A Comparative Study of the Translator’s Style
—— A Corpus-based Case Study of Lianghuiwang〔*〕

Zheng Xinmin,Ning Qiang

(Institute of Linguisticsl Shanghai International Studies University, Shanghai 200081)

Ⅰ.Introduction

Mona Baker (1993) is the first researcher who advocated to apply corpus to translation studies.In recent years,researchers began to employ the corpus and the corpus analysis tool to research the translator’s style.Hermans (1996) holds that “that other voice (i.e.the translator’s) is there in the text itself,in every word of it”.Mona Baker (2000) defines translator’s style as “a kind of thumbprint that is expressed in a range of linguistic — as well as non-linguistic — features”.Zhang Meifang (2002) stresses that “based on the corpus,a description,analysis,comparison and interpretation of the intangible and inconspicuous linguistic habits can convincingly illustrate the existence of the translators’ traces.” Hu Kaibao (2011) defines the translator’s style as translator’s preference in language use,or the recurrent linguistic pattern in the target text.

In the present study,Wordsmith 6.0 and Brown corpus were employed to make a quantitative analysis of the linguistic features of the two English versions of Lianghuiwang — the first chapter of Mencius.Irene Bloom’s version (2009) and Wu Guozhen’s (2015) version were selected for the present study.The two translators represent the nonnative and native translators of the traditional Chinese classics and their different language and cultural background,professional practice and language proficiency in both the source and the target language are considered to have left traces in their translation.Therefore,the findings of the comparative study are supposed to be able to reveal their translators’ style to a certain extent.

Ⅱ.Comparison at the lexical level

The analysis at the lexical level generally focuses on tokens,types,type/token ratio,word frequency,and word length etc.Different translators hold their own translation beliefs and principles,and their word choice in the translation is different,which composes a very important aspect of the translator’s style.

1.Type/Token Ratio

Token refers to the total number of words in a text and type is the number of different word forms in a text.Type/Token Ratio (TTR.),which is the number of types divided by the number of tokens,is a measure of vocabulary variation within a written text.A high TTR means that the translator uses a wider range of words while a low TTR indicates that the translator’s word choice is restricted.It is recommended that standardized type/token ratio figures be adopted when the texts used are of different lengths (Baker,2000).The relevant figures for the two versions are listed in Table 1.

Table1Type/TokenRatiocomparison

Example 1

ST.孟子見梁襄王，出，語人曰：“望之不似人君，就之而不見所畏焉?！?/p>

Bloom:Mencius saw King Xiang of Liang.On emerging he said to someone,“Seeing him from a distance,he does not appear to be a ruler of men;approaching him,one sees nothing imposing about him.” (35 words)

Wu:Mencius had an audience with King Xiang of Liang.When he came out,he said to others,“When I saw him at a distance,he did not look like a king to me.When I approached him,I found nothing in him to be in awe of.” (47 words)

Example 2

ST:王見之，曰：‘牛何之？’

Bloom：On seeing this,the king asked where the ox was going.

Wu:You saw this and asked,“What are you going to do with the bull?”

The above examples illustrate that Bloom follows the style of being concise and formal of the source text and tries not to use conjunctions as cohesive device.Furthermore,he even translates literally with the ox as the subject of the sentence,which actually is of the absolutive case in Chinese.In comparison,Wu not only adopts cohesive conjunctions but also makes his translation more target-reader-friendly.In the second example,Wu treated the bull (ox in Bloom’s version) as the object and adds the hidden subject (you) to make the meaning of the sentence more explicit.

Table2Wordfrequencystatistics

2.Word frequency,Keywords and word length

Word frequency refers to the number of occurrence of the words in a given text.By comparing the word frequency of the text,the characteristics of the translator’s word choice can be further revealed.Table 2 shows the statistics of the word frequency of both versions.

Table 2 presents the top ten most frequent used words by the two translators.The top five,namely “the”,“to”,“and”,“of”,and “a”,though ranks a little bit different,are all function words.It corresponds with the top five high-frequency words of the TEC(Olohan,2004).In addition,out of ten top high-frequency words,eight words are identical,which demonstrates the similarities in the word choice between the two translators.

According to Feng Qinghua (2008),the high frequency of “the” and “of” generally proves the formality of the translated text.Apparently,Bloom’s version is more formal than Wu’s version,which again verifies the conclusion drawn from the TTR statistics.

What is also noteworthy is that there are two pronouns (you and your) in Wu’s version and two nouns (King and people) in Bloom’s version.A keyword analysis of both translation texts further reveals the translators’s word choice in the dealing with the theme of the source text.

Table3Keywordstatistics

Mu Lei(2011) believes that KeyWords of Wordsmith is an important program to investigate the content and linguistic features of both the source and the translation texts.Key words are obtained by comparing the word frequencies in a given text against their expected frequencies derived in a lager referential corpora through a statistical test (e.g.,loglinear or chisqured).Keyness refers to the context-specific quality of being key of a word or phrase.Though the frequency is different,most of the key words of both versions overlap.The key words (Mencius,king,people,etc.) of Table 3 vividly demonstrates that both versions reflect the content and theme of the source text faithfully,which is the dialogue between King and Mencius about how to govern a state.However,from the top 15 keyword list,it can be concluded that the theme of Wu’s version is more prominent in that it covers the “benevolent” and the “governance”.

Besides,as found in the high-frequency wordlist,the major difference between the two versions lies in the frequency of “king”,“you” and “one”.By employing the concordance program of wordsmith,the example context of these three words are displayed.See example 3 and 4.

Example 3

ST:“緣木求魚，雖不得魚，無后災。以若所為，求若所欲，盡心力而為之，后必有災?！?/p>

Bloom:“When one climbs a tree in search of fish,though one gets no fish,no disaster will ensue.But if one acts in the way you do in pursuit of ….”

Wu:“If you climb a tree to catch fish and get none,there will be no disastrous consequence.But if you exert all your energy to get what you desire in this way,you are sure to ….”

Example 4

ST:“王好戰(zhàn)……”

Wu:“Since you have interest in warfare…”

Bloom:“the king is fond of war…”

In the examples above,Wu uses “you” (the second person pronoun) while Bloom uses the indefinite pronoun (one) and noun (the king) to refer to King Hui of Liang respectively.The adoption of different pronouns and nouns is also an indicator of the different style of the translators.Wu’s second person pronoun closes the distance between Mencius and King Hui of Liang and is more dialogue-like while Bloom’s indefinite pronoun and nouns make Mencius’s words more preaching-like.

3.Word length and lexical density

Word Length is the number of letters used in a word.The length of words influences the difficulty of the text and therefore the reader’s reading experience accordingly.Table 4 presents the average word length of the two selected translated texts.

Table 4 shows that 3-letter words accounts for the highest percentage of both versions.The total percentage of 2-letter words to 7-letter words accounts for 87.51% (Wu’s version) and 88.02% (Bloom’s version) respectively,which indicates that both versions are more of a dialogue style with simple words.The word length difference is insignificant and the relatively high percentage of shorter word length of both versions indicates less reading difficulty.

Table4Wordlengthstatisticsofthemainwords

4.Lexical density and text difficulty

Another way to examine the readability is to calculate the lexical density of a text.The lexical density of a text can be obtained through the formula:Lexical density = (number of lexical words/total number of words) * 100.A high lexical density indicates more information-carrying words and a low lexical density fewer information-carrying words.Thus,lexical density can serve as a measure of the vocabulary richness of the translation text.See table 5 for the lexical density of both versions.

Though the lexical density of both versions are pretty close,the slight difference in the lexical words and pronouns proves the word choice feature of the two translators again.Bloom uses more nouns and verbs while Wu use relatively more pronouns,adjectives and adverbs.Bloom tends to follow more close to the source text and translate most of the “王” as “king” while Wu adopts “you” to represent “王” in the dialogue between Mencius and King Hui of Liang.Wu’s lexical density is higher than that of Bloom’s in that he uses more adjectives and adverbs which adds vividness of the translation text.

Table5Lexicaldensitystatistics

Table6Vocabularyprofiler-results

As for the translation text difficulty,it can be examined through the vocabulary profiler program,which presents the level distribution of words in the text.Table 6 shows the results of the two version’s vocabulary distribution percentage at different levels.

As seen from Table 6,about 88% vocabulary of Wu’s translation and about 89% vocabulary of Bloom’s translation belong to the most frequently-used 2,000 words.Thus,it can be safely concluded that the reading difficulty of both versions was lower and this reflects the feature of lexical simplification of the translated language.

Ⅲ.Comparison at the syntactic level

1.Sentence length

As far as the syntactic level is concerned,the average sentence length,the use of conjunctions and the reporting clauses are often compared and contrasted.Average sentence length is the number of words in a sentence,which reveals the translators’ personal habit of sentence making.Through the analysis of the sentence length,the translated text can be compared and contrasted.

Table7Sentencenumber&averagesentencelength

Table 7 shows that both the sentence number and the sentence length between the two versions are close.Both translator ors employ shorter sentences in their translation and make the text more reader friendly.Using shorter sentences to translate also mirrors the influence of the source text on the translated text in that the source classical Chinese text characterizes conciseness.

Example:

ST:仁者無敵。

Wu:The benevolent have no enemy.(5 words)

Bloom:The humane man has no enemy.(6 words)

Example:

ST:是以君子遠庖廚。

Wu:That’s why a superior man stays away from the kitchen.(11 words)

Bloom:And so the noble person stays far away from the kitchen.(11 words)

2.Use of conjunctions

Compared with the English,the Chinese uses fewer conjunctions.The logical relationship between clauses and sentences is usually hidden and expected to be understood through the word order,the context or the shared knowledge of the reader and the listener.But in English,generally,the logical relationship has to be shown by conjunctions.Therefore,to investigate the use of conjunction in the translation is an effective way to identify the translator tor’s sentence making choice.

As Table 8 shows,the percentage of conjunction use of Bloom’s version is higher than that of Wu’s version,while the difference is rather small.It can be said that both translators follow the linguistic rules of the target language,however,Bloom uses more complex sentences than Wu,which makes her translation more formal and increase the reading difficulty to a certain extent.

Table8Conjunctionfrequencydifference

3.Reporting clauses

The reporting clause refers to a clause which indicates that someone is talking about what someone said or thought.For example,in ‘She said that she was’ ‘She said’ is a reporting clause.Lianghuiwang,the first chapter of Mencius,is mainly the record of conversations between King Hui of Liang and Mencius.Therefore,it is full of direct speeches with reported and reporting clauses.The reporting clause refer to a clause which indicates that someone is talking about what someone said or thought.“Mencius said” is a reporting clause.The study of the main reporting verbs like “曰” (say) and “對曰” (reply) etc.can reveal the translator’s preference of sentence making at the syntactic level.Through the concord program of Wordsmith,the concordances of the reporting clauses were obtained.

Table9Positionofconcordancesofthereportingclauses

From Table 9,it can be seen that there is a big difference between the two translators in their use and positioning of the reporting clauses.Though,overall,the total number of reporting clauses of the two versions is close (Wu:90;Bloom:95),the positioning of them is quite different.Wu seems to have purposefully placed his reporting clauses evenly to avoid the repetition while Bloom seems to have a preference for the initial position.After verifying the source text,it is found that Bloom’s version is greatly influenced by the style of the source text which places constantly the reporting clause and reporting verbs at the initial position.Besides the even positioning of reporting clauses,Wu also employs different words to bring vivacity to his translation.

Example

ST:曰：“不可……”

Bloom:The king said,“That would not do…”

Wu:“No mocking,” he commented…

ST:曰：“王如知此……

Bloom:Mencius said,“If the king understands this…”

Wu:“you see the point of the metaphor,Your Majesty.” continued Mencius...

Ⅳ.Conclusion

Through the statistical analysis of the two translation texts from the linguistic perspective,it is found that Wu’s version is of higher TTR and thus a greater lexical variety than Bloom’s version,but the latter’s high frequency words indicates the feature of a relatively formal text.In terms of keywords,both version faithfully present the content and theme of the source text,but Wu’s version is more theme prominent.As for as the word length is concerned,both versions take the dialogue style with simple words,while Wu’s version demonstrates a higher lexical density with more adjectives and adverbs,which adds vividness of the translation text.However,Wu’s richer vocabulary also increases the reading difficulty of his text.At the syntactic level,though Wu uses longer sentences,his use of conjunctions is less than Bloom,whose version seems more target-language-like.In the respect of reporting clauses,Wu positions his reporting clauses evenly,while Bloom follows the source text closely.Overall,Bloom seems more source-text centered in his translation while Wu takes a more target-text oriented style.

Notes:

〔1〕穆雷：《翻譯研究方法概論》，外語教學與研究出版社，2011年，第153頁。

〔2〕張美芳：《利用語料庫調(diào)查譯者的問題——貝克研究新法評介》，《解放軍外國語學院學報》2002年第3期，第54-57頁。

〔3〕胡開寶：《語料庫翻譯學概論》，上海交通大學出版社，2011年，第109頁。

〔4〕馮慶華：《母語文化下的譯者風格》，上海外語教育出版社，2008年，第225頁。

〔5〕Baker,M.,Towards a methodology for investigating the style of a literary translator, Target,2000(2),pp.241-266.

〔6〕Hermans,Theo.The translator’s voice in translated narrative, Target,1996(1),pp.23-48.

〔7〕Baker,M.,Corpus linguistics and translation studies:implications and applications,In M.Baker,et al (eds.),Text and Technology,Amsterdam:Benjamins,1993,pp.233-252.

〔8〕吳國珍：《孟子》(最新英文全譯全注本)，福建教育出版社，2015年。

〔9〕Irene Bloom, Mencius, 2009,Columbia University Press.