Design 3.1 Research Corpus and Questions 3.1.1 Research Corpus This study selects five chapters from Neijing that are related to shallow needling: Pibu Lun from Suwen, which establishes the theoretical basis of shallow needling, and Jiuzhen Shier Yuan, Guan Zhen, Xiao Zhen Jie, and Zhong Shi from Lingshu (灵枢 ), which discuss instruments, techniques, and clinical applications. Since discussions of shallow needling are scattered throughout Neijing, chapters with fewer mentions were excluded. The study uses two complete English translations of Neijing: the version by Prof. Li Zhaoguo from Shanghai University of Traditional Chinese Medicine (Li, 2005; 2008), hereafter referred to as “Li’s Version”, and the version by Chinese-American acupuncturists Wu Liansheng and Wu Qi (Wu, 1997), hereafter “Wu’s Version”. The total word count of the extracted chapters is 15,156 in the Li Version and 12,556 in the Wu Version. 3.1.2 Research Questions This study aims to examine similarities and differences in vocabulary, syntax, and discourse cohesion between the two translations, explore the reasons behind their differences, and summarize implications for future English translations and international communication of shallow needling. Specifically, the following questions are addressed: 1) What are the similarities and differences between the two translations in terms of lexical, syntactic, and discourse levels? 2) What are the causes of these translation differences? 3) What insights can these differences provide for the future translation and global dissemination of shallow needling? 3.2 Research Tools and Indicators 3.2.1 Research Tools ABBYY FineReader 12 was used to convert Neijing into UTF-8 formatted TXT files. Relevant content www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 54 was cleaned using EmEditor. The corpus was analyzed using WordSmith 4.0 and AntConc 3.4.4w for lexical, syntactic, and discourse-level features. SPSS 26 was used for quantitative analysis. For indicators that met the assumption of normality (e.g., type-token ratio, mean sentence length, high-frequency conjunctions), independent samples t-tests were conducted. Non-parametric tests were used for variables not meeting normality. Ridit analysis was applied to compare word length distributions. A significance level of P < 0.05 was used. 3.2.2 Research Indicators Based on the framework by Lv Pengfei et al. (2021), this study selects the following indicators: 1) Lexical level: type-token ratio, lexical density, average word length, high-frequency words, and culturally loaded terms 2) Syntactic level: mean sentence length, sentence length standard deviation 3) Discourse level: conjunction usage, coherence, and cohesion in the text 4 Statistical Results and Analysis of Causes 4.1 Lexical Level The lexical features of the translations are analyzed from the perspectives of Standardized Type-Token Ratio (STTR), lexical density, average word length, high-frequency words, and culturally loaded terms. 4.1.1 Standardized Type-Token Ratio (STTR) and Lexical Density In corpus translation studies, a type refers to a unique word, while a token refers to every instance of a word, including repetitions. The ratio of types to tokens (TTR) reflects lexical richness or lexical density (Zhang, 2024). Generally, a higher TTR indicates a more diverse vocabulary, whereas a lower TTR reflects limited lexical variety. Lexical density refers to the proportion of content words (nouns, verbs, adjectives, adverbs) within a text. A higher lexical density suggests more information-rich content, which may enhance the accuracy of conveying the original meaning but might also result in verbosity. Conversely, a lower lexical density may imply a more concise translation but could risk omitting implicit meaning embedded in the classical Chinese language in Neijing. However, when the corpus is long, an increase in function words can significantly affect the TTR, making it less reliable for evaluating lexical richness. Therefore, Scott (2001) proposed the Standardized Type-Token Ratio (STTR), which calculates the TTR at fixed intervals (usually every 1,000 words) and then averages the values, thereby minimizing the impact of corpus length and improving the accuracy of lexical density assessment. Table 1. STTR Comparison between the Two Translations Li’s Version Wu’s Version Types 1728 1557 Tokens 15156 12556 www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 55 TTR 11 12 STTR 30.27 28.26* Note. Compared with the Li version, P > 0.05. As shown in Table 1, the number of tokens in Li’s version is considerably higher than that in Wu’s version. The primary reason lies in the translation strategies adopted by the two versions. Li frequently employs a combination of transliteration and literal translation for traditional Chinese medicine (TCM) terms, followed by explanatory notes appended to the end of each chapter. This approach leads to a higher number of tokens. In contrast, Wu does not provide any annotations and tends to adopt a more liberal translation strategy. For example, the term Fuci (浮刺 ), referring to a technique within the shallow needling method, is translated in Wu’s version as “superficial-puncture,” using a free translation strategy. However, in Li’s version, it appears as “Fuci (floating needling),” accompanied by a footnote: “Fuci (浮刺): It refers to a kind of oblique and shallow needling.” This not only adds transliterated terms but also lengthens the translated text through additional explanations. Figure 1. STTR Distribution across Five Chapters in the Two Translations The STTR for the five chapters in Li's version is 30.27, with an average STTR of 29.45. In Wu's version, the average STTR across the five chapters is 28.26, with an overall average STTR of 27.45. An independent samples t-test was conducted using SPSS 26, and the results indicated no statistically significant difference in lexical richness between the two translations (t = 0.971, P > 0.05). This suggests that the two versions do not differ significantly in terms of vocabulary quantity or lexical diversity. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 56 4.1.2 Average Word Length and Its Distribution Another way to examine lexical features is through the average word length, which reflects the average number of letters per word in the corpus. This metric provides insight into the translator's lexical choices and stylistic preferences. In this study, WordSmith 4.0 was used to calculate the average word length and the distribution of words of various lengths in both translations. A bar chart was generated to illustrate the distribution of word lengths and to highlight the lexical differences between Wu's version and Li's version. Figure 2. Word Length Distribution in the Translations The corpus data show that the average word length in Li’s version is 4.53 with a standard deviation of 2.54, while Wu’s version has an average word length of 4.56 and a standard deviation of 2.38. As illustrated in Figure 2, the majority of words in both translations consist of 2 to 4 letters. In Li’s version, two-letter words are the most frequent, accounting for 21.02% of the total, whereas three-letter words dominate in Wu’s version, making up 25.89%. Words with fewer than 8 letters constitute over 91% of all words in both versions. A Ridit analysis conducted via SPSS 26 on the raw word-length data revealed a statistically significant difference between the two versions (r = 0.51, t = 5.24, P < 0.01), indicating a considerable disparity in word length. Specifically, Wu’s version contains longer and more complex words compared to Li’s version. This variation is primarily attributed to differences in word choice and stylistic preferences between the translators. 4.1.3 High-Frequency Words and Culturally Loaded Terms 1) High-Frequency Words High-frequency words can reflect a translator’s linguistic tendencies and stylistic preferences. In this study, AntConc 3.4.4w was used to extract the top 50 high-frequency words in each translation. The www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 57 statistical results are presented as follows: Table 2. Comparison of Top 50 High-Frequency Words in the Two Translations No. Li’s Version Wu’s Version Words Frequency Percentage Words Frequency Percentage 1 the 1544 10.19% the 1876 14.94% 2 of 614 4.05% of 504 4.01% 3 and 495 3.27% and 494 3.93% 4 to 472 3.11% is 390 3.11% 5 is 431 2.84% to 259 2.06% 6 in 275 1.81% when 259 2.06% 7 qi 259 1.71% energy 242 1.93% 8 that 217 1.43% be 201 1.60% 9 be 199 1.31% in 188 1.50% 10 needling 180 1.19% it 170 1.35% 11 it 160 1.06% channel 169 1.35% 12 needle 143 0.94% will 127 1.01% 13 are 122 0.80% needle 123 0.98% 14 means 122 0.80% pricking 123 0.98% 15 channel 109 0.72% called 118 0.94% 16 a 100 0.66% are 104 0.83% 17 if 98 0.65% disease 103 0.82% 18 when 97 0.64% a 102 0.81% 19 channels 91 0.60% evil 96 0.76% 20 yang 88 0.58% one 95 0.76% 21 yin 88 0.58% should 95 0.76% 22 disease 86 0.57% can 94 0.75% 23 should 85 0.56% so 93 0.74% 24 can 82 0.54% patient 92 0.73% 25 by 79 0.52% with 90 0.72% 26 which 74 0.49% not 88 0.70% 27 called 73 0.48% which 78 0.62% 28 acupoints 71 0.47% yang 71 0.57% www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 58 29 on 69 0.46% yin 67 0.53% 30 as 63 0.42% means 66 0.53% 31 will 60 0.40% on 62 0.49% 32 with 60 0.40% prick 62 0.49% 33 used 58 0.38% as 60 0.48% 34 treat 57 0.38% must 57 0.45% 35 evil 56 0.37% skin 50 0.40% 36 one 54 0.36% foot 47 0.37% 37 into 52 0.34% if 47 0.37% 38 pulse 51 0.34% pulse 46 0.37% 39 reducing 51 0.34% or 44 0.35% 40 not 50 0.33% kind 42 0.33% 41 part 50 0.33% he 39 0.31% 42 body 49 0.32% pricked 38 0.30% 43 techniques 49 0.32% collaterals 37 0.29% 44 pathogenic 47 0.31% five 37 0.29% 45 located 46 0.30% purging 37 0.29% 46 idea 45 0.30% acupoints 36 0.29% 47 or 45 0.30% channels 36 0.29% 48 organs 45 0.30% has 35 0.28% 49 reinforcing 45 0.30% treat 35 0.28% 50 this 44 0.29% large 34 0.27% According to the data in Table 2, the top five high-frequency words in both translations are “the,” “and,” “of,” “is,” and “to.” Words such as “of,” “and,” “if,” “when,” “to,” “in,” “on,” “which,” “or,” “as,” and “with” are present in both versions, while “that,” “by,” and “into” are unique to Li’s version, and “so” appears only in Wu’s version. These words primarily serve to convey various grammatical and logical relations—such as coordination, possession, condition, location, and accompaniment. Overall, both translations demonstrate an emphasis on conjunctions and prepositions to clearly represent relationships within the text. Both versions frequently use the pronoun “it” to refer to previously mentioned content, thus avoiding redundancy. The term “经络 ” is consistently translated as “channels and collaterals,” “阴阳 ” is transliterated as “yin and yang,” and “脉” is rendered as “pulse” in both translations. The modal verb “will” is also found among the top 50 high-frequency words in both versions and is mainly used to www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 59 express future possibility or hypothetical scenarios. Wu’s version uses it more frequently, often in conjunction with subordinate clauses introduced by “when” or “if.” The pronoun “one” appears in both translations, mainly functioning anaphorically and occasionally as a numeral. Wu’s version uses it more often, typically positioning “one” as the subject of a sentence, occasionally replaced with “patient,” and frequently referring to “one” using the pronoun “he.” The word “evil” appears in both translations as the equivalent of the traditional Chinese concept “邪”. Wu’s version adopts a direct translation—“evil”—while Li’s version uses a hybrid transliteration and translation, such as “Xie (Evil),” and sometimes opts for the term “pathogenic.” Consequently, “pathogenic” is among the top 50 high-frequency words only in Li’s version. “Acupoints” is also included in the top 50 in both translations to represent “穴位”, but appears less frequently in Wu’s version, which also uses the more general term “point.” “Energy” in Wu’s version and “qi” in Li’s version are tied for seventh place, illustrating a divergence in their lexical choices for the term “气”. Wu translates terms like “邪气” and “病气” as “evil energy” or “pathogenic energy,” whereas Li uses the transliteration “qi.” The author argues that “energy” does not fully convey the inherent cultural and conceptual richness of “气 ,” and that Li’s use of “qi” better preserves the cultural uniqueness of TCM. The word “that” only appears in Li’s top 50 list, where it primarily functions as a clause introducer—e.g., in predicative or attributive clauses—and occasionally serves as a pronoun. The relatively high frequency of “by” and “used” in Li’s version is due to a preference for passive constructions such as “be treated by” or “be used to,” which contribute to a more objective tone. The preposition “into” is frequently used in Li’s version with verbs like “get” or “insert” to describe either the internal invasion of pathogenic factors or the insertion of needles. Words like “part” and “body” appear more frequently in Li’s version, typically in expressions such as “part of the body.” The word “located” is often used to describe the position of acupoints or pathological conditions. “Idea” is used to enhance logical flow or clarity, as in “[The idea that]…” “Organs” is employed by Li to translate “脏腑 ”, as in “Five-Zang Organs,” while Wu alternates between “organ” and “viscera.” The word “techniques” appears only in Li’s high-frequency list, reflecting his translation strategy for reinforcing and reducing methods, commonly rendered as “reinforcing techniques” and “reducing techniques.” This emphasizes the procedural nature of these methods. Wu, on the other hand, uses “invigorating” and “purging” and rarely translates “techniques” directly. The word “so” is found only in Wu’s top 50 list. It is often used in conjunction with “called” to form adjectival phrases such as “The so-called ‘easy to relate’…” or as a conjunction introducing adverbial clauses. Both versions use “called” to introduce acupuncture instruments or acupoints. Wu also uses “kind” to categorize types of instruments, and “large” to express magnitudes, such as in “four times as large as,” whereas Li prefers the term “great.” The verb “pricked” appears frequently in Wu’s version as a translation of “刺”, often combined with www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 60 modal verbs like “should,” “can,” or “must” (e.g., “should be pricked”). In translating the term “针刺”, Li’s version favors “needling,” while Wu prefers “pricking.” For instance, “浅刺 ” is translated as “shallow pricking” in Wu’s version, and “shallow needling” in Li’s. The author contends that Wu’s use of “pricking” merely conveys the action of piercing the skin with a sharp object and lacks specificity. In contrast, Li’s use of “needling” makes the use of a needle explicit, which facilitates international understanding and dissemination of the shallow needling technique. 2) Culturally Loaded Terms Culturally loaded terms refer to words, phrases, or idioms that represent concepts unique to a specific culture. These terms encapsulate practices, beliefs, and values that have been shaped over long periods and are distinct from those of other cultures (Nida, 1964). This study, using the original chapters on shallow needling from the Neijing as a reference, identifies several culturally loaded terms related to shallow needling. With the assistance of the corpus tool AntConc 3.4.4w, the following keyword list was compiled: Table 3. Translation of Culturally Loaded Terms Li’s Version Wu’s Version 络刺 (Luoci) Luoci blood-letting-needling 大泻刺 (Daxieci) Daxieci drainage-needling 毛刺 (Maoci) Maoci skin-needling 浮刺 (Fuci) Fuci (floating needling) superficial-puncture 直针刺 (Zhizhenci) Zhizhenci (direct needling) perpendicular-needling 锋针 (Fengzhen) Fengzhen (sharp needle) ensiform needle 井 (Jing) Jing - Well Jing (well) 荥 (Ying) Ying - Spring Xing From Table 3, it can be seen that Li’s version tends to adopt a foreignization strategy, frequently employing transliteration or a combination of transliteration and literal translation, supplemented by annotations to explain the culturally loaded terms. In contrast, Wu’s version is more inclined toward domestication, typically using a single translation strategy—either free translation or transliteration—without providing further cultural explanations. This divergence is mainly related to the translators’ purposes. Li’s version places more emphasis on preserving the philosophical and cultural ideas of Neijing, aiming to promote traditional Chinese medical culture. Wu’s version, however, was originally created to provide a complete English translation for international learners of Chinese medicine, thus placing more focus on the medical content’s comprehensibility for foreign readers (Li, 2020). In terms of culturally loaded terms, the foreignization strategy is more conducive to the international www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 61 dissemination of Chinese culture. Although this strategy may affect comprehension to some extent, such limitations can be compensated for through the use of annotations (Chen & Wang, 2022). Therefore, Li’s version can be seen as striking a balance between understandability and cultural transmission. 4.2 Syntactic Level At the syntactic level, stylistic features are analyzed through average sentence length and sentence length standard deviation. Average sentence length is calculated by dividing the total number of tokens by the number of sentence-final punctuation marks, such as periods, question marks, and exclamation points. A longer average sentence length indicates more complex sentence structures and higher syntactic maturity, whereas a shorter one reflects simpler structures. The sentence length standard deviation reveals the variation in sentence lengths throughout the text. A larger standard deviation indicates greater fluctuation in sentence length, while a smaller one suggests more consistency and stability in sentence structure. Table 4. Syntactic Features of the Two Translations 李本 吴本 The Number of Sentences 658 346 Sentence Length（ x ±s） ） 23±15.28 36±26.24* Note. Compared to Li’s version, *P<0.05. As shown in Table 4, the average sentence length in Wu’s version is 13 words longer than that in Li’s version, indicating a substantial difference. An independent-samples t-test conducted using SPSS 26 on the average sentence lengths across each chapter revealed a significant difference between the two translations (t = -3.95, P < 0.01). This suggests that Wu’s version generally favors longer sentences, reflecting greater syntactic maturity and complexity. Additionally, Wu’s version also shows a higher standard deviation in sentence length, indicating a wider variation across sentences. In contrast, Li’s version exhibits more uniform sentence lengths, demonstrating structural balance that to some extent maintains the format of the original text. This difference again correlates with the translators’ respective purposes. For instance, in the chapter Xiao Zhen Jie, the following sentence illustrates this stylistic divergence. Original Text:“针太深则邪气反沉”者，言浅浮之病，不欲深刺也，深则邪气从之入，故曰反沉也。 “皮肉筋脉各有所处”者，言经络各有所主也。 Li’s Version: [The idea that to use] deep needling [to treat the disease that is superficial] leads to internal invasion of Xieqi (Evil - Qi) means that mild disease should not be [treated by] deep needling lest Xieqi (Evil - Qi) deepen [its invasion into the body] with the thrusting [of the needle]. That is why it is said that [deep needling may] worsen the disease. [The idea that] the locations of [the diseases in] www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 62 the skin, muscles, sinews and vessels are different means that [the skin, muscles, sinews and vessels] are controlled respectively by Channels. Wu’s Version: The so called “when the pricking is too deep, the evil energy will, on the contrary, become sunken” means to the disease of which its evil energy is on a shallow position, do not prick deeply, otherwise, it will, cause the evil energy to penetrate inside, so, it is on the contrary to become sunken. The so called “the skin, muscle, tendon and channel are having their regular position” means the skin, muscle, tendon are on their regular positions respectively, and there are the definite positions in the channels and collaterals for treating the various diseases. In this example, Li’s version uses round brackets to provide current translations or explanations immediately after transliterated terms or culturally specific concepts, while square brackets are employed to differentiate annotations from the main text. This approach aligns with Li’s pre-translation principle: “Translate classical texts in a classical style, with no modern adornments.” thereby preserving the structural integrity of the original. In contrast, Wu’s version employs more conjunctions, resulting in longer sentences that convey more detailed information. For instance, the phrase “for treating the various diseases” in Wu’s version makes explicit a meaning that is only implied in the original text, thereby reflecting the translator’s intention to convey medical experience and instructional purpose (Chen, 2021). Wu’s translation does not follow the original sentence structure and places greater emphasis on the transmission of information. 4.3 Discourse Level Cohesion is a semantic concept achieved through grammatical and lexical means, allowing sentences to be logically connected. Coherence, also a semantic concept, refers to the resulting semantic relationships established by cohesive ties. Both cohesion and coherence are crucial in forming a comprehensible discourse. English, being a hypotactic language, heavily relies on explicit cohesive devices such as conjunctions. In contrast, Neijing, as a classical Chinese text, features parataxis with a high degree of implicitness. Thus, when translating such texts into English, cohesive markers like conjunctions are often necessary to articulate semantic relationships between clauses and sentences. At the discourse level, this study analyzes the coherence and cohesion of the two versions by examining the use of conjunctions. Using the Wordlist function in AntConc 3.4.4w, the five most frequently used conjunctions in each version were identified and analyzed. Table 5. Top Five Conjunctions in Each Translation Li’s Version Wu’s Version Frequency Percentage (%) Frequency Percentage (%) and 495 3.27% 494 3.93% when 97 0.64% 259* 2.06% if 98 0.65% 47 0.37% www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 63 or 45 0.30% 44 0.35% but 22 0.15% 16 0.13% 总计 757 4.99% 860 6.85% Note. *P < 0.05 compared to Li’s frequency of "when". A non-parametric test using SPSS 26 showed no statistically significant differences between the two datasets (z = -0.104, P > 0.05), suggesting that both translators emphasized the use of conjunctions to convey semantic relationships in the source text, including coordination, contrast, time, and condition. However, an independent sample t-test on the conjunction “when” revealed a significant difference between the two versions (t = -4.511, P < 0.05), with Wu’s version using it more frequently. Wu commonly uses “when” to introduce subordinate clauses, while Li tends to use “if” or omit the conjunction altogether. This reflects differences in stylistic preference and expression. For instance, in the chapter Pibu (The Skin Regions), the following original sentence is translated differently. Original Text: 邪中之则腠理开，开则入客于络脉，留而不去，传入于经，留而不去，传入于府， 廪于肠胃……。 Li’s Version: Attacked by Xie (Evil), the Couli (muscular interstice) opens, giving rise to the invasion of Xie (Evil) into the Collaterals. [If it] lingers [in the Collaterals], it enters the Channel; [if it] lingers [in the Channels], it enters the Fu-Organs and accumulates in the intestines and stomach... Wu’s Version：When the evil attacks the skin, the striae will be kept openned and the evil-energy will invade the collaterals, when the evil retains long, it will be transmited into the channel inside, when it retains in the channel long, it will be transmited into the intestine and stomach… From the above, we see that in translating “邪中之则腠理开，开则入客于络脉,” Wu’s version uses "when" to introduce adverbial clauses, while Li’s version omits conjunctions. For the subsequent sentences, both versions use different conjunctions to express conditionality, but Wu prefers "when" whereas Li prefers "if." This accounts for the higher frequency of "when" in Wu’s translation. 5 Conclusion and Implications The Neijing contains extensive discussions on shallow needling, and analyzing the translation features and differences in the two English versions provides valuable insights for future translation practices. This study finds that both translations share common features: there are no significant differences in lexical density, vocabulary size, and richness; at the discourse level, both emphasize the use of conjunctions. The differences lie in several areas: at the lexical level, Li often employs a foreignization strategy using transliteration with literal translation and annotation, preserving cultural nuances. Wu, on the other hand, typically favors free translation, leaning toward domestication. At the sentence level, Li’s version has lower average sentence length and smaller standard deviation, resulting in more www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 64 balanced sentences that better preserve the original structure. At the discourse level, Wu frequently uses "when" to introduce clauses, while Li prefers "if" to express conditionality. These differences are primarily due to the translators’ goals, strategies, and stylistic preferences. Based on the findings, several implications emerge: First, due to inherent conceptual differences between Chinese and Western medicine and cultural thinking, translating traditional Chinese medical terminology remains a challenge. Li’s strategy of combining transliteration, literal translation, and annotations better preserves cultural meaning and facilitates cultural transmission. Wu’s free translation approach may sometimes lead to mismatches—for instance, equating "Qi" with "energy"—which can cause conceptual inaccuracies. Second, the proper application of shallow needling is highly conditional, requiring precise expression of logical relationships between clauses in translation. Differences between parataxis in Chinese and hypotaxis in English necessitate explicit cohesion in English translations. The use of conjunctions by both translators enhances coherence and provides useful models for translating medical classical texts and classical Chinese literature more broadly. Lastly, regarding translation purpose and methods, Li focuses on preserving the cultural, philosophical, and structural aspects of Neijing, using foreignization strategies to appeal to scholars interested in the cultural and philosophical foundations of shallow needling. Wu emphasizes the transmission of medical knowledge, adopting domestication strategies to enhance readability, aimed at learners seeking technical understanding. Therefore, Li’s approach is more suitable for cultural dissemination, while Wu’s emphasis on explicating medical concepts also offers important insights for specialized medical translation. References Baker, M. (1993). Corpus linguistics and translation studies: Implications and application. In M. Baker, G. Francis, & E. Tognini-Bonelli (Eds.), Text and Technology: In Honour of John Sinclair (pp. 233-252). Amsterdam & Philadelphia: John Benjamins. https://doi.org/10.1075/z.64.15bak Cong, L. Y., Li, D., & Fu, Y. (2022). Shallow needling on skin for 30 cases of functional dyspepsia with spleen deficiency and qi stagnation. Chinese Acupuncture & Moxibustion, 42(2), 143-144. Gong, T. T., & Wang, C. (2023). A review of mechanism and modern clinical application of shallow acupuncture therapy. Clinical Journal of Traditional Chinese Medicine, 35(1), 184-188. He, N. N., Xu, J. Y., Li, P. P., et al. (2018). Corpus-based study on translation of disease terms of brain system in Huangdi Neijing. China Journal of Traditional Chinese Medicine and Pharmacy, 33(5), 1958-1961. Li, D. H., Ren, Y. L., Liang, H. et al. (2022). Professor Liang Fanrong′s experience in treating peripheral facial paralysis by phases based on the muscle meridian theory. World Chinese Medicine, 17(18), 2632-2634, 2639. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 65 Li, D., Cong, L. Y., & Fu, Y. (2021). Clinical observation on treatment of paraesthesia pharyngis with shallow thorn acupuncture method. China Journal of Traditional Chinese Medicine and Pharmacy, 36(12), 7479-7482. Li, P., Meng, X. J., Cheng, W. J. et al. (2022). Professor LU Fan’s clinical experience in shallow needling technique of acupuncture. Chinese Acupuncture & Moxibustion, 42(2), 194-196. Li, Z. G. (Trans.), & Liu, X. R. (Rev.). (2005). Yellow emperor’s canon of medicine: Plain Conversation. Xi’an: World Publishing Corporation. Li, Z. G. (Trans.), & Liu, X. R. (Rev.). (2008). Yellow emperor’s canon of medicine: Spiritual Pivot. Xi’an: World Publishing Corporation. Liao, Q. Y. (2000). Explorations in contemporary Western translation theory. Nanjing: Yilin Press. Liu, C. M. (2016). Promoting English translation of TCM by self-compiled bilingual parallel corpus of TCM classics. Journal of Basic Chinese Medicine, 22(11), 1545-1547. Lu, D. C., Qu, K., & Chen, Z. (2024). Analysis of “Qi” in Veith’s translation of Huangdi Neijing Suwen using corpus analysis. Journal of Basic Chinese Medicine, 30(8), 1399-1403. Mo, X. Y., & Xu, Z. H. (2019). Acupuncture depth in Huangdi Neijing. China Journal of Traditional Chinese Medicine and Pharmacy, 34(12), 5842-5845. Pan, H. O. (2022). Corpus study on Yin- Yang Yingxiang of Suwen. Journal of Basic Chinese Medicine, 28(4), 630-634, 666. Qiu, H. R., & Yao, X. (2017). Translator- centered translation in the English versions of Huangdi Neijing. Journal of Basic Chinese Medicine, 23(3), 415-417+420. Qiu, H. R., & Yao, X. (2018). Translation comparison of treatment policies in Huangdi Neijing. Chinese Journal of Integrated Traditional Chinese and Western Medicine, 38(10), 1263-1267. Scott, M. (2004). Oxford WordSmith Tools 4.0 Manual. Oxford: Oxford University Press. Su, L., & Zhou, E. (2023). English translation of metaphoric disease names in Huangdi Neijing: Current status and countermeasures. Journal of Basic Chinese Medicine, 29(9), 1535-1539. Ure, J. (1971). Lexical density and register differentiation. In G. Perren & J. L. M. Trim (Eds.), Applications of Linguistics (pp. 443-452). Cambridge: Cambridge University Press. Wang, S. B. (2022). On “sugar- needling” comfort acupuncture. Guiding Journal of Traditional Chinese Medicine and Pharmacy, 28(1), 1-5. Wu, G. Q., Wang, Z. D., Zhao, Y. B. et al. (2024). Progress in dose- effect relationship of acupuncture. Global Traditional Chinese Medicine, 17(1), 166-172. Wu, L. S., & Wu, Q. (Trans.). (1997). Yellow Emperor’s Canon Internal Medicine. Beijing: China Science and Technology Press. Yang, H. Z. (1993). Corpus linguistics and machine translation. Journal of Shanghai Jiao Tong University (Philosophy and Social Sciences), (1), 98-111. Yang, M., & Bai, N. (2010). A survey of corpus- based translation studies in China, 1993-2009. Chinese Translators Journal, 31(6), 46-50. www.scholink.org/ojs/index.php/eltls English Language Teaching and Linguistics Studies Vol. 7, No. 4, 2025 Published by SCHOLINK INC. 66 Ye, X., & Dong, M. H. (2015). On translating pulse diagnostic terms in Huangdi Neijing. Journal of Basic Chinese Medicine, (1), 94-96. Yu, W. Y., Ma, L. X., & Song, Y. et al. (2021). Significance of shallow needling in treatment of pain. Journal of Clinical Acupuncture and Moxibustion, 37(5), 93-97. Zheng, D. H. (2016). Translation of culture- bound terms in China’s cultural export. Shanghai Journal of Translators, (2), 53-56. Zhu, J. F. (2015). Corpus-based translation studies on Yellow Emperor's Canon of Medicine. Journal of Basic Chinese Medicine, 21(9), 1161-1164. A Corpus-based Study on the Translation of Shallow Zhiqiong Ye1*, Liangbing Wu2, Yingzhu Chen2, & Qua 1 Guangzhou Songtan Polytechnic College, Guangzhou Abstract Objectives This study aims to analyze chapters rel Keywords