A Re-examination of Syntactic Complexity by Investigating the Internal Structure Variations of Adverbial Clauses across Speech and Writing

A Re-examination of Syntactic Complexity by Investigating the Internal Structure Variations of Adverbial Clauses across Speech and Writing Mingyu Wan Dept. of Linguistics and Translation City University of Hong Kong mywan4-c@my.city.edu.hk Alex Chengyu Fang Dept. of Linguistics and Translation City University of Hong Kong acfang@cityu.edu.hk Abstract This paper re-examines the debatable issue about adverbial clause (A,CL) whether it in- dexes a complex discourse of contemporary English and how it distributes across Speech (S) and Writing (W) with different subtypes of structures by investigating the internal struc- ture variations. A Finite-State-Machine model is adopted for retrieving the internal struc- tures. Empirical results show that A,CL pre- vails in W than in S with a higher occurrence rate (W: 31.20% vs. S: 14.80%), which con- firms its function of indexing a complex dis- course; but the standard token-type-ratio of its internal structures shows an opposite dis- tribution (W: 12.89% vs. S: 16.66%), which suggests a higher structural density/variation of the spoken mode; besides, five subtypes of internal structures are identified with vari- ous distributions across S and W: S employs a higher proportion of subordinator-overt sub- ordinating A,CL, while W adopts more in- finitive A,CL, including to-infinitives, present and past participles; coordinated embeddings and subordinator-covert finite A,CL are com- monly found in both modes. Despite of the individual variance of internal subtypes, sta- tistical test indicates a less noticed fact that the overall structural variation of A,CL between S and W is not significant (p-value = 0.5245). 1 Introduction The definition of language complexity is generally quite broad and has not reached much consensus among researchers of various fields, such as in lin- guistics, cognitive linguistics, machine translation and L2 learning. People may interchangeably refer to it as text readability (Just and Carpenter, 1980; Rayner, 1998; Vajjala et al. , 2016; Charles et al. , 2007), or language preparedness/elaboration (Fang, 2006; Fang and Cao, 2015), or writing quality ( Crossley et al. , 2010; Crossley and Danielle, 2014), or language proficiency (Piln et al. , 2016; Ortega, 2003; Larsen-Freeman, 2006; Housen and Bram , 2012). Pallotti (2015) underlines the polysemy of the term complexity in the linguistic literature and sum- marizes the different notions of complexity in this field by referring to three main meanings: • Structural complexity, a formal property of texts and linguistic systems having to do with the number of their elements and their rela- tional patterns. • Cognitive complexity, having to do with the processing costs associated with linguistic structures. • Developmental complexity, the order in which linguistic structures emerge and are mastered in second (and, possibly, first) language acqui- sition. Not only the definitions are debatable, but also the measurements. Biber (1988) did an influential work on reporting 67 linguistic variations across Speech and Writing based on multi-dimensional factor anal- ysis. Heylighen and Dewaele (1999) developed a simple but effective formula to calculate the fuzzi- ness/complexity of language in terms of formal- ity as in“F=(noun freq.+adjective freq.+preposition PACLIC 32 715 32nd Pacific Asia Conference on Language, Information …

Content truncated to 500 words