A Quantitative View of Short Utterances in Daily Conversation: A Case Study of Thats right, Thats true and Thats correct

PACLIC 28 !378 A Quantitative View of Short Utterances in Daily Conversation: A Case Study of That’s right, That’s true and That’s correct Yanjiao Li Department of Linguistics and Translation City University of Hong Kong Hong Kong SAR yanjiaoli2-c @my.cityu.edu.hk Alex C. Fang Department of Linguistics and Translation City University of Hong Kong Hong Kong SAR acfang@cityu.edu.hk Jing Cao School of Foreign Languages Zhongnan University of Economics and Law Wuhan, P. R. China cecilia_cao @znufe.edu.cn Abstract Short utterances serve a multitude of different communicative functions in interactive speech and have attracted due attention in recent re- search in dialogue acts. This paper presents a quantitative description of three short utteranc- es i.e. that’s right, that’s true, that’s correct and their variations based on the Switchboard Dialogue Act Corpus. Particularly, it offers an overview to account for how they are deployed by native speakers in daily conversation. At the same time, it attempts to provide a comparative account of that’s right and that’s true, showing that while almost 75% of them are mutually exchangeable, they nonetheless exhibit prefer- ences in interactive speech. This insight is ex- pected to form a useful approach towards automatic dialogue act tagging. 1 Introduction Dialogue act (DA), defined as “communicative activity of a dialogue participant, interpreted as having a certain communicative function and se- mantic content” (ISO 24617-2, 2012: 2), plays a key role in the interpretation of the communicative behaviour of dialogue participants and offer valua- ble insight into the design of human-machine dia- logue system (Bunt et al. 2010). With the goal of facilitating automatic DA tagging, this paper de- scribes a corpus-based investigation into that’s right, that’s true, that’s correct and their variations in the Switchboard Dialogue Act (SWBD) Corpus, in order to answer questions about the communica- tive functions they mainly perform in daily conver- sation. These utterances deserve our particular attention in research considering that, like other brief responses (e.g. Oh, Uh huh, Mm, Okay), they serve as important feedback to the main speaker and they usually occur as overlapping speech. They are particularly problematic to interpret be- cause they demonstrate a drastically different func- tional or pragmatic meaning from the semantic meaning of the component tokens. Consider Ex- ample 1. Example 1 sd B.54 utt1: -- {C and } I like that because [ it's a, + it's ] real easy to, {F uh, } follow for her, / sd B.54 utt2: {D you know, } {F uh, } {D gosh, } [ if, + if ] I read straight out of the Bible to her she'd <laughter> never understand any of it. / sd A.55 utt1: {D Well, } it's hard for me. / ba B.56 utt1: That's right <laughter>. / sw_0263_2226.utt This is one excerpt retrieved from the targeted cor- pus, which will be further illustrated in section 2. A and B, two speakers, are talking about books and literature, where B is describing one of her daugh- ter’s book, “real easy to follow”. The last utterance Copyright …

Content truncated to 500 words