A Corpus-based Study Of Split Infinitive

The split infinitive is considered as one of the most controversial grammar structures in English. On one hand, many grammarians have severely criticised and opposed its usage in English, especially in written English, because they regard it as an ungrammatical structure and think that people should not break the traditional rule—"never split the infinitive". On the other hand, this structure has been employed more and more frequently in many kinds of domains and language genres. After reviewing the related literatures, the author found that in resent years, more and more attentions have been paid to split infinitive, but most of the researches are based on a few examples that occasionally obtained by the researchers and analysed according to their language instincts. No systematic study of this subject, especially the research on the frequency and features of the construction that occurs in real languages, has been found so far. This makes the application of the structure lack of persuasive power. Now, with the development of the corpus linguistics and large-scale corpora, scholars have obtained a brand-new research method—the corpus-based method—to analyze English grammar. The present research will apply this method to studying the exceptional case of grammar– the split infinitive.The main purpose of this research is to count, analyze and summarize the usage of split infinitives in real languages by applying the corpus-based research method, to further concluding the situation of its proper use and support the use of split infinitives, and to provide objective basis for the application of corpus linguistics in the study of grammar.The corpus used for the research is the British National Corpus (the BNC), which contains both written and spoken form of texts. After inspecting the codes used in BNC, the author designs the retrieval sentences according to the part-of-speech feature of split infinitives, and then applies the software AntConc3.2.1 to retrieving the construction of split infinitives in BNC and getting the related results. The combination of quantitative and qualitative methods has been used to systemically study the situation of its usage in the whole corpus and its sub-corpora. To be specific, the quantitative method is adopted to summarize the frequency and distribution of split infinitives among different domains and genres; the qualitative method is adopted to check the examples obtained from BNC. Several features of split infinitives have been concluded, such as, the forms, types, and functions of split infinitives.The analysis of split infinitives found in BNC has brought the major findings as follows:a. There exists large number of split infinitives in BNC and the total number reaches 4,572. As to the relative occurrence of split infinitives (tokens per million words), the construction occurs more often in spoken English than in written English;b. Split infinitives appear in all sub-corpora (containing 13 language domains) of BNC. In written materials, the construction most frequently exists in the domain of applied science. In spoken materials, split infinitives appear very frequently in four sub-corpora: business, public affairs, education and entertainment, except in the demographic sub-corpus;c. In the sub-corpora categorized according to written genres, split infinitives appear most frequently in the genre of newspaper. This result indicates that the openness and inclusiveness of media create a loose environment for the use of split infinitives;d. There are mainly three existing forms of split infinitives: the simple split infinitives, the compound split infinitives and the split negative infinitives. The first kind occurs most frequently in BNC, the number of simple split infinitives has reached over 90% of the total occurrence, and the second and the third kinds rarely appear. However, after searching BNC, the author does find some examples of these two kinds and most of them are properly used;e. After studying the splitters used to split the infinitives, the author found that almost all adverbs can work as splitters, except the location adverbs. The adverbs of manner, frequency and degree are the most frequently used types. As to the occurrence of certain adverbs or adverbial phrases, the author found that the words that can be widely used to split the infinitives are relatively concentrated. The number of the top 10 splitters amount to about 43.7% of the total (500) , and more than half of the 500 splitters only appeared once or twice;f. In the discussion part, the author studied split infinitives in the context, and found that this construction has its special meanings in the context: firstly, it avoids ambiguity and makes the sentence more clear; secondly, compared with other alternative constructions, split infinitives put emphasis on the infinitive verbs; last but not least, the application of split infinitives will enhance the rhythm;Considering the relative materials and the present research findings, the author further proposes several suggestions on the usage of split infinitives:a. More attention should be paid to distinguish what is a real split infinitive structure and what is not;b. When the split infinitive can be easily altered by other structures (replacing the adverb before or after the infinitive), and there is no meaning differences between the two structures, the split infinitive is better to be avoided;c. When the splitting element is too long, the sentence will become awkward and clumsy. Therefore, it is usually the case that if the splitter contains more than two words, the split infinitive must be avoided;d. Due to the historical reasons, the argument for the use of split infinitives will continue for a relatively long time. However, if split infinitives are required to meet the requirements of communication, we need to boldly split the construction but not to rigidly adhere to the old forms.The most significant innovation of the current study is the attempt to apply large-scale corpus and the combination of quantitative and qualitative methods to studying the structure and functions of split infinitives. It has not only given guidance and objective support to the proper use of split infinitive construction, but also shed some light on the application of corpus linguistics to the study of grammar. This research has also implied that, grammatical rules still have to keep pace with the times, and the descriptive grammar is better in reflecting the current situation of English grammar. Meanwhile, because it is the first time that the author has dealt with this area, there are still many shortcomings. For instance, the research and analysis methods still need to be improved on in future researches.