A Corpus-Based Study of Nominalizations across Chinese Media English and British Media English


Student thesis: Doctoral Thesis

View graph of relations


  • Ying LIU


Awarding Institution
Award date4 Aug 2015


Variation is inherent in human language and linguistic variation is crucial for the study of language use. The topic of linguistic variation in different English varieties has received much scholarly attention over the past three decades within the context of world Englishes (e.g. Kachru, 1982, 1992; Kirkpatrick, 2007). Considerable research has been done to explore the cross-variety differences across English varieties. Previous studies of English variety variation have concentrated on comparing two native varieties – British English and American English or on comparing a few non-native varieties like Indian English with the two native varieties. There are far fewer empirical studies of English varieties in the expanding circle such as China English. With regard to the linguistic features under comparison, previous studies have primarily focused on the lexical aspect (such as articles, particle verbs and prepositions) and the grammatical system (such as modality, tense and aspect). Comparisons of syntactic features are rarely found apart from several studies of verb complementation patterns (e.g. Mukherjee and Hoffmann, 2006; Mukherjee and Gries, 2009), thus leaving syntactic variation in different English varieties a largely ignored area.
Nominalization is pervasive in language and has received much attention in linguistic enquiry from various perspectives, covering aspects of its form, function and meaning. Among other things, it is of close relevance to language variation studies. One of the effects of nominalizations is that information is presented through phrasal structures rather than through clauses that incorporate verbs. Thus, nominalization, as a syntactic construction, is found to have the function of distinguishing a nominal and compressed style from a colloquial one and its distribution varies a lot across registers and genres (e.g. Biber, 1986; Greenbaum, 1988, etc). But until now, there is only a small amount of literature concerning corpus-based study of nominalizations, and most extant studies deal with this phenomenon in a relatively superficial and practically easy way. Moreover, how the uses of nominalization vary across different English varieties is still an underexplored area.
Based on the above rationale, the present study sets out to explore variations at the level of nominalization across China English and British English. We will examine those constructions describable as “a noun phrase such as the quarrel over pay which has a systematic correspondence with a clause structure and the noun head of such a phrase is normally related morphologically to a verb” (Quirk et al., 1985: 1288) by using data from two corpora – the Chinese Media English Corpus and the British Media English Corpus. Specifically, this study is interested in the quantitative comparisons with an emphasis on whether or not the two varieties of English demonstrate any significant differences regarding this particular type of syntactic construction. Nominalization involves morphology, syntax and semantics. Thus, we will present the quantitative and distributional results in terms of morphological structure, syntactic structure and argument structure. Variation will be shown at the level of the overall corpora and also across their different text categories.
It is observed that there are substantial intervarietal differences between Chinese Media English and British Media English. In terms of the morphological structure, Chinese Media English uses significantly more suffixed nominalizations than BME overall and also across the five categories (except social life), indicating a more nominal and formal style in CME. In contrast, British Media English has significantly more uses of converted nominalizations overall and in all the five categories, which might have something to do with the possible association between converted nominalizations and informality of writing.
Additionally, Chinese Media English and British Media English also differ markedly in terms of the syntactic structures of nominalizations. The results show that there are much more complex nominalizations in CME but simple nominalizations are more common in BME. Furthermore, CME has more uses of premodified nominalizations and phrasal postmodified nominalization and develops a reliance on compressed and phrasal types of modification, whilst BME tends to use more postmodified nominalizations and develops a reliance on expanded and clausal types of modification.
Lastly, there are also variations in the argument structure of nominalizations between Chinese and British Media English. The total number of semantic arguments taken by nominalizations in Chinese Media English is significantly larger than those in its British counterpart, especially in categories of business, editorial and news report. These two English varieties also differ in the complexity of argument patterns. Chinese Media English is found to have a more complex argument pattern, evidenced by more uses of nominalizations taking one or more semantic arguments than British Media English. In addition, CME tends to have a left-branching of arguments manifested in the more uses of premodified nouns and adjectives to express arguments, but BME tends to have a right-branching of arguments manifested in the more uses of postmodified infinitive clauses and finite clauses to express arguments.
Large-scale quantitative analyses of nominalization from a structural approach across different English varieties have, to our knowledge, not yet been carried out. Such quantitative differences in usage are not immediately accessible to intuition and can only be identified by analyzing large amounts of natural data. It is hoped that this research will offer useful insights on syntactic variation across different English varieties and also on the understanding of nominalization as a syntactic feature and the two English varieties under study. Besides, the research results are expected to be applied to computational technologies such as automatic text classification and semantic role labeling of nominal predicates.