Chapter OneIntroduction
1.1Background of the Study
The word "chunk" can be used as an umbrella term to refer to the multi-wordcombinations or units that play an important role in language acquisition，processingand use. The study of chunks can be traced back to 150 years ago when scholars beganto realize the importance of multi-word combinations in child's language acquisition.From then on，chunks have been characterized in a variety of ways，and terms such as“prefabricated chunks", “formulaic language"and "lexical phrases" have been usedto refer to different types of chunks. Given its importance, researchers in recent yearshave shown great interest and enthusiasm in the studies of linguistic chunks.According to Miller (1956)，"chunk" was first used as a psycholinguistic andcognitive psychological term to refer to the unit that forms human beings' memoryspan. Human beings can increase the number of bits of information that the memorycontains simply by building larger and larger chunks, with each chunk containingmore information than before. Then the concept was transferred to the research fieldof language to refer to the fixed or semi-fixed structures or combinations used as thesmallest units of human communication (Becker，1975). They are on the continuumbetween grammar and lexicon, and are composites of grammar, lexicon and function.In the 1970s, the study of chunks has become an important area in second languageacquisition research. The multi-faceted nature of chunks has been characterized basedon their relationship with novel language, and on form, fiinction, lexical, syntactic andsemantic properties.
……………
1.2Statement of the Problem
As has been mentioned, the role of chunks has long been recognized and re?mained as an important focal area for researchers in English language acquisition,processing and use for both English native speakers and ESL/EFL learners. In secondlanguage acquisition, for example，chunks can be used as communicative strategy,learning strategy, and production strategy both in oral and written language. But interms of accuracy, fluency, and complexity of language production, chunk use is sel?dom used as a measure to determine the quality of linguistic output.Literature shows that most studies on chunks mainly focus on spoken languageand have been limited to English native speakers only，or one group of ESL/EFLlearners. Though comparative results are available concerning the use of chunks between the English native speakers and one group of ESL/EFL learners, few studieshave taken students with different LI backgrounds as the target groups to study, letalone the comparisons of chunk use by these learners who have involved differentmother tongues in their English learning.As to the studies on the output quality of English writings, most of them tookaccuracy，fluency, and complexity as separate focus, and seldom studied the three as?pects as a triad in a whole.In addition, there is a lack of studies on the relationship between the overallchunk use and students, writing proficiency. More comprehensive and in-depth studiesare needed in this aspect to find out how strong the correlation is between chunk useand students' writing proficiency.
………
Chapter TwoLiterature Review
2.1Introduction
This chapter mainly reviews previous studies on chunks, accuracy, fluency, andcomplexity in language production and the relationship between chunk use and accu?racy, fluency, and complexity. First of all, it reviews studies on the definitions and thedefining criteria of chunks and the role of chunks in LI and L2 acquisition and lan?guage pedagogy. Then, previous studies on the definitions of accuracy, fluency，andcomplexity in language production and the measures for accuracy, fluency, and com?plexity are reviewed. Studies on accuracy in language pedagogy，fluency in linguisticoutput, and complexity in language pedagogy are also reviewed. The relationship be?tween chunk use and accuracy，fluency, and complexity is another important part ofthe review. The working definitions of chunks, accuracy，fluency, and complexity, aswell as the working criteria for chunk identification are provided to set up the basicconceptual framework of the present study. In the last part of the chapter, researchgaps are pointed out and research questions are raised.
……………
2.2Studies on Chunks in Language Production
Previous studies in language production include studies on the definition ofchunks, the defining criteria of chunks such as fixedness, idiomaticity, institutionali?zation; studies on the role of chunks in LI and L2 acquisition; and studies on the roleof chunks in language pedagogy. A number of scholars have attempted to define chunks from different perspec?tives, Among these scholars, there are two groups deserving more attention，one being"lexical" in the label, and the other "formulaic". Despite the different terms they use,the key concept they refer to is more or less the same thing in nature. For example, Pawley and Syder (1983) used the term "lexicalized sentencestems” to refer to the sentence length units with grammatical form and lexical contentlargely or completely fixed. It is lexicalized in the sense that its fixed elements make astandard label for culturally recognized concept. They used "semi-lexicalized se?quence" to express a sequence between lexicalized formations and nonce forms, that is，a lexicalized sentence stem with permissible expansions and substitutions. They foundthat possessing a large number of sentence stems could add to the fluency of language,and that the formation of these sentence stems was closely related to the institutional?ization and lexicalization processes of language.Nattinger and DeCarrico (1992) labeled prefabricated chunks as "lexical phrases"to distinguish them from other conventionalized or frozen forms such as idioms andcliches. To them，lexical phrases also differ from collocations that are chunked setsof lexical items with no particular pragmatic functions. They are the multi-word lex?ical phenomena that exist on the continuum between lexicon and syntax，consistingof larger stretches of language with their meaning derivable from the inpidual con?stituents. The term “lexical” is used in the label in the sense that the phrases haveimplicatures and speech act functions besides their literal sense, and that these addi?tional meanings are conventionalized or institutionalized and，in some sense, arbi?trary.
…………
Chapter Three Methodology
3.1Introduction..........56
3.2The Corpus Used in the Study..........56
3.3Tools Used in the Study..........57
3.4Tagging of the Corpus..........58
3.5Analysis of the Data..........63
3.6Chapter Summary..........66
Chapter Four Shared Features in Chunk Use by All Language Groups
4.1Introduction..........67
4.2Shared Features in Three Major Categories of Chunks..........67
4.3Shared Features in Using Part-of-speech-based Chunks..........69
4.4Shared Features in Using Preposition-based Chunks..........73
4.5Shared Features in the Use of Clause-based Chunks..........76
4.6The Partially Shared Features in the Use of Verb-based chunks..........78
4.7Chapter Summary..........80
Chapter Five Discrepancies in Chunk Use Between Language Groups
5.1Introduction..........82
5.2Differences in Chunk Use by Total Number..........82
5.3Differences in Chunk Use by Main Category..........84
5.4Differences in the Use of Part-of-speech-based Chunks..........85
5.5Differences in the Use of Clause-based Chunks..........93
5.6Chapter Summary.....94
Chapter EightDiscussion
8.1Introduction
This chapter provides a discussion of the findings based on the results reportedfrom Chapter Four to Chapter Seven. It mainly discusses the position of chunks as partof natural language use，the specific features of chunks used by students with differentLI backgrounds, a hierarchical model that reflects the shared patterns of chunk use，the role of chunks in facilitating writing accuracy, fluency，and complexity, and therole of chunks in predicting the English writing proficiency. The results and findingsare discussed by relating to results of previous studies and relevant theories. It has been found in the present study that both the English language group andthe non-English language groups use chunks definitely. The statistics presented pre?viously show that the proportion of the total number of chunks identified in the pre?sent study in relation to the total running words of the corpus is 3.85%. The proportionof chunks used by the Chinese, the Spanish，the Arabic, and the English languagegroups in relation to the total running words in the corpus is respectively 1%, 0.81%，0.91%，and 1.13%. And the proportions of chunks used by the Chinese, the Spanish,the Arabic, and the English language groups in relation to the total running words ofthe respective sub-corpora are respectively 3,78%, 3.79%，3.8%, and 3.98%. Both theEnglish language group and the non-English language groups use chunks definitelyalthough the non-English groups overuse, underuse，and misuse chunks.
…………
Conclusion
More studies need to be conducted in order to provide a better understanding ofthe use of chunks in the English writings by students with different LI backgrounds.The future studies may extend the present study in a number of ways.It would be better for the future study to replicate the present study in a muchbigger corpus, a corpus consisting of more samples from more language groups. Ifmore students with more varied LI backgrounds could join future studies，a morethorough picture of chunk use by both native and non-native students could be gained.Since the present study is cross-sectional in nature, it would be better for the fu?ture study to incorporate longitudinal research to obtain more developmental featuresand trends in the use of chunks in the English writings of different language groups.In the future, qualitative analysis could be conducted to have a closer examina?tion at the focused samples and particular cases of chunk use in English writings.More follow-up procedures such as questionnaire investigations and face-to-face interviews are needed if the subjects are available.
……………
Reference (omitted)