1．Introduction 1
1.1．Purpose of the present study 1
1.2．Raison d'être 2
1.3．Overview of the present study and major terms defined 4
2．Relevant Theoretical and Practical Issues in Perspectives 6
2.1．IL and relevant issues in SLA 6
2.1.1．Interlanguage 8
2.1.2．Error analysis 10
2.1.3．Linguistic Universals and L1 Transfer 13
2.2．Corpus linguistics and its applications 15
2.2.1．Corpus and computational linguistics 16
2.2.2．Applied corpus linguistics and the Internet 19
2.2.3．Word sense disambiguation 25
2.2.4．WordNet,Semcor and FrameNet 30
2.2.5．Cobuild and TACTweb:online aids to corpora studies 33
2.3．The role of lexis in L2 teaching and leaming 35
2.3.1．Ahistorical perspective on vocabulary teaching and research through vocabulary selection 35
2.3.2．West's A General Service List of English Words 39
2.4．Verb classifications 43
3．Research Design and Methodological Issues 47
3.1．Research Design 48
3.1.1．Rationale 48
3.1.1.1．Why a contrastive study between West's list and Chinese leamers'performance? 49
3.1.1.2．Why choosing verbs as a model construction interface? 51
3.1.1.3．Similarities and dissimilarities 52
3.1.2．Hypotheses and predictions 53
3.1.2.1．Hypothesis one:Correlations in L1 and L2 data in sense selection are positively corresponded to the level of L2 acquisition and leaming 53
3.1.2.2．Hypothesis two:Clustering in L1 and L2 data in sense selection is pattemed according to leamer levels in L2 acquisition and leaming 55
3.1.2.3．Hypothesis three:There should be more collocates in L1 data than in L2 data 56
3.1.3．Data 57
3.1.3.1．Groupings of CLEC 58
3.1.3.2．Groupings of verbs 59
3.1.3.2.1．A complete list of the chosen verbs 59
3.1.3.2.2．Grouping of the verbs according to their meanings 63
3.2．Methodological issues 73
3.2.1．Tagging of meanings 74
3.2.2．Comparison of semantic frequency 80
3.2.3．Studies of collocates 82
3.2.4．Identification of similarities and dissimilarities by statistical means 85
3.3．The corpora involved 86
3.3.1．CLEC 86
3.3.2．LOB 87
3.3.3．BNC 88
3.3.4．Longman corpus 88
3.4．Textual analysis tools 89
3.4.1．MicroConcord 90
3.4.2．TACT 91
3.4.3．WordSmith Tools 92
4．Findings and Discussion 94
4.1．Quantitative Analytical Procedures 94
4.1.1．Comparison with West's Service List 94
4.1.2．Comparison against Longman Corpus 98
4.1.3．Comparison against LOB 104
4.2．Qualitative procedures:results from case by case examinations 110
4.3．Keyword studies 116
4.4．Findings in summmary 120
4.5．Related Issues in Perspectives 122
4.5.1．The vocabulary lists in curriculum guides examined 123
4.5.2．Proposal for newer vocabulary guides 126
5．Conclusion 129
5.1．Summary 129
5.1.1．Conclusions regarding Hypothesis One:Pattems hidden in the seemingly different L1 and L2 verb uses 129
5.1.2．Conclusions regarding Hypothesis Two:L2 leamers'repertoire of verbs comparable to L1 speakers' 130
5.1.3．Conclusions regarding Hypothesis Three:Analysis of verbs sense selection pattems point to tendency of better semantic than syntactic attainment for L2 leamers 131
5.2．Implications for language teaching and leaming 133
5.3．Recommendations for further studies 134
Glossary 137
Bibliography 144
Appendix A:Verbs by Semantic Domains from West and the CLEC 151
Appendix B:Verb lists from BNC ＆ West's Service List 158
Appendix C:Percentage curves for top 10 verbs 168
Appendix D:Comparison of Verb Sense Uses in Percentages between 380 verbs in both West(1953)and the CLEC 169
Appendix E:Key Verb Uses Compared in CLECMA against BROWN and CLECMA against LOB 182