在獨白裡，說話者大多只專注於自己的發言，鮮少與聽話者有口語上的互動。為了維護說話者的面子以及演說的流暢度，即便是報告內容有誤，聽話者也不常主動糾正。因此，說話者必須積極檢視自己的言談，以確保談話內容正確、適切。說話者在執行自我監聽時，會下意識地發出某些特定的言談標記；本文的重心便是在探討這些自我監視標記在獨白中的語用功能、言談分佈、使用頻率與其語義的歷史演變過程。本文的研究對象為大學教授與其修課學生；語料取自於課堂中的長篇口頭報告，收集到的語料包含中文標記「嘿」、「好」、「嗯」、「對」以及英文「alright」、「right」、「okay」、「yeah」 與「yep」。這些顯示說話者執行自我監聽的言談標記依功能可分為四類：自我證實標記、自我確認標記、全篇話語完成標記、篇中話語完成標記。研究結果顯示，篇中話語完成標記使用頻率最高。此外，這些標記的言談語義與其原始意義有密切的關聯。並且，說話者對某語言的熟悉度會影響該語言言談標記的使用頻率。最後，獨立標記在言談中的分佈決定複合標記的構成方式。

In a monologue narrative, the speech floor is by and large occupied by anindividual speaker; rarely are verbal interactions between interlocutors observed.Inaccurate utterances, if there are any, are seldom corrected by addressees so that the speaker’s positive face can be maintained in public. Without listeners’ active utterance-checking activities to ensure the appropriateness of their speech, addressers tend to self-monitor their words attentively while doing their presentations. As speakersare monitoring their speeches, a number of pragmatics particles are uttered sub-consciously. In this study, function, distribution, and frequency of those self-monitoring devices in monologue narratives are focused on. The data of the present investigation were collected from students’ presentations and professors’lectures in a university located in northern Taiwan. In this corpus, the Chinese markers heh, hao, N1, and dui as well as the English markers alright, right, okay, yeah, and yepare observed. These particles can be functionally classified into fourgroups—self-confirmation markers, self-assurance markers, current-utterance completeness markers, and utterance-internal completeness markers. These markers’pragmatic functions are not arbitrarily derived; instead, they are highly grounded on their lexical interpretations. Restrictions on forming compound monitoring mechanisms and how those pragmatic functions are derived from their lexical interpretations are also discussed.