Master of Arts (Applied Linguistics)

This study attempts to investigate the linguistic landscape (LL) of the Public Housing and Development Board (HDB) estates of Singapore. Two HDB estates are selected as the research sites and a total of 486 signs are collected as data for analysis. This empirical study utilizes both quantitative and qualitative methods to reveal how the three mother tongue languages are represented in the LL of the housing areas, the difference between top-down and bottom-up signs, and lastly, to identify any disparities in language usage between the two selected HDB estates. The study draws on theoretical frameworks such as geosemiotics by Scollon & Scollon (2003), and Reh’s (2004) model for types of multilingual information to aid in analysis.

The study affirms prior research indicating that Singapore is an English-dominated society, a fact that is also reflected in the LL of the two HDB estates. English is the predominant language on signage, alongside the presence of the three mother tongue languages: Chinese, Malay, and Tamil, reflecting the nation’s multilingual identity. Chinese emerges as the second most prevalent language in these estates, while Malay and Tamil appear somewhat marginalized.

In both HDB estates, the LL exhibits a clear prevalence of English. The study also uncovers disparities in the LL between the two HDB estates, in terms of aspects like language combination patterns and appearance of traditional Chinese script, suggesting inconsistency across Singapore’s housing areas. This implies that the impact of the official language policy on signage may vary by region, emphasizing the dynamic nature of LL in different locales.

This dissertation discussed the complex representation of the Russian-Ukrain crisis in Chinese social media to reveal the ideologies and linguistic features of different political stances using a corpus-based Critical Discourse Analysis method. The analysis consists of two levels. The first is a transitivity analysis of the syntactic structure of clauses to identify the roles allocated to different entities mentioned in the corpus. The second is a comprehensive frequency analysis of lexicogrammatical issues to reveal the linguistic patterns of the representations.

As a result, this dissertation revealed how Russia supporters and Ukraine supporters post under the “positive us” and “negative them” structure of van Dijk (2011). Remarkable differences also appeared on the political affiliation of the users. The second part of this dissertation found that both Russia supporters and Ukraine supporters post positively about certain groups of culture and political tendency.

An English textbook is an essential guide for English language learners. In order to better understand the structure of textbooks, researchers have applied several approaches to analyze them. The corpus-based approach has gained momentum due to the benefits of enabling empirical studies on language use and the development of various computational tools that make the analysis more efficient. Traditionally, researchers intend to evaluate a whole textbook, or to concentrate on one aspect of the textbook, such as vocabulary, phraseology, or syntactic contents. However, the listening components of Chinese senior secondary school English textbooks have never been subjected to any systematic analysis. To this end, this study aimed to investigate the senior secondary textbook listening parts, applying additive Multidimensional Analysis (MDA) (Biber, 1988/1995) to analyze the linguistic features and functions of the listening corpus from 46 senior high school textbooks.

The data used in the present study consist of 431 texts from six Chinese state-authorized publishers of the textbooks’ listening sections. The Multidimensional Analysis Tagger (MAT) (Nini, 2019), which replicates Biber’s (1988/1995) MDA, was used to conduct the analysis on the texts. The texts were first tagged by using MAT and then mapped onto Biber’s (1988/1995) six dimensions. The registers and text types were also used as independent variables in MANOVA to explore their impact on dimension scores. The additive MDA revealed that textbook conversations although involved, non-narrative, and overt expression of persuasion, are not close to natural conversations of the linguistic features on Dimension 1 (Involved versus Informational production), Dimension 2 (Narrative versus Non-narrative concerns), and Dimension 4 (Overt expression of persuasion). Textbook public speaking is involved, non-narrative, and explicit; textbook short passage is informational, non-narrative, explicit, and abstract. The results showed that registers or text types can impact on dimension scores, while the interaction between registers and text types had no impact on dimension scores.

Implications of the current study can be considered from methodological and pedagogical perspectives. The additive MDA approach to analyze textbook listening texts provides detailed insights into the lexical, syntactic, and grammar structures. Researchers can use these findings to compare differences with other corpora to gain a more comprehensive understanding of the corpus. On the other hand, teachers can also use the pedagogical value of the findings to include supplementary authentic listening materials to enhance students’ listening capabilities.

Based on the findings, I argue that applying the additive MDA approach in English textbooks would be of help for teachers, textbook authors, and textbook publishers.