A Study On Inter-textual Vocabulary Growth Patterns For Maritime Convention English

Based on the maritime convention corpus, this study researches the inter-textual vocabulary growth pattern, tests the goodness of fit of existing mathematic models and finds out the most suitable model for vocabulary growth pattern of the maritime convention corpus. This paper utilizes Perl and SPSS to process and analyze data on the basis of quantitative linguistics and addresses three issues, that is, the lexical distribution of individual texts, the characteristics of vocabulary growth of maritime convention English compared with general English from BNC, and the mathematic models of vocabulary growth for the maritime convention corpus.The total number of tokens of maritime convention is372266, and this corpus was cut into360individual texts. Each text has approximately1000tokens. As is indicated in this research, these vocabulary sizes of individual texts fall into normal distribution with a95%prediction interval. After calculating type-token ratio, the conclusion is drawn that the TTR of maritime convention is much smaller than that of general English. This research tests three existing mathematic models (Brunet’s model, Tuldava’s model and Herdan’s model). It is revealed that all these three models fit the vocabulary growth of maritime convention well, yet Herdan’s model is the most suitable one, not only concerning the goodness of fit but also the brevity of expression.