论文部分内容阅读
【Abstract】Business English Correspondence （BEC） is a discourse in a written format for exchanging information in business activities. Nowadays this act of communication is conducted mainly by means of business letters， emails and faxes. The 7C’s principles are internationally acknowledged as the norm for writing BEC， namely “Clearness”， “Conciseness”， “Correctness”， “Consideration”， “Completeness”， “Concreteness” and “Courtesy”. To comply with the 7C’s principles， BEC has formed its own linguistic features， including nominalization. This paper makes a corpus-based analysis of nominalization in BEC and investigates the functions of nominalization in BEC.
【Key words】corpus-based analysis； nominalization； Business English Correspondence （BEC）； functions of nominalization
【作者簡介】He Xiangzhen， Nanjing Normal University Zhongbei College.
1. Definition and Linguistic Features of BEC
Regarding business correspondence， many Chinese books and articles are using the definition “a piece of conversation by post” given by Gan Hong （1979： 45）， who has written several professional BEC teaching materials. However， considering that conversations follow no strict principles while business correspondence is restricted by the 7C’s， Gan’s definition that equates business correspondence with conversations can arouse criticism.
In Register， Genre and Style， Biber and Conrad （2009： 182） have investigated the linguistic characteristics of emails based on a corpus of 76 emails with 15，840 tokens， which compares the frequency of lexical verbs， pronouns and nouns in emails， conversation and academic prose. According their study， emails are quite similar to academic prose in the overall use of nouns， which is much more frequent than in conversations.
Since the linguistic features may vary within a register， Biber and Conrad （2009： 187） have further distinguished among different types of emails to examine register variations. They classified all the emails in the corpus into three sub-categories： emails among friends and family on personal topics， emails among colleagues on professional topics， and emails among strangers on professional topics. Their conclusion is that with the decreasing degree of intimacy between participants， the frequency of nouns is increasing， making emails among colleagues and strangers more similar to standard written prose.
As business correspondence mostly takes place among colleagues and strangers， nouns can be more frequently used in BEC than in conversations. Thus it is inappropriate to define business correspondence as “a piece of conversation by post”. A more suitable definition may be “a piece of well-planned communication by post”. Biber and Conrad （2009： 188） then made a further investigation on nominalizations across the three email sub-registers. According to their findings， from emails among friends and family to those among colleagues and finally to those among strangers， with the increasing frequency of nouns， the frequency of nominalizations is also increasing in proportion.
Since most business correspondence is among colleagues and strangers， one assumption can be concluded – nominalization is a salient feature of BEC， deviating it from conversations. In the following part， analysis of nominalization will be made based on a BEC corpus to testify this assumption.
2. A Corpus-based Analysis of Nominalization in BEC
2.1 Methodology
For the analysis， a small BEC corpus of 150 business letters and emails has been established. There are totally 18，028 tokens. Except for several on-line samples， most of the texts are from a professional BEC textbook. Though the scale of this corpus is small， the texts in the corpus cover almost every topic of business activities， from the initial enquiry to the final shipment， so it can in some degree be considered as representative of business English correspondence. Therefore， the analysis results will be convincing and reliable.
After the corpus was established， the next step was identifying nominalizations. In the book Corpus Linguistics， Biber， Conrad and Reppen （2006： 59） provide the following definition – “Nominalizations are nouns that are related to verbs or adjectives morphologically； the nominalization is said to be ‘derived’ from the verb or adjective”. And adding suffixes to verbs and adjectives is the major way to form nominalizations. In light of previous studies on nominalization， the most frequent nominalizations are those with suffixes -ment， -tion/-sion， -ty， -ness， -ance/-ancy， -ence/-ency （including plural forms） and gerunds with a special suffix -ing. Since these two categories can already take up a large proportion of all nominalizations， only these nominalizations have been searched for.
It should be noted that not all tokens with the above suffixes are nominalizations. Non-nominalizations ending with the same suffixes need to be sorted out， such as attention， moment， quality， business， advance， agency， etc. For gerunds with the special suffix -ing， present participles need to be excluded， so this analysis has only searched for the three kinds of gerunds below.
（1） v-ing of （2） personal pronoun v-ing
（my/me/our/us/your/you/his/him/her/their/them）
（3） preposition v-ing
（by/on/upon/of/at/before/after/with/to/in/into）
2.2 Frequency of Nominalizations in the BEC Corpus
With the aid of the software AntConc， nominalizations of the above two categories were picked out and non-nominalized words with the same suffixes were manually deleted in the Concordance list.
As the result presents， there are totally 547 hits of nominalizations， taking up 3.03% of all tokens. And the suffix -ment is the most productive one in this corpus.
Besides quantitative analysis， a comparative study has also been conducted. The proportion of nominalizations in the BEC corpus has been compared with that in British National Corpus of Written English （BNCW） and British National Corpus of Spoken Language （BNCS）. The proportion of nominalizations in the BEC corpus （3.03%） is quite similar to that in BNCW （2.98%）， and is about three times that in BNCS （0.97%）. From the above analysis， a conclusion that “nominalizations are frequently used in BEC， deviating it from conversations” can be reached.
2.3 Pervasiveness of Nominalizations in the BEC Corpus
As Biber states in Register， Genre and Style （2006： 16）， the defining characteristics of a register lie in both frequent and pervasive distribution of the linguistic features. So apart from frequency， pervasiveness of the nominalizations in the corpus also needs analyzing.
As mentioned above， the most productive suffix in the BEC corpus is -ment. So the distribution of the nominalizations with -ment can to some extent represent that of all nominalizations. After all the nominalizations with -ment were extracted from each topic. Of the total 150 texts， nominalizations with the suffix -ment appear in more than half of them （84 texts）. And these nominalizations are distributed in every process in business activities.
As both frequency and pervasiveness of nominalizations have been examined， the assumption that “nominalization is a salient feature in BEC” can be fully testified. In the following part， this paper will analyze the underlying reasons for this phenomenon.
3. Functions of Nominalization in BEC
Nowadays， the 7C’s principles have become the norm for writing BEC， generating its linguistic features. Four major functions of nominalization can help to achieve the 7C’s principles， namely conciseness， formality， cohesion and objectivity， and that is why nominalizations are frequently and pervasively used in BEC. 3.1 Conciseness and Formality
A concise message can save time for both sides， so conciseness is one important principle among the 7C’s， which can be realized by nominalization. Halliday （1994） compares the nominalization process to “package”. By nominalization， one original simple clause can be “packed” into a nominal group or several clauses into one clause. The information loaded in the original two or more clauses can be equally conveyed by a condensed nominal group or one clause， which means the lexical density is raised. So nominalization can help to achieve the “Conciseness” principle to avoid clumsy and wordy expressions.
In Halliday’s study of nominalization in science and technical registers （1994）， he argues nominalizations can turn a discourse into a kind of “expert’s language”， with higher level of formality. Nominalizations in BEC actually have similar function. By nominalization， the spoken language is turned into abstract written language， which tends to have bigger lexical density. Halliday （1994） also argues that the degree of formality is in positive proportion to the lexical density. So by raising lexical density， nominalization can make business correspondence more formal and professional. The “Correctness” principle in 7C’s indicates not only correct use of grammar， spelling and punctuations， but also the right level of language. So the formality function of nominalization agrees with the “Correctness” principle.
3.2 Cohesion
As Halliday and Hasan puts in Cohesion in English （1976）， cohesion is “non-structural resource for discourse”. And “there are four ways by which cohesion is created in English： by reference， ellipsis， conjunction and lexical organization”. Though different in form， nominalizations share the same root and have the same semantic meaning as the original verbs or adjective. So nominalization can be seen as a repetition device for “lexical organization”. Below is one example from the corpus.
e.g.： We have examined the goods against order No.1563. It was found upon examination that nearly 20% of the packages were broken.
In this example， “examine” is nominalized into “examination”， which conveys the same semantic meaning. Through this kind of repetition， lexical cohesion of the two sentences has been fulfilled.
Besides repetition， cohesion can also be realized by using nominalizations to construct the “Theme-Rheme” pattern. A sentence can be divided into two parts： Theme and Rheme. Theme is “the starting point of the clause message”. Rheme is the remainder that develops the Theme （Halliday 1994： 38）. The nominalization process can develop one element of the Rheme in the previous sentence （R1） into the Theme of the following sentence （T2）. Here is one example taken from the corpus. e.g.： Therefore we have cabled you this afternoon， asking you to amend your L/C. The amendment should be made without delay.
In this example， the Theme of sentence 2 “amendment” is nominalized from the Rheme of sentence 1 “amend”. Via nominalization， the two sentences are tightly connected. This cohesion function of nominalization makes a text clearer and easier to follow， thus it complies with the “Clarity” principle.
3.3 Objectivity
Nominalization is also a powerful device to achieve objectivity. As put by Jespersen （1924）， nominalization makes a discourse more abstract by removing some animating elements of the verb. Nominalization from verbs can omit the subject or object of a motion， as well as indications of time or modality. Example A is from the corpus and I have rewritten it to its congruent form B.
e.g. A： Punctual execution of the order will guarantee enough time for trial assembly.
e.g. B： You should execute the order punctually to guarantee that we have enough time for trial assembly.
It can be seen from comparison that in the nominalization process， the executor “you” and the subjective mood indicated in the modal verb “should” are deleted， which creates an effect of objectivity. The impolite tone of sentence 5b is tactfully avoided and the request is made in a more courteous way by omitting the strong emotional factors through nominalization. Therefore in this case， objectivity function of nominalization is in conformity with the “Courtesy” principle.
4. Conclusion
This paper aimed to investigate nominalization in BEC. After proposing that nominalization is a salient feature in BEC， this paper made verification through quantitative and comparative analysis of a BEC corpus with AntConc. With representative samples， functions of nominalization as conciseness， formality， cohesion and objectivity were illustrated. Since all those functions conform to the 7C’s principles， this analysis can be applied in practice， to guide the writing of BEC and facilitate effective business communication.
However， this paper has some limitations. First， regarding the BEC corpus， most texts are from one textbook. Though it can offer some guidance， it cannot fully reflect the real business world. Secondly， regarding this analysis， it has just taken some most productive suffixes into consideration but neglected other suffixes like -age and -ship. Thirdly， AntConc cannot search for nominalizations from direct conversion， like “order” and “delay”. Finally， from the beginning to the final process of business， the intimacy among participants will change. So there can be register variations evolving with the rising intimacy. This point deserves exploring in further studies. References：
[1]Biber， D.， Conrad， S. Register， Genre， and Style[M]. Cambridge： Cambridge University Press，2009.
[2]Biber， D.， Conrad， S.
【Key words】corpus-based analysis； nominalization； Business English Correspondence （BEC）； functions of nominalization
【作者簡介】He Xiangzhen， Nanjing Normal University Zhongbei College.
1. Definition and Linguistic Features of BEC
Regarding business correspondence， many Chinese books and articles are using the definition “a piece of conversation by post” given by Gan Hong （1979： 45）， who has written several professional BEC teaching materials. However， considering that conversations follow no strict principles while business correspondence is restricted by the 7C’s， Gan’s definition that equates business correspondence with conversations can arouse criticism.
In Register， Genre and Style， Biber and Conrad （2009： 182） have investigated the linguistic characteristics of emails based on a corpus of 76 emails with 15，840 tokens， which compares the frequency of lexical verbs， pronouns and nouns in emails， conversation and academic prose. According their study， emails are quite similar to academic prose in the overall use of nouns， which is much more frequent than in conversations.
Since the linguistic features may vary within a register， Biber and Conrad （2009： 187） have further distinguished among different types of emails to examine register variations. They classified all the emails in the corpus into three sub-categories： emails among friends and family on personal topics， emails among colleagues on professional topics， and emails among strangers on professional topics. Their conclusion is that with the decreasing degree of intimacy between participants， the frequency of nouns is increasing， making emails among colleagues and strangers more similar to standard written prose.
As business correspondence mostly takes place among colleagues and strangers， nouns can be more frequently used in BEC than in conversations. Thus it is inappropriate to define business correspondence as “a piece of conversation by post”. A more suitable definition may be “a piece of well-planned communication by post”. Biber and Conrad （2009： 188） then made a further investigation on nominalizations across the three email sub-registers. According to their findings， from emails among friends and family to those among colleagues and finally to those among strangers， with the increasing frequency of nouns， the frequency of nominalizations is also increasing in proportion.
Since most business correspondence is among colleagues and strangers， one assumption can be concluded – nominalization is a salient feature of BEC， deviating it from conversations. In the following part， analysis of nominalization will be made based on a BEC corpus to testify this assumption.
2. A Corpus-based Analysis of Nominalization in BEC
2.1 Methodology
For the analysis， a small BEC corpus of 150 business letters and emails has been established. There are totally 18，028 tokens. Except for several on-line samples， most of the texts are from a professional BEC textbook. Though the scale of this corpus is small， the texts in the corpus cover almost every topic of business activities， from the initial enquiry to the final shipment， so it can in some degree be considered as representative of business English correspondence. Therefore， the analysis results will be convincing and reliable.
After the corpus was established， the next step was identifying nominalizations. In the book Corpus Linguistics， Biber， Conrad and Reppen （2006： 59） provide the following definition – “Nominalizations are nouns that are related to verbs or adjectives morphologically； the nominalization is said to be ‘derived’ from the verb or adjective”. And adding suffixes to verbs and adjectives is the major way to form nominalizations. In light of previous studies on nominalization， the most frequent nominalizations are those with suffixes -ment， -tion/-sion， -ty， -ness， -ance/-ancy， -ence/-ency （including plural forms） and gerunds with a special suffix -ing. Since these two categories can already take up a large proportion of all nominalizations， only these nominalizations have been searched for.
It should be noted that not all tokens with the above suffixes are nominalizations. Non-nominalizations ending with the same suffixes need to be sorted out， such as attention， moment， quality， business， advance， agency， etc. For gerunds with the special suffix -ing， present participles need to be excluded， so this analysis has only searched for the three kinds of gerunds below.
（1） v-ing of （2） personal pronoun v-ing
（my/me/our/us/your/you/his/him/her/their/them）
（3） preposition v-ing
（by/on/upon/of/at/before/after/with/to/in/into）
2.2 Frequency of Nominalizations in the BEC Corpus
With the aid of the software AntConc， nominalizations of the above two categories were picked out and non-nominalized words with the same suffixes were manually deleted in the Concordance list.
As the result presents， there are totally 547 hits of nominalizations， taking up 3.03% of all tokens. And the suffix -ment is the most productive one in this corpus.
Besides quantitative analysis， a comparative study has also been conducted. The proportion of nominalizations in the BEC corpus has been compared with that in British National Corpus of Written English （BNCW） and British National Corpus of Spoken Language （BNCS）. The proportion of nominalizations in the BEC corpus （3.03%） is quite similar to that in BNCW （2.98%）， and is about three times that in BNCS （0.97%）. From the above analysis， a conclusion that “nominalizations are frequently used in BEC， deviating it from conversations” can be reached.
2.3 Pervasiveness of Nominalizations in the BEC Corpus
As Biber states in Register， Genre and Style （2006： 16）， the defining characteristics of a register lie in both frequent and pervasive distribution of the linguistic features. So apart from frequency， pervasiveness of the nominalizations in the corpus also needs analyzing.
As mentioned above， the most productive suffix in the BEC corpus is -ment. So the distribution of the nominalizations with -ment can to some extent represent that of all nominalizations. After all the nominalizations with -ment were extracted from each topic. Of the total 150 texts， nominalizations with the suffix -ment appear in more than half of them （84 texts）. And these nominalizations are distributed in every process in business activities.
As both frequency and pervasiveness of nominalizations have been examined， the assumption that “nominalization is a salient feature in BEC” can be fully testified. In the following part， this paper will analyze the underlying reasons for this phenomenon.
3. Functions of Nominalization in BEC
Nowadays， the 7C’s principles have become the norm for writing BEC， generating its linguistic features. Four major functions of nominalization can help to achieve the 7C’s principles， namely conciseness， formality， cohesion and objectivity， and that is why nominalizations are frequently and pervasively used in BEC. 3.1 Conciseness and Formality
A concise message can save time for both sides， so conciseness is one important principle among the 7C’s， which can be realized by nominalization. Halliday （1994） compares the nominalization process to “package”. By nominalization， one original simple clause can be “packed” into a nominal group or several clauses into one clause. The information loaded in the original two or more clauses can be equally conveyed by a condensed nominal group or one clause， which means the lexical density is raised. So nominalization can help to achieve the “Conciseness” principle to avoid clumsy and wordy expressions.
In Halliday’s study of nominalization in science and technical registers （1994）， he argues nominalizations can turn a discourse into a kind of “expert’s language”， with higher level of formality. Nominalizations in BEC actually have similar function. By nominalization， the spoken language is turned into abstract written language， which tends to have bigger lexical density. Halliday （1994） also argues that the degree of formality is in positive proportion to the lexical density. So by raising lexical density， nominalization can make business correspondence more formal and professional. The “Correctness” principle in 7C’s indicates not only correct use of grammar， spelling and punctuations， but also the right level of language. So the formality function of nominalization agrees with the “Correctness” principle.
3.2 Cohesion
As Halliday and Hasan puts in Cohesion in English （1976）， cohesion is “non-structural resource for discourse”. And “there are four ways by which cohesion is created in English： by reference， ellipsis， conjunction and lexical organization”. Though different in form， nominalizations share the same root and have the same semantic meaning as the original verbs or adjective. So nominalization can be seen as a repetition device for “lexical organization”. Below is one example from the corpus.
e.g.： We have examined the goods against order No.1563. It was found upon examination that nearly 20% of the packages were broken.
In this example， “examine” is nominalized into “examination”， which conveys the same semantic meaning. Through this kind of repetition， lexical cohesion of the two sentences has been fulfilled.
Besides repetition， cohesion can also be realized by using nominalizations to construct the “Theme-Rheme” pattern. A sentence can be divided into two parts： Theme and Rheme. Theme is “the starting point of the clause message”. Rheme is the remainder that develops the Theme （Halliday 1994： 38）. The nominalization process can develop one element of the Rheme in the previous sentence （R1） into the Theme of the following sentence （T2）. Here is one example taken from the corpus. e.g.： Therefore we have cabled you this afternoon， asking you to amend your L/C. The amendment should be made without delay.
In this example， the Theme of sentence 2 “amendment” is nominalized from the Rheme of sentence 1 “amend”. Via nominalization， the two sentences are tightly connected. This cohesion function of nominalization makes a text clearer and easier to follow， thus it complies with the “Clarity” principle.
3.3 Objectivity
Nominalization is also a powerful device to achieve objectivity. As put by Jespersen （1924）， nominalization makes a discourse more abstract by removing some animating elements of the verb. Nominalization from verbs can omit the subject or object of a motion， as well as indications of time or modality. Example A is from the corpus and I have rewritten it to its congruent form B.
e.g. A： Punctual execution of the order will guarantee enough time for trial assembly.
e.g. B： You should execute the order punctually to guarantee that we have enough time for trial assembly.
It can be seen from comparison that in the nominalization process， the executor “you” and the subjective mood indicated in the modal verb “should” are deleted， which creates an effect of objectivity. The impolite tone of sentence 5b is tactfully avoided and the request is made in a more courteous way by omitting the strong emotional factors through nominalization. Therefore in this case， objectivity function of nominalization is in conformity with the “Courtesy” principle.
4. Conclusion
This paper aimed to investigate nominalization in BEC. After proposing that nominalization is a salient feature in BEC， this paper made verification through quantitative and comparative analysis of a BEC corpus with AntConc. With representative samples， functions of nominalization as conciseness， formality， cohesion and objectivity were illustrated. Since all those functions conform to the 7C’s principles， this analysis can be applied in practice， to guide the writing of BEC and facilitate effective business communication.
However， this paper has some limitations. First， regarding the BEC corpus， most texts are from one textbook. Though it can offer some guidance， it cannot fully reflect the real business world. Secondly， regarding this analysis， it has just taken some most productive suffixes into consideration but neglected other suffixes like -age and -ship. Thirdly， AntConc cannot search for nominalizations from direct conversion， like “order” and “delay”. Finally， from the beginning to the final process of business， the intimacy among participants will change. So there can be register variations evolving with the rising intimacy. This point deserves exploring in further studies. References：
[1]Biber， D.， Conrad， S. Register， Genre， and Style[M]. Cambridge： Cambridge University Press，2009.
[2]Biber， D.， Conrad， S.