Abstract
In the business context, effectively responding to negative reviews is critical for a hotel to maintain reputation and customer relations. To explore the linguistic devices employed in addressing guest complaints, a corpus-based study is conducted on the use of interactional metadiscourse and identity construction in responses to negative online reviews of Chinese and British Hotels. Drawing upon the statistical results of the usage of interactional metadiscourse and the analysis of discourse examples, this study delves into the frequency and similarities/differences in the employment of five subcategories of interactional metadiscourse across the respective corpora of 100 responses to negative reviews from hotels in Beijing and hotels in London. Furthermore, the study examines the characteristics and similarities/ differences of the identity construction of manager, communicator, doer and advisor with the use of interactional metadiscourse. The findings reveal that there are significant differences between the two sides in the use of self-mentions, boosters, hedges and positive attitude markers, while there is little difference in the use of engagement markers and negative attitude markers. The most constructed identity by both sides is the communicator, and the least is the adviser, with little difference. The identity of manager is significantly more prevalent in responses from hotels in Beijing, whereas hotel responders in London exhibit a notably greater tendency to construct the identity of doer. The similarities and variances of interactional metadiscourse use and identity construction indicate the two sides' distinctive priorities in interactions with guests and different cultural values, which provide valuable insights for hotels on the effective use of metadiscourse to construct multiple identities, revealing that the strategically crafted responses play a pivotal role in shaping favorable images, fostering harmonious relationships with customers and promoting sustainable development of the hotels.