Chapter One Introduction
1.1Research Background
Liu Runqing (1996) pointed out that English teaching in the future will be connectedmore and more with the specialized knowledge or a particular aspect of a subject. Thecombination of specialized knowledge and English teaching that is known as the English forSpecific Purpose (ESP) teaching will become the mainstream of the English teaching in the21st century. With the rapid development of China's economic, scientific and technologicalcommunication with the whole world, the single skilled learners can not adapt to the socialdemands anymore, and the talents who have solid technical knowledge and higher foreignlanguage ability are more and more popular. Thus, the role of cultivating inter-disciplinarytalents for the society of the regular colleges and universities is becoming more and moreobverse. In this kind of demanding circumstance, the original English teaching model in thecolleges and universities could no longer meet the requirements of social development, whichasks urgently the ESP teaching, which is an effective solution to the problems mentionedabove. Currently, in the universities where ESP teaching is being carried out, there is also abottle neck that the ESP teachers are mainly composed of college English teachers, whoselack of technical knowledge especially technical vocabulary makes ESP teaching lesseffective. (Zhang Mei，2011)
In the field of corpus linguistics, specialized corpus refers to the corpus that is built forsome specific study purposes, and it is established by collecting the sample texts of aparticular field or subject (Liang Maocheng，2010). In the 1980s, professor Yang Huizhongtogether with the project team members in Shanghai Jiao Tong University designed anddeveloped the JDEST academic corpus, which marked the beginning of corpus linguistics inChina (Zhen Fengchao，2010). Special English corpus, the perfect combination of corpuslinguistics and ESP teaching, is quite useful in solving the problem of lacking of technicalvocabulary during the ESP teaching process.The development of corpus linguistics and the grown of ESP teaching mutually blossomand complement each other. Corpus linguistics provides a new method and a new perspectivefor the study of ESP teaching. In return, the development of ESP teaching makes requests onthe linguistics theories, and its new research topics are therefore put forward. At the sametime, the use of specialized English corpus in ESP teaching and researching also enriches the content of corpus linguistics (Chen Jiansheng & Tan Qingsong, 2009). Thurstun & Candlin(1998) stated more precisely that the major influence of corpus linguistics on the ESPteaching and researching is reflected by the vocabulary. (Zhang Jihua, Gao Qin & WangBeilei，2009)
……………
Chapter Two Literature Review
2.1 Corpus and Corpus Linguistics
Scholars at home and abroad attempted to define the term corpus. Leech (1992) definedthe corpora as generally assembled with particular purposes in mind, and is often assembledto be representative of some language or text type. Sinclair (1996) thought that a collection ofpieces of language that are selected and ordered according to explicit linguistic criteria inorder to be used as a sample of the language is a corpus. Richards et al. (2000) held theopinion that a corpus is a collection of materials that has been made for a particular purpose,such as a set of textbooks which are being analyzed and compared or a sample of sentences orutterances which are being analyzed for their linguistic features. And McEnery et al. (2005)once said in their new book that corpus as used in modem linguistics can best be defined as acollection of sampled texts, written or spoken, in machine-readable form which may beannotated with various forms of linguistic information. (Zhao Xiaodong, 2008) LiangMaocheng et al. (2010) concluded that the term corpus came from Latin, and its originalmeaning is a body. Nowadays, what we talk about corpus refers to a collection of texts storedin an electronic database, which is collected following some particular standards and canrepresent the variants of a language or several similar languages, while the collection of asmall amount of texts can not be called a corpus. Thus the researches on a corpus based on thedata source could be treated as that on the language features or variants that the language canreflect，therefore the results gotten from the analysis of one single corpus can be extended tothe whole language.
……………
2.2 Specialized Corpus and ESP
2.2.1 Development of Specialized Corpus
Specialized corpus also called special purpose corpus, opposite to the general corpus isdesigned and established with a special aim by collecting the texts in some particular domains.And during the research process, specialized corpus is often compared with general corpus toidentify the characteristics of certain language, besides specialized corpus is also the idealmaterial resources for the compiling of reference books.I Specialized corpus began emerging in the late 1960s abroad, and it matures now, whichcan be proved in the following statements: (1) numerous achievements: from 1980 to 2008，ifcorpus was set as a keyword，more than 200 articles about specialized corpus could be foundin the journal named as ESP that was presided by American Academy of English in the early1980s (Liu Mantang, 2004); (2) wide range of research: there were lots of outcomes, such asAHI Corpus for lexicography; Helsinki Corpus of English Texts for the study of Englishlanguage variation; Wales Institute of Technology Corpus，Child Language InformationIExchange System Corpus and International Learner English Corpus for language acquisitionresearch; Nijmegen Corpus for English grammar study and English Translation Corpusestablished by University of Manchester for translation study; (3) comprehensive content: theresearch content of specialized corpus research involved machinery, construction, medicalscience and other topics. With the deepening of corpus researches and the innovation ofcorpus technology, development of overseas specialized corpus appeared to be of twocharacteristics, which were specialization/small sized and multi-modal of corpus data. (1)
………………
Chapter Three Research Methodology..........19
3.1Corpus Establishment.........19
3.2Data Collection.........20
3.2.1Text Collection.........20
3.2.2Text Editing.........22
3.2.3Annotation.........23
3.3Data Processing.........26
Chapter Four Results and Discussion.........30
4.1Macroscopic Lexical Features.........30
4.1.1Word Frequency.........30
4.1.2Part of Speech.........31
4.1.3Word Difficulty.........32
4.2General Agriculture Words.........33
4.3Technical Agriculture Words.........34
4.4Keywords.........37
4.5Colloquialization.........39
Chapter Five Conclusion.........40
5.1Major Findings.........40
5.2Implications.........41
5.3Limitations and Suggestions for Further Study.........42
Chapter Four Results and Discussion
4.1 Macroscopic Lexical Features
Word list 1 is actually a word frequency list of all the 6304 types of the VOAARC. Bypiding all these words and expressions into five degrees: the most frequently used, the morefrequently used, frequently used，the less frequently used as well as the least frequently usedas can be seen in the below table, a basic word description overview is revealed. From Table 4-1, it can be found that the number of the most frequently used words andexpressions is the least while the number of the least frequently used word and expressions isthe most so we can come to the conclusions that the word frequency is inversely proportionalwith the word quantity. What's more, more than half of the words or expressions appear nomore than ten times in the list. The most frequently used words rank from No 1 to No 19which include the function words as auxiliary verbs: can, conjunctions; and, determiners: aand the, prepositions: of, to, in, for，but and from etc, pronouns: it, they，that and this etc, adverb: not and some verbs: plant and say etc; the words with the highest frequency aremostly the function words, which shows the basic function of language, transferring basicinformation. And it coincides right with the statement of Liang Maocheng (2010). The rankNo 1 word the's frequency is 12550, the second word he's is 7103 and the third's, the fourth'sas well as the fifth's is respectively 5448，4849 and 4497. So the gap between each two is5447，1655，599 and 352 and they are decreasing roughly, which shows that the higher thefrequency is，the larger the gap between these words with the frequencies is.
……………
Conclusion
Firstly, the macroscopic lexical features: the VOAARC contains 515 VOA AgricultureReport texts from 2002 to 2011，with 6304 types and 197,983 tokens. Its TTR and STTR arerespectively 3.18 and 42.31, which are all smaller than that of the Brown Corpus. Though theaverage word length of VOAARC is a little bit bigger than that of the Brown Corpus, inconclusion the VOA agriculture English is less complicated than general English on lexicallevel. As for the word frequency，the number of the words with the highest frequency is lowwhile the number of the words with the lowest frequency is high，which means frequency of aword or an expression is inversely proportional to its rank in the frequency table of theVOAARC. As for the part of speech, compared with the general corpus, the content wordsused in the specialized corpus the VOAARC play a much more important role, and the morethe theme of a specialized corpus is limited, the more the content words are used and thespecialized corpus is more valuable in the study of the content words compared with thegeneral corpus.
……………
Reference (omitted)