turn-initial bushi in BCC Corpus（1296 token）
Description
The first stage utilizes a large-scale corpus to observe the overall distributional characteristics of turn-initial bushi repair sequences and to categorize their subsequent interactional trajectories. The second stage uses audio-recorded corpora of natural conversation to conduct micro-level sequential conversation analysis on the identified trajectories, thereby scrutinizing their specific organizational mechanisms. The data for the first stage are drawn from the Spoken Sub-corpus of the BCC Corpus (Beijing Language and Culture University). Interactional instances containing turn-initial bushi were first retrieved and then manually screened, only cases with a complete sequential structure were retained. That is, instances were selected only if they fully manifested the prior understanding (T1), the bushi repair (T2), and the subsequent response (T3), yielding a total of 1,296 tokens. Some materials within the BCC Spoken Sub-corpus originate from online interactional texts such as Weibo and therefore do not strictly constitute audio-recorded natural conversation. The purpose of utilizing this corpus segment is not to conduct fine-grained conversation analysis, but rather to observe the interactional distribution of turn-initial bushi repair sequences across a large-scale sample, and to identify their subsequent interactional trajectories through constant comparative analysis.
Files
Institutions
- Hunan Normal UniversityHunan, Changsha