|Abstract: ||本論文以二○一六年「英國脫離歐盟公投」(簡稱：英國脫歐)為題，運用Hyland (2005)交互後設論述的五項子分類，比較並分析英文報紙與英語新聞雜誌報導英國脫歐事件的交互後設論述用字及其頻率。本研究從《每日郵報》與新聞雜誌《經濟學人》線上資料庫中，篩選並隨機挑選五十篇相關文章，作為研究文本資料庫。研究結果發現，報紙與新聞雜誌在不同文體中使用交互後設論述符號的用途各異。以頻率而言，「強調」(booster)在不同文體中的的用法因內容而相似，「表態」(attitude marker)的用法因內容而相異，而「規避」(hedge)、「介入」(engagement)和「自稱標記」(self mention)的用法則因文體而異。質的方面，五種交互後設論述符號的用途皆因文體而異。本研究結果可啟迪未來研究，比較新聞文本中的交互後設論述，並應用於教學及實務：教師可依據交互後設論述符號在文體間的用途差異設計教材，新進記者則可瞭解前輩在報紙及新聞雜誌中，如何運用交互後設論述符號達成語用目的。最後，本研究盼能喚醒讀者的思辨意識，協助讀者留意筆者在新聞文章中，運用交互後設論述達到與讀者互動，進而影響讀者見解之效。|
Metadiscourse has been a popular research topic in recent years to explore resourceful insights of written texts. Specifically, interests of interactional metadiscourse have been growing. Interactional metadiscourse plays an important role in the interaction between writers and readers (Hyland, 2005). Therefore, the writers of journalistic articles widely employ interactional metadiscourse markers to persuade readers by conveying writers’ attitudes and stances to readers. Previous studies have examined academic, non-academic genres (e.g., literature and business) and journalistic genres (e.g., opinions, editorials and columns) across languages. Yet little attention has been paid to the use of interactional metadiscourse between journalistic genres. Therefore, this thesis investigated and compared the frequencies and functions of the interactional metadiscourse markers in political news articles that reported the Brexit referendum from an English-language newspaper, Daily Mail, and an English-language news magazine, The Economist. Fifty news articles were randomly selected from the online archives respectively. The results of this study revealed differences in most of the interactional metadiscourse markers. Quantitatively, similarities appeared only in boosters but differences were discovered in hedges, attitude markers, engagements and self mentions between the two genres. Further analyses revealed that genres contributed more to the differences in hedges, engagements and self mentions, yet the nature or types of articles might play a more important role in the frequency of boosters and attitude markers. Qualitatively, the uses of all interactional metadiscourse were different between the two genres. The findings can shed light on future research, pedagogies and practices regarding metadiscourse in journalistic writing. This study can serve as a foundation to future research on comparison of metadiscourse in different genres. In addition, instructors can take the findings as references to design teaching materials by presenting the different uses of interactional metadiscourse across genres. Moreover, novice journalists can learn to use interactional metadiscourse markers to serve pragmatic purposes. Finally, the findings can help general readers notice the rhetorical purposes of interactional metadiscourse markers in journalistic articles.