**Advances in (Im)politeness Studies** Series Editor: Chaoqun Xie

## Mary Shin Kim Editor

# Exploring Korean Politeness Across Online and Offline Interactions

## **Advances in (Im)politeness Studies**

#### **Series Editor**

Chaoqun Xie, School of English Studies, Zhejiang International Studies University, Hangzhou, Zhejiang, China

The book series *Advances in (Im)politeness Studies* advances new perspectives, challenges and insights on (im)politeness studies and, in so doing, furthers understanding and interpretation of human worlds (online and offline) and human beings. (Im)politeness has, over the last several decades, become a multidisciplinary, interdisciplinary and transdisciplinary endeavor. (Im)politeness can be seen as a complex system, the production, perception, and evaluation of which may involve various components, linguistic, behavioral, cognitive, social, contextual, emotional, moral, historical, cultural and ethical. A full understanding of the (im)politeness system may only be reached by looking into the complex, fluid and dynamic interaction among those components.

The series invites innovative monographs and edited volumes that contribute to charting and shaping the discipline with respect to contemporary (im)politeness practice and research, that experiment with new and creative approaches to describing and explaining specific (im)politeness phenomena in either face-to-face communication or mediated interaction, or expound philosophical dimensions and implications of (im)politeness as a critical and essential lens through which to examine the full complexities and intricacies of human interpersonal interaction and human nature. Both experienced researchers and young enterprising scholars are welcome to submit their book proposals. The volumes in this series appeal to scholars and students of social interaction in general and pragmatics, sociolinguistics, philosophy, psychology, language teaching and learning in particular.

For inquiries and submission of proposals authors can contact the editor-in-chief, Chaoqun Xie, via: chaoqunby@hotmail.com

Mary Shin Kim Editor

# Exploring Korean Politeness Across Online and Offline Interactions

*Editor* Mary Shin Kim Department of East Asian Languages and Literatures University of Hawaii at Manoa Honolulu, HI, USA

ISSN 2524-4000 ISSN 2524-4019 (electronic) Advances in (Im)politeness Studies ISBN 978-3-031-50697-0 ISBN 978-3-031-50698-7 (eBook) https://doi.org/10.1007/978-3-031-50698-7

© The Editor(s) (if applicable) and The Author(s) 2024. This book is an open access publication.

**Open Access** This book is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this book are included in the book's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Paper in this product is recyclable.

## **Acknowledgements**

The inception of this book project can be attributed to the insightful proposal made by Dr. Chaoqun Xie, the editor-in-chief of the Springer Nature Book Series on Advances in (Im)politeness studies, who suggested and supported the publication of a volume dedicated to the most recent research on Korean politeness. I wish to express my deep appreciation to all the authors from various disciplines and research backgrounds who have generously shared their valuable research contributions. Their collective efforts have enriched this volume with a diverse array of studies on Korean politeness, spanning a wide range of interactional contexts and communication platforms. Additionally, each author offered valuable feedback on the chapters contributed by their fellow authors, thereby enhancing the significance of this collaborative project. I extend my appreciation to all the anonymous reviewers for their valuable and constructive comments and suggestions. I would also like to express my heartfelt thanks to Ann Tai Choe and Tyler Miyashiro for their assistance in meticulously editing and formatting each chapter. Due to the generous support of the Academy of Korean Studies 2023 Scholarly Book Publication Grant (AKS-2023-P004), this volume has been successfully published as an openaccess book.

## **Contents**



## **Editor and Contributors**

#### **About the Editor**

Mary Shin Kim (PhD University of California, Los Angeles) is Professor in the Department of East Asian Languages and Literatures at the University of Hawaiʻi at Mānoa. Her research areas are interactional linguistics, pragmatics, and second language pedagogy. She examines language use in social interaction — how speakers collaboratively construct, intersubjectively understand, and mutually respond to their actions in real time. She further investigates how to apply these findings to language teaching. Her work has appeared in *Discourse Processes*, *Discourse Studies*, *Journal of Pragmatics*, and *Research on Language and Social Interaction*. Mary is also a co-author of *Integrated Korean Workbook* (Beginning and Intermediate Levels).

#### **Contributors**

**Eun Young Bae** Department of Korean Language Education & Korean Language Education Research Institute, Seoul National University, Seoul, South Korea

**Lucien Brown** Korean Studies, Monash University, Melbourne, Victoria, Australia

**Hanwool Choe** School of English, The University of Hong Kong, Hong Kong

**Seunggon, Jeong** Department of Korean Language Education, Seoul National University, Seoul, South Korea

**Jaehyun Jo** Department of Asian Languages and Cultures, Rutgers University, New Brunswick, USA

**Sujin Kang** The Prompt Company, Seoul, South Korea

**Jieun Kiaer** Asian and Middle Eastern Studies, Oxford, England, United Kingdom

**Ji Young Kim** Communication Program, School of Communication & Information, College of Social Sciences, University of Hawaiʻi at Mānoa, Honolulu, USA

**Loli Kim** Asian and Middle Eastern Studies, Oxford, England, United Kingdom

**Mary Shin Kim** Department of East Asian Languages and Literatures, University of Hawaiʻi at Mānoa, Honolulu, USA

**Soung-U Kim** Dipartimento Asia Africa e Mediterraneo (DAAM), Università di Napoli L'Orientale, Italy

**Alfred W. T. Lo** Asian and Middle Eastern Studies, Oxford, England, United Kingdom

**Miriam A. Locher** Department of Languages and Literatures, University of Basel, Basel, Switzerland

**Thomas C. Messerli** Department of Languages and Literatures, University of Basel, Basel, Switzerland

**Tyler Miyashiro** Department of East Asian Languages and Literatures, University of Hawaiʻi at Mānoa, Honolulu, USA

**Mee-Jeong Park** Department of East Asian Languages and Literatures, University of Hawaiʻi at Mānoa, Honolulu, USA

**Yujong Park** English Literature and Language Department, Sungkyunkwan University, Seoul, South Korea

**Gahye Song** The World Languages Center, Northeastern University, Boston, Massachusetts, USA

## **Introduction: K-Politeness Across Online and Offline Interactions**

#### **Mary Shin Kim<sup>1</sup>**

This book investigates how Korean politeness ("K-politeness") is perceived, practiced, and negotiated across diverse interactional contexts and communication platforms, both online and offline. The research examines K-politeness from not only the standpoint of Korean speakers but also from the vantage point of K-wave fans, viewers, and learners of Korean language and culture. The research offers a comprehensive and emic perspective on K-politeness, incorporating multidisciplinary and multimodal analysis.

#### **1 K-Politeness**

Korean features one of the most intricate systems of honorifics. When delving into the realm of Korean politeness, naturally, research concentrates on scrutinizing these elaborate honorific linguistic resources. There have been many studies on Korean speech styles, honorific expressions, and address and reference terms: examining their morphosyntactic and lexical aspects, as well as understanding how speakers deploy these resources based on factors such as age difference, social status, or the level of interpersonal familiarity (Brown 2015; Sohn 1986; Sohn 1999). Expanding upon the groundwork laid by previous studies, this edited volume assembles the most current research on Korean (im)politeness examining it as a set of dynamically interactive practices, drawing from multidisciplinary and multimodal perspectives encompassing a diverse array of interactional contexts and communication platforms.

The volume first examines how Korean language speakers perceive, practice, and utilize politeness or impoliteness as interactional tools or practices during daily interaction, online and offline. The studies not only include intimate interactions between family members and friends, but also institutional interactions between doctors and patients, business vendors and business clients, talk show hosts and their guests, and between politicians via public discourse. The volume also examines an outsiders' view of Korean (im)politeness. As the number of K-wave viewers and fandom rapidly spreads globally, increasing numbers of fans, learners, and consumers of Korean language and culture gather online and offline (Locher & Messerli, 2020). The studies include discussion of the perception by Korean (im)politeness of K-wave fans and Korea language learners, and how they understand, identify, and relate to Korean (im)politeness (Brown, 2011; Byon 2004; Locher 2020). The perception of fans and learners is undeniably shaped by Korean media, and as a result, the role of media is also explored within this volume.

e-mail: maryskim@hawaii.edu

<sup>1</sup> M. S. Kim ()

Department of East Asian Languages and Literatures, University of Hawaiʻi at Mānoa, Honolulu, Hawaiʻi, USA.

This exploration encompasses how media reflects and influences the way speakers perceive and engage in Korean (im)politeness.

As we communicate through an increasing number of platforms on a daily basis, the volume also examines these realities by not only looking at face-to-face interaction, but also interaction through instant text messages, chat boxes in livestreaming sites, online chat boxes with business vendors, as well as other communication channels.

This volume provides new insights into Korean (im)politeness with contributions from researchers from different disciplines, including communications, sociolinguistics, conversation analysis, discourse analysis, and pragmatics. Although the disciplines and methodologies may vary, all the studies are based on empirical research drawn from interactions between participants or interviews (Locher & Watts, 2005). Taking an emic and embodied approach, these studies examine how speakers identify, make sense of, and engage in (im)politeness by drawing upon verbal and non-verbal resources (Brown & Winter, 2019).

#### **2 This Volume: K-Politeness Across Online and Offline Interactions**

#### *What do Korean speakers know about politeness?* (**Soung-U Kim and Lucien Brown**, Chapter 2)

Drawing on interviews with Korean participants across different age groups, this chapter discusses what politeness means to Korean speakers, and how they conceptualize commonly known Korean politeness metaconcepts. The survey showed how speakers perceive and talk about "politeness" and related metaconcepts and metalexemes. Prior accounts of politeness metaconcepts in Korean have typically focused on vertical and hierarchical aspects of politeness. According to this study, across many different metaconcepts, only *yeyuy paluta* 'possess correct civility' is perceived predominantly as a hierarchical concept. Most participants viewed *concwunghata* 'be respectful' and *mwusihata* 'be disrespectful' as horizontal, while *(pwul) chincelhata* 'be (dis)courteous' is sensitive to social distance rather than hierarchical power, since it only applies to interactions with strangers. In contrast to prior studies which viewed *concwung* 'respect' as "elder respect," this study showed that it was primarily defined as demonstrating respect for someone as a human being and treating others as one would like to be treated oneself. Moreover, participants viewed emotional attunement as an important underlying motive for politeness, instead of face. Showing care for others' emotions and the concept of comfort (*phyenhata*) are key to achieving politeness.

The interviews further revealed that participants viewed (im)politeness as an inherently multimodal concept. Participants viewed politeness as an embodied attitude (*thayto*) that was employed to maintain emotional attunement during interaction. Respecting others (*concwung*) by acknowledging (*inceng*) their emotional needs and social position, all while recognizing their universal value as human beings, is essential in all interactions. In the line with other chapters in this volume which aim at questioning and enriching conventional views on Korean (im)politeness, Kim & Brown's study complexifies conventional portrayals of Korean politeness as a mainly vertical and hierarchy-oriented notion by taking into account participants' universalist (i.e., essentially horizontal) and inherently multimodal understandings of (im)politeness.

#### *Korean honorific speech level markers as contextualization cues in family instant messages* (**Hanwool Choe**, Chapter 3)

While Chapter 2 examined what Korean speakers thought of politeness through interviews, Chapter 3 analyzed how Korean speakers actually use honorific speech level markers – deferential and polite – in their online family interaction via KakaoTalk, an instant messaging application. Using interactional sociolinguistics, the study focused on how Korean family members construct discursive meanings of the speech level markers. The use of honorific speech level markers in family group chats neither always adheres to politeness nor is necessarily regulated by traditional social factors such as age, roles, and status. The data illustrates how the deferential and polite speech level markers, as contextualization cues, can serve as four different markers. To be specific, the honorific speech level markers are used as 1) an egalitarian marker between a married couple of the same age, whereas they serve as 2) a face-saving marker, 3) a footing marker, or 4) an affective marker between mothers-in-law and daughters-in-law, especially when they share everyday photos and videos of children of the family. This study contributes to showing how the honorific speech level markers signal the ways in which family members manage and negotiate power and solidarity, while also constructing and performing their family-related identities, in family group chats.

#### *Addressee honorifics as an interactional resource for socialization in Korean adult-child interaction* (**Eun Young Bae, Gahye Song, and Seunggon Jeong**, Chapter 4)

While Chapter 3 investigated how honorifics are used among adult family members, Chapter 4 examined how they are used in adult-child interaction. The study in particular analyzed adults' use of Korean addressee honorifics to children, such as the deferential sentence ender -*supnita* and the polite sentence ender -*yo* (sometimes accompanying the subject honorific suffix *-si*). These are normally considered as grammatical resources for displaying deference and respect. However, in adult-child interaction, adults use these for different practical reasons. Deploying multimodal discourse analytic and language socialization frameworks, the study reveals that addressee honorifics serve as a tool for socializing children into socio-moral values of Korean society. They are recurrently observed in three contexts, giving compliments, showing gratitude, and issuing directives. In these interactional environments, addressee honorifics are used alongside semiotic resources to evaluate children's behaviors and draw attention to the locally meaningful social roles, norms, and expectations. By doing so, it provides children with an opportunity to reflect on their behaviors within the proposed frame of interpretation. This study contributes to uncovering creative or performative functions of addressee honorifics for fostering social awareness, social responsiveness, and courtesy in children. Such practice is significant in becoming competent members of Korean society who think, feel, and act in accordance with Korean cultural norms and expectations.

#### *Solidarity through negotiated interactional identities in Korean* (**Mee-Jeong Park**, Chapter 5)

Chapter 5 also directed its attention toward various Korean speech styles and their utilization by newly acquainted Korean interlocutors during media discourse. Selecting a speech style in Korean is far from simple; it entails a multifaceted process. Participants in a conversation consistently assess the dynamics of their interaction and engage in negotiations to establish a

suitable level of rapport. This process is facilitated by adapting their choice of address terms and/or speech style to conform to the identity they have recently constructed. An examination of a large corpus of TV talk show and reality show data presents three frequently encountered agebased categories of negotiated identities: (a) the identity of 'friends' (*chinkwu*) when both newly acquainted interlocutors are of the same age; (b) the identity of 'siblings,' encompassing terms like *hyeng/oppa* (older brother) or *nwuna/enni* (older sister) when the two interlocutors differ in age; (c) 'senior' (*senpay*) or 'junior' (*hwupay*) identities, which are adopted if they share the same professional field or have attended or currently attend the same school. This practice of negotiating interactional identities is aimed at preventing conflicts that may emerge when they fail to achieve a shared understanding of intimacy levels while engaging in routine activities.

#### *"What does hyung mean please?": Moments of teaching and learning about Korean (im)politeness on an online streaming platform of Korean TV drama* (**Miriam A. Locher and Thomas C. Messerli**, Chapter 6)

While previous chapters examined Korean (im)politeness through the lens of L1 Korean speakers, this chapter examines how viewers or fans of Korean dramas perceive Korean (im)politeness. The study examined the streaming platform Viki.com, which allows fans to translate the Korean original in subtitles and to interact with each other in written comments on the episodes they stream. According to the study, an average of 2.9 scenes per episode include moments of relational work. The significance of relational work negotiations in Korean society cannot be ignored by translators and viewers. While translators act as cultural mediators in their orientation towards the source culture, viewers actively engage and grow interest in Korean culture through commenting. The study shows that viewers do pick up on (im)politeness negotiations in linguistic (e.g., Korean borrowings, address terms) and embodied, multimodal form (i.e., bowing and lowering your gaze). Such discussions can be used as a starting point for gaining awareness of the complexities of the Korean (im)politeness system. Thus, the scenes from the drama which show relational work and viewers' comments on relational work provide evidence that there are moments of teaching and learning about Korean (im)politeness in this online fan community.

#### *"Koreans are always nodding or bowing": K-wave fandom's perception and learning of nonverbal politeness* (**Jieun Kiaer, Loli Kim, and Alfred W. T. Lo**, Chapter 7)

Chapter 7 explores how K-wave language learners develop perceptions and understanding of politeness through their exposure to Korean dramas and reality TV shows. The study utilizes a multimodal qualitative design to explore non-verbal politeness, and the subjects of the study were learners in a multilingual Korean classroom in the United Kingdom. The learners completed multiple tasks and activities, such as demographic surveys, interactive brainstorming, K-film roundtable discussions, and the intercultural awareness reflective journal. The findings reveal that non-verbal politeness is perceived differently by different levels of learners. Beginning level learners lacked an understanding of Korean pragmatics and the importance of semiotic resources. Intermediate participants had a better grasp of non-verbal behavior, but they still lack nuance, whereas advanced participants understood it well. Since pragmatics and politeness are essential for successful communication in Korean, students should learn about

these non-verbal features from the outset. The chapter further discusses how integrating media in language learning could be particularly advantageous for learning East Asian languages.

#### *Negotiating age, epistemic stance, and category memberships in Korean talk* shows (**Mary Shin Kim and Jaehyun Jo**, Chapter 8)

Chapter 8, following the insights from Chapter 5, which underscores the significance of age in Korean politeness, delves into the practical applications of age and its relevance to speakers during their interactions. Utilizing membership categorization analysis, this study identifies a categorial practice associated with age within a collection of data segments derived from various talk show interviews. The results of the analysis reveal that speakers frequently reference or imply age when conveying their epistemic stance, whether it involves asserting or disclaiming their rights to knowledge on the topic under discussion. Age is used as a justification or authorization for possessing or lacking certain knowledge or information. Moreover, the study sheds light on how speakers classify both themselves and others into age-based categories (e.g., the young, the old, *acessi*, grandpa) based on their epistemic status or rights concerning specific matters. Notably, these categories are not tied to the speakers' chronological age or life stages; instead, they are spontaneously constructed within the context of the interaction and can be subject to challenges, resistance, and negotiation by the speakers involved. This study demonstrates how speakers utilize age as a tool in their interactions to negotiate their epistemic positions and category memberships. The findings of the study further unveil an intriguing contrast in how politeness is practiced by Korean speakers regarding age. Korean society holds a deep-rooted tradition of demonstrating respect and politeness towards older individuals. However, an alternate trend surfaces when observing media discourse.

#### *Exploring frames and negativity strategies in the news during an election campaign* (**Ji Young Kim**, Chapter 9)

This chapter examines the practice of impoliteness as a strategic campaign resource in Korean political discourse. Negativity is the primary focus of the study, which has become a popular campaign strategy in today's election contexts due to its appeal to the media and the public. Negative messages have been shown in previous studies to attract more media attention and engage the public. This study examined the news frames and negative remarks in the news coverage of two mainstream newspapers during Korea's 2022 presidential election. The results revealed that the frame types change over time. Before the official campaign period, announcements, investigations, and personal stories were the most frequent election frames. During the official campaign period, the election frames most frequently used were strategies, public engagement, issue position, conflict, race, and critique. Negative politeness occurred more often closer to the election. Particularly, newspapers reported a campaign message when it explicitly associated a negative aspect to the opponent. One of the positive impoliteness strategies that can harm a person's positive face desire is name-calling, which was frequently reported. The Korean news coverage frequently used ideological negative language, such as populism, communism, feminism, or political revenge, during the presidential election campaign.

#### *Facework in patient requests for treatment recommendations in Korean medical interactions: The use of main clause omission* (**Yujong Park**, Chapter 10)

Chapter 10 examines how Korean speakers carry out potentially face-threatening actions during medical interactions. Doctor-patient interaction is characterized by a knowledge-based asymmetry or epistemic imbalance. Accordingly, patients' act of proposing a specific treatment plan can be considered a highly face-threatening act which imposes on the doctor's authority. The study shows how patients strategically design their turns to manage these delicate, problematic actions. Drawn from a corpus of videotaped primary care visits collected from Korean medical practitioners, this study examines politeness as a social practice in the collected interactions adopting a conversation analytic approach as applied to institutional interactions. The analysis indicates that patients frequently use main clause omissions to advocate for a specific treatment plan or request a specific prescription. Patients can hint at a problematic action instead of fully articulating the request due to the absence of main clauses. With the epistemic asymmetry present, patients can pursue the matter and insist without being overly demanding. The use of face and facework concepts can effectively account for this practice. In terms of understanding politeness in interaction, the analysis highlighted the importance of the two interrelated analytical notions, incrementality (how patients produce their requests incrementally in light of the doctors' response) and sequentiality (current turns are always understood relative to prior and subsequent talk).

#### *Challenges in managing Korean online service requests and complaints via business chat* (**Mary Shin Kim, Sujin Kang, and Tyler Miyashiro**, Chapter 11)

The last chapter discusses how Korean speakers manage and negotiate face-threatening situations in business context. The chapter in particular examines business-to-business online communication via chat. Restaurants and food-related businesses initiate online chats, reaching out to address delivery issues or delays. The recipient of these requests is a delivery application company, which occupies a unique position as the intermediary. This company must not only respond to its food industry clients but also transmit their requests to local delivery offices. A collection of 65 online chats between business clients and service provider agents unveils that the course of an initial request can branch out and intensify, leading to increased frustration and criticism, depending on how both parties respond to each other. Within these service-related exchanges, clients show a preference for clear and direct language, ensuring the efficient fulfillment of their service requests. This inclination toward explicit reassurance regarding the swift resolution of delivery issues takes precedence over the customary use of restrained and indirect expressions, often associated with politeness, by the agents in their interactions bridging clients and the delivery dispatch office. Agents adapt their response strategies by enhancing their sense of agency, certainty, and immediacy when fulfilling requests. These response approaches align with client preferences. The research findings suggest that what is deemed appropriate or polite in face-to-face interactions may differ in the realm of online business-to-business communication.

#### **References**


Byon, A. S. (2004). Learning linguistic politeness. *Applied Language Learning*, *14*(1), 37–62.


Sohn, H-M. (1999). *The Korean Language*. Cambridge University Press.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **What Do Korean Speakers Know About Politeness?**

#### **Soung-U Kim and Lucien Brown<sup>1</sup>**

**Abstract** This chapter explores how Korean speakers understand and talk about "politeness" and related metaconcepts. Data is drawn from qualitative interviews with 20 Korean speakers (10 students in their 20s and 10 professors aged 40 and above). We asked them to provide their own personal definitions of six metaconcepts: *yeyuy paluta* 'possess correct civility', *yeyuy epsta*  'lack civility', *concwunghata* 'be respectful', *musihata* 'be disrespectful', *chincelhata* 'be courteous', and *pwulcincelhata* 'be discourteous.' We analyse how the participants defined these six concepts, and then look at four other important metalexemes that emerged in the data: *phyenhata* 'comfortable', *thayto* 'embodied attitude', *paylye* 'consideration' and *inceng* 'recognition'. Through analysis of these different concepts, we are able to construct an emic (participant-oriented) picture of what politeness means across two different generations of Korean speakers. We conclude by discussing how the results compare to previous descriptions of politeness in Korean, and also to the models of politeness offered in previous politeness theories.

**Keywords** Politeness Metapragmatics Politeness metaconcepts Korean Politeness Metalinguistic knowledge Politeness and multimodality Qualitative interviews

<sup>1</sup>S-U. Kim

L. Brown () Korean Studies, Monash University, Melbourne, Victoria, Australia e-mail: Lucien.Brown@monash.edu

Dipartimento Asia Africa e Mediterraneo, Università di Napoli L'Orientale, Napoli, Italy. e-mail: soungu.kim@unior.it

#### **1 Introduction**

Politeness is not only something that we perform during interaction, but also something that we think and talk about (Kádár & Haugh, 2013). In every language and culture, speakers have their own lexicon for talking about politeness and related concepts such as civility, respect, kindness, rudeness and so forth. These lexical items are to various extents culture specific, and may have quite different semantic mappings in different languages (Pizziconi, 2007). They may also be understood in different ways by people from different social groups and generations (Fukushima & Haugh, 2014). Investigating these metapragmatic understandings of politeness across different cultures is increasingly becoming recognized as an important goal for politeness research given the shift towards researching politeness from an emic (i.e., participant-oriented) perspective (Eelen, 2001). As noted by Verschueren (1999, p. 196), "there is no way of understanding forms of behaviour without gaining insight into the way in which the social actors themselves habitually conceptualize what it is they are doing."

In the context of Korean, various politeness-related metaconcepts have been proposed in previous literature (see Section 2.2 for full discussion). The term politeness has frequently been translated into Korean as 공손 *kongson* (see L. Brown, 2011b), whereas the concept of face has been equated with 체면 *cheymyen* (Lim & Choi, 1996). Studies have also talked about the importance of 존대 *contay* 'deference' (Hwang, 1975; Sohn, 1986), and the centrality of showing respect to status superiors (윗사람 *wissalam*) according to neo-Confucian slogans such as 경로사상 *kyenglosasang* 'respecting the elderly' (Yoon, 2004, p. 198). However, to date, discussions of politeness metaconcepts in Korean have relied almost exclusively on the intuitions of individual scholars. What is therefore missing is a study that asks non-expert speakers of Korean for their insights into how they understand politeness.

The current study addresses this gap in the literature by carrying out interviews with two groups of Korean speakers (university professors and university students) to explore their understandings of politeness in Korean. Our analysis focusses on how these speakers understood six interrelated metaconcepts: *yeyuy paluta* 'possess correct civility', *yeyuy epsta* 'lack civility', *concwunghata* 'be respectful', *musihata* 'be disrespectful', *chincelhata* 'be courteous', and *pwulcincelhata* 'be discourteous.' We also discuss other politeness-related concepts that emerged during the conversations that we had with these speakers. Our primary goal is to establish perspectives of what politeness means to Korean speakers, and, secondarily, to show how these perspectives might contrast with the claims made by researchers in previous studies.

#### **2 Background**

Before discussing the data collection, we pause briefly to set up the background of the study. Section 2.1 explains the notion of politeness metaconcepts, with a subsequent overview of previous research on Korean politeness metaconcepts in Section 2.2. In Section 2.3, we summarise the core Korean metaconcepts that we focused on during data collection.

#### *2.1 Politeness Metaconcepts*

Early theories of (im)politeness that adopted a universal perspective assumed that "politeness" , or a concept closely analogous to it, existed in all cultures. Brown and Levinson (1987, p. 60), for instance, began their universal account of politeness by noting "extraordinary parallelism in the linguistic minutae … in quite unrelated cultures" which relate to the isolated motive of politeness. In such early proposals, two properties underlying politeness are assumed to be universal across cultures: rationality and face. The latter, defined in terms of "public self-image", became a key concept in pragmatics and subjected to various analyses and critiques (see Haugh & Bargiela-Chiappini, 2010).

However, the idea that politeness was a universal concept that existed with little variation across different cultures was soon challenged, as did the ideas that politeness relied on individual rationality and face. Most vocal in their critiques were scholars of Japanese including Ide (1989) and Matsumoto (1988) who claimed that the focus on individual rationality in the Brown and Levinson framework was incompatible with languages that feature grammaticalised politeness systems, the usage of which was controlled by social convention rather than individual rationality and face. Ide (1989) proposed that the practice of using polite behaviour according to social convention corresponded to the Japanese concept of *wakimae* 'discernment', which involved the appropriate understanding and expression of one's place or role in given situations (p. 230).

Although the idea that Japanese politeness relied on social convention more than strategy was criticized in subsequent studies (e.g., Pizziconi, 2003), the work of Ide (1989) and Matsumoto (1988) laid the foundations for culture-specific investigation of emic politeness related concepts. Subsequent studies went on to propose a vast number of culture-specific politeness concepts for Japanese (e.g., Haugh, 2005 - *basho* 'place'), Chinese (Gu, 1990 - *limào* 'polite appearance'), Thai (Intachakra, 2012 - *k <sup>h</sup>waːmkreːŋaj* 'fear of hearts'), and Persian (Koutlaki, 2002 - tæ'arof 'ritual politeness'), among many others. More broadly, politenessrelated practices started to be analysed as culturally-embedded rituals enacted to maintain the moral order in social interaction (Kádár, 2017).

In recognition of the need to further explore the emic ways that politeness is perceived and talked about across different cultures, a number of recent studies have adopted empirical methods to establish what politeness means to speakers of different languages, and to tap into speaker intuitions of the meanings of politeness metalexemes. Pizziconi (2007) used a questionnaire that asked British English and Japanese participants to judge the level of similarity of ten politeness metalexemes in their respective languages. The results showed important crosscultural differences, for example, 'friendliness' was homologous with politeness in British English, whereas Japanese speakers associated politeness with modesty and restraint.

Our own paper belongs to an emerging group of studies that have resorted to interviews to investigate speakers' metapragmatic knowledge of politeness. While Ogiermann and Suszczyńska (2011) interviewed speakers of Polish and Hungarian to explore how notions of politeness were changing after the fall of the iron curtain, Su (2019) explored the conceptual understanding and semantic field of *limào* 'politeness' among Taiwanese living in China. Moreover, Fukushima and Haugh's (2014) study explored the metaconcepts of "attentiveness, empathy and anticipatory inference" in Japanese and Chinese (see also Fukushima, 2020).

#### *2.2 Politeness Metaconcepts in Korean*

Traditional descriptions of politeness in Korean (Dredge, 1983; Hwang, 1975, 1990; Sohn, 1986) claim that Korean has a clear distinction between "politeness" (which is typically translated into the Korean term 공손 *kongson*) and "deference" (존대 *contay*) (see also A. H.-O. Kim, 2011 for discussion). The former basically refers to Brown and Levinson's (1987) concept of politeness based on rationality and face, which is described as "an open pattern of language usage" including "prosodic means, modal elements, softening adverbials, etc" (Sohn, 1986, p. 411) that speakers can employ strategically. The latter, *contay* 'deference' basically refers to the use of honorifics, which Sohn (1986, p. 411) notes is "a closed, language-specific system". The distinction between politeness and deference is exemplified in these studies by examples such as the following that are claimed to show that these two concepts work as independent systems (Hwang, 1990, p. 48):


b. *inswu-ya, ikes com chiwe cwul-lay?* Insoo-VOC this please clear BEN-SUG 'In-soo, will you put this away for me?'

Sentence (1a) contains subject honorifics, but is a bold imperative. According to Hwang, this sentence is "deferential but impolite". In contrast, (1b) contains no honorifics, but is a benefactive suggestion rather than an imperative. This makes it "non-deferential but polite." Hwang (1990, p. 42) sees politeness as "a matter of speaker's psychology", but deference as "a matter of social code which is imposed upon the participants in communicative interactions."

Previous studies model the importance of honorifics and "deference" as being embedded in the hierarchical structure of Korean society. Yoon (2004, p. 194) points out that "Koreans believe they are not equal in status, either in the family or in other social groups to which they belong", and that Korean ideologies regarding social relations focus on the need to exhibit respect when dealing with elders. Respecting elders is said to be tied up with pervasive neo-Confucian social slogans such as 경로사상 敬老思想 *kyenglosasang* 'respecting the elderly' (Yoon, 2004, p. 198) and 장유유서 長幼有序 *cangyuyuse* 'the old and the young know their place' (Lee & Ramsey, 2001, p. 267). Yoon (2004, p. 194) notes that "Koreans believe that people are not equal in status, either in the family or in other large social groups to which they belong", whereas Brown (2011a, p. 80) notes that disagreeing with or causing discomfort to elders is taboo in Korean culture. With this emphasis on the importance of respecting elders, previous studies rarely if ever mention how politeness or respect might work in horizontal relationships, or for the more powerful or older party in hierarchical relationships. As we will discuss later on, our study witnessed an emergence of horizontal layers of understanding politeness by our interviewees, which may either point to a need to differentiate views such as Yoon's (2004), or to the need to study possible societal changes that may have occurred during the last decades.

Scholars working on Korean followed Brown and Levinson (1987) in seeing face, or 체면 *cheymyen*, as the underlying motive for politeness (De Mente, 1998; Lim & Choi, 1996; Oak & Martin, 2000). According to Lim (1995), *cheymyen* involves the adherence to form and societal expectations, which are identified as being key Confucian values. Behind the need to conform to social expectations is a high concern for how others perceive one's behaviour, or "the eyes and ears of others" as Kim (2001, p. 123) puts it.

In sum, studies that have described Korean politeness metaconcepts have foregrounded the division between strategic politeness (*kongson*) and deference (*contay*), as well as the importance of respecting elders and adhering to social expectations. However, it is unclear whether the distinction between politeness and deference is substantiated in the actual ways that Korean lay people understand politeness, nor whether they view elder respect and social conformity as the driving forces behind politeness. In fact, the terms *kongson* and *contay* are both rather infrequent in everyday conversation (see L. Brown, 2013) and there is no evidence beyond the invented examples of linguists (see example 1, above) that the concept of *kongson* is separated from the use of honorifics. There is therefore a need for us to consult with Korean speakers and explore the ways in which they talk about politeness.

#### *2.3 Focal Concepts Explored in This Study*

In order to address the need to explore Korean politeness metaconcepts from the perspective of Korean language users, we carried out interviews with two groups of Korean speakers (university professors and university students; see Section 3 below). We asked them to provide their own personal definitions of six focal metaconcepts, which we will introduce in this section. These six metaconcepts formed three pairs of terms, where one of the two terms related to politeness and one was the lexical opposite that referred to impoliteness (see Table 1). These six metaconcepts were selected since they are all frequently used terms in everyday speech, and since they reflect three distinct dimensions of (im)politeness.


#### **Table 1** Focal metaconcepts we investigated

The first two metaconcepts that we explored were *yeyuy paluta* 'possess correct civility' and the closest lexical opposite *yeyuy epsta* 'lack civility'. These commonly used terms were chosen for investigation since they appeared to us to represent the idea of upkeeping social norms for the purposes of general courtesy. The term *yeyuy* captures the idea of politeness as decorum (Jung, 2005), whereby both senior and juniors contribute to the preservation of socially desirable norms of interaction.

We then explored the notion of respect via the metalexemes *concwunghata* 'be respectful' and *mwusihata* 'be disrespectful'. These two items were selected since respect is commonly discussed in previous accounts of Korean politeness (see L. Brown, 2011a), and these two lexemes are frequently used in everyday speech. Previous studies tend to equate respect in Korean with displaying deference to status superiors, which is also captured in the dictionary definition of *concwunghata* as 높이어 귀중하게 대하다, 'elevate one's counterpart and treat them as very important' (which in turn is a rather literal translation of the Sino-Koreanic origin of this word, 尊重) 1 . As for *mwusihata*, in addition to a lack of respect for a superior's rank, Kim and Brown (2019) note that this term also refers to failure to maintain a certain minimum level of respect for someone's position, even if their social position is relatively low.

Finally, *chincelhata* 'be courteous' and *pwulcincelhata* 'be discourteous' are high frequency lexemes that refer to politeness as kindness, friendliness or hospitableness. According to Yang et al. (2013), *chincel* is a multimodal phenomenon comprising of the control of speech, facial expression, behaviour and appearance, which characterizes service industry interactions. The improvement of this public face of politeness has been encouraged on the national level by public campaigns such as the 2015 친절한 대한민국 캠페인 *Chincelhan Tayhanminkwuk Campaign, 'The Friendly South Korea Campaign'* (Han, 2015), which utilized the slogans "Korea smiles on you."

#### **3 Method**

In this Section, we explain our data collection methodology (Section 3.1), present some important details about our participants (Section 3.2) and explicate our analytic procedure (Section 3.3).

#### *3.1 Data Collection*

We conducted semi-structured interviews with 20 Korean speakers at a large university in Seoul.<sup>2</sup> Interviews have been widely used in previous studies exploring metapragmatic understandings of politeness, including Chang and Haugh (2011), Fukushima and Haugh (2014), and Fukushima (2020).

The interviews were conducted by the first author and recorded using a Canon XA-11 camera with a top-mounted, directional Rode NTG-2 mic, accompanied by parallel audio recordings on a Zoom H2n recorder. The average length of the recordings was 44 minutes and 44 seconds.

Our interviews relied on a questionnaire for semi-structured interviews, which was developed to include three thematic areas of questions. In the first, we asked general questions regarding factors that were important in human interaction in order to encourage participants to talk about politeness using their own terms. In the second section, we invited participants to

<sup>1</sup> See Phyocwunkwuketaysacen (Standard Korean Language Dictionary): https://stdict.korean.go.kr [last accessed 2022-07-22]

<sup>2</sup> A sub-set of the data presented in this paper was previously analysed in Brown et al. (2022) where we looked at just two of the participants, and focussed on the embodied ways that participants enacted their politeness narratives rather than on their understandings of metaconcepts).

reflect on their personal understandings of the six politeness metaconcepts described above in Section 2.3, using the following questions as a guide:


The first author of this paper took on the role of the interviewer. The objective was to create a conversational atmosphere, during which all questions of the questionnaire were supposed to be covered yet leaving room for participants to steer the interview into individual directions, highlight what was meaningful to them, and to engage in deep description of their metapragmatic understanding. Accordingly, our questionnaire was taken as a rough guide for orientation for the interviewer, rather than a strict, tasked-based template.

Finally, in the third section we asked about their perceptions of politeness variation across different generations, genders and dialect areas. In the current paper, we focus on the answers provided in Section 2. A full list of interview questions can be found in the Appendix.

#### *3.2 Participants and Researchers*

All 20 participants worked or studied at a large university in Seoul, with 10 of them being students at this university and 10 of them being professors. We collected data from two different generations of speakers in order to include possible cross-generational differences in understandings of politeness, inspired by Fukushima and Haugh (2014).

The average age of the students was 21 (range: 20 to 24) and for the professors it was 51 (range: 44 to 62). There were equal numbers of males and females in each group. All participants self-reported that they were speakers of Standard South Korean, including two speakers who said that they also spoke a regional dialect (one student spoke Jeolla dialect and one professor spoke Gyeongsang dialect). All of the professors had experience living overseas (eight in the US, one in France and one in China), as had four of the students (four in the US and one in New Zealand). In the subsequent analysis, participants are referred to by "S" for "student" and "P" for "professor" followed by their participant number (e.g., S1 = participant 1, who is a student).

The other participant in the data is the interviewer, who is the first author of the paper (referred to herein by his initials "SC"). Although the analysis that follows will focus primarily on the productions of the interviewees, we acknowledge that these productions are in fact coconstructed by the interviewer, and that the methodology and analysis are unavoidably coloured by the academic and personal biases of our research team. Both authors see politeness as multiplicious and contested, and metapragmatic talk as a type of situated and stylized performance, and this perspective was inevitably reflected in the kinds of questions we included in the interview. Whereas the second author is a pragmatician who speaks Korean as a second language, the first author is a documentary and descriptive linguist and a native speaker of Korean (raised and educated in Germany).

#### *3.3 Analysis*

The interviews were transcribed in ELAN (Version 5.7; 2019), which is a tool for the multimodal annotation of video resources.

The data was then coded thematically, adopting an inductive methodology based on grounded theory (Auerbach & Silverstein, 2003). This approach involves an iterative, bottom-up procedure whereby so-called "repeating ideas" in the data are coded and increasingly broader themes and concepts are abstracted from them in an inductive approach.

For the current paper, we focus on two types of codes that emerged in the data. First, we look at codes that relate to the six focal metaconcepts of this study: *yeyuy paluta* 'possess correct civility', *yeyuy epsta* 'lack civility', *concwunghata* 'be respectful', *musihata* 'be disrespectful', *chincelhata* 'be courteous', and *pwulcincelhata* 'be discourteous' (see Section 2.3 above). These codes mostly emerged in the second section of the interview, during which we explicitly asked participants to define these terms. Second, we look at codes that relate to four other metaconcepts that occurred frequently during the interviews and which emerged as key ideas in the way that our participants understand politeness: 편하다 *phyenhata* 'comfortable', 태도 *thayto* 'embodied attitude', 배려 *paylye* 'consideration' and 인정 *inceng* 'recognition'. We grouped together codes that related to these terms, and abstracted from these codes the underlying ways that our participants understood them. For the six focal metaconcepts, given the importance of hierarchy and respect for elders noted in previous research on Korean politeness, we began by quantifying how many of our participants understood each concept as hierarchical or vertical.

#### **4 Data Presentation**

We begin by presenting findings for the six focal metaconcepts (Section 4.1) in pairs (politeness related term and corresponding impoliteness term). Then, in Section 4.2, we examine the other politeness-related terms that emerged in the data.

#### *4.1 Focal Metaconcepts*

#### **4.1.1** *Yeyuy Paluta* **'Possess Correct Civility' and** *Yeyuy Epsta* **'Lack Civility'**

Twelve of the twenty participants (seven students and five professors) viewed *yeyuy paluta* 'possess correct civility' as a vertical concept that is performed towards elders or superiors, whereas five saw it as horizontal, and three saw it as working both vertically and horizontally. Interestingly, *yeyuy epsta* 'lack civility' was viewed in less hierarchical terms – only seven of the participants viewed it as vertical (five students and two professors), whereas nine saw it as horizontal, and two as both (a further two participants failed to provide a definition for this item). Participants therefore do not seem to view *yeyuy paluta* 'possess correct civility' and *yeyuy epsta* 'lack civility' as exact conceptual opposites.



**Table 3** Understandings of *yeyuy epsta*

\* = Two participants did not provide a definition for *yeyuy epsta*

Participants saw *yeyuy paluta* as involving being careful (조심하다) and controlled with your own behaviours in order to show respect (존중) to others and avoid making them feel bad (언짢다). Participant P2 described *yeyuy paluta* as 'exhibiting self-censorship' (스스로 센서십을 발휘하는 것), while Participant S13 conceptualized it in terms of suppressing certain behaviours, such as yawning or taking out your phone when interacting with a status superior. S20 noted that exercising *yeyuy* was uncomfortable and required perseverance and endurance to maintain: 'when we say that someone is *yeyuy paluta*, we mean that someone knows how to endure things (그래서 더 예의 바르다라는 표현을 하면은 그 얘긴 그냥 더 참을 줄 아는 거고).' P6 explained how 예의 바르다 arises when you listen respectfully (경청하다) to what the other person is saying and respond appropriately (적절하게 호응을 하다). On the other hand, *yeyuy epsta* is a result of not knowing when to speak and when to listen (participant S9), inconveniencing other people (participant S14) and hurting the feelings of others (participant S1).

As noted above, whereas some participants viewed *yeyuy paluta* as involving a general consideration for other human beings, others connected it specifically with elder respect, with this hierarchical understanding of the concept being more common among the students. Participant S1, for instance, explicitly linked *yeyuy paluta* to the use of honorifics and to placing yourself beneath status superiors (자신을 낮춘다). By using honorifics, elders are said to feel more comfortable (편안하다). On the other hand, as noted by S5, reckless behaviour towards elders, including using inappropriate language results in *yeyuy epsta*.

Although half of the professors also defined *yeyuy paluta* in terms of showing respect towards superiors, they noted that status superiors as well needed to show *yeyuy* towards their juniors – an idea that never occurred in the student data.<sup>3</sup> P17 described *yeyuy epsta* as a lack of respect (존중) towards younger generations, or others who are lower in status than oneself:

(1) P17 (22:50) 1 내가 더 지위가 높다든지 'Whether you're of higher status' 2 뭐 돈이 더 많다든지

'or you have more money'

<sup>3</sup> As noted by an anonymous reviewer, professors might have heightened awareness of the need to treat their juniors with more sensitivity due to receiving faculty training in these areas. Although this is certainly a possibility, this was never explicitly mentioned in the interviews.

3 내가 나이가 많다든지

	- 'taking whatever social power you have and trying to push down on me'

```
'I think all of these manifestations are yeyuy epsta'
```
P09 noted that although you would not tell an elderly person to their face that they are *yeyuy epsta*, it would still be possible to refer to an elder in these terms when they are not present. Based on this, she concluded that *yeyuy epsta* was a term that could be used to describe the behaviour of elders as well as juniors.

Although some speakers mentioned linguistic devices in their description of *yeyuy paluta/epsta* such as the use of honorifics, more emphasis was placed on the content of speech, as well as on non-verbal aspects of politeness. Participants described *yeyuy paluta* as involving the avoidance of boasting, or asking inappropriate questions (P6). *Yeyuy* resides in *thayto* 'embodied attitude', which included correct posture and the control of facial expressions and the 'tone of the voice' (말투).

Some participants, particularly among the professors, saw *yeyuy* as a socially-imposed ritual or convention, as well as the adherence to a strict and complex set of rules, referred to by P16 as 예의범절 *yeyuypemcel* 'the rules of etiquette'. For P10, the driving force behind upkeeping these rules was self-presentation and the desire to avoid having yourself judged as *yeyuy epsta*:

(2) P10 (15:44)

1 저는 본질은 자기 보호라고 생각해요. 'I think the essence [of *yeyuy epsta*] is self-preservation' […] 2 '버릇 없다' 또는 '예의 없다' 뭐 이런 그 판단을 받을 수-

'you could be judged as *pelus epsta* ('rude') or *yeyuy epsta* ('lacking civility')'

3 그 상대로부터 어 받을 수가 있기 때문에-

'since you could receive [such evaluations] from the interlocutor…'

Both groups of participants noted that status inferiors were more susceptible to having their behaviour judged as *yeyuy epsta*, meaning that they had to exercise additional care in their behaviour. S20 described *yeyuy* as "an expression from the past" (옛날 말) and as a linguistic device used when older people are judging the behaviour of younger people. However, a more general interpretation of *yeyuy paluta* was that people adhered to show consideration (배려) for others and to avoid hurting their feelings (상대방이 기분 나쁘게 하지 않다 – P02). *Yeyuy epsta*  occurred when your behaviour results in others being hurt (상대방의 마음이 서운하다 (S01).

#### **4.1.2** *Concwunghata* **'Be Respectful' and** *Mwusihata* **'Be Disrespectful'**

As noted above in Section 2.2, previous accounts of politeness in the Korean context tend to emphasize the hierarchical aspects of respect, equating the concept with showing deference to status superiors. In contrast to this, only three of our participants (one student and two professors) described *concwunghata* 'be respectful' as a hierarchical concept. The remaining 17 participants all saw it as something that was horizontal and mutual, and equated it with respect for fellow human beings (인격체). As for *mwusihata* 'be disrespectful', almost all (nine out of ten) the professors saw this too as being horizontal, that is, committing an act of *mwusi,*  'disrespect' meant that one disrespects the other's intrinsic value as a human being. As for the students, five of the nine students who provided an answer on this item saw it as hierarchical.


\* = One participant did not provide a definition for *mwusi*

*Concwung* involves treating others as you would want to be treated yourself and seeing things from the perspective of others, which is connected to the concept of consideration (배려) mentioned in Sub-section 4.1.1 above. As described by S20, "if you don't want to suffer something yourself, then other people won't want to suffer it either (내가 당하고 싶지 않으면 다른 사람이 당하고 싶지도 않다)." According to S03 and P06, *concwung* is not a matter of who is younger or older or elevating those of higher status, but recognizing (인정) or showing interest (관심) in others for who they are:

(3) S03 (20:39)

1 존중은 머 윗사람 아랫사람 상관없이 'concwung, it doesn't matter whether you are the older party or the younger party' 2 어떤… 한 개인으로 이제 인정해 주는 거를 이야기할 수 있을 것 같아요. 'I think it's about recognizing someone else as an individual'

(4) P06 (17:43)

1 꼭 이렇게 상대방을 나를 뭐 높여달라 이런 차원이 아니라

'it's not about the interlocutor asking me to elevate them' 2 상대방에 대해 인제 관심을 가지고

'it's about showing interest in the interlocutor'

3 그 사람이 인제 어떻게 보면 굉장히 편안하게- 사람을 편안하게 해 줄 수 있는 것 같고

'it's about being able to make the other person feel comfortable'

	- 'isn't that what *concwung* is really about?'

P06 further noted that taking an interest in others can be effortful in cases where their interests are different to your own. Nonetheless, for *concwung* to be achieved, your display of interest must not be fake (가식적).

*Mwusihata*, on the other hand, occurs when you fail to acknowledge the individual perspectives or positionality of others. As noted by P17, *mwusihata* is "thinking that people who are different from myself are wrong (나와 다른 것을 틀리다고 생각하는 거죠)", or, as observed by S08, "not respecting the values of others (그 사람이 가지고 있는 가치를 존중 하지 않는다 그런 게 무시한다고 생각해요)." *Mwusihata* occurs "when we judge others based on our own standards (상대방을 제 기준에 맞춰서 판단하는 걸 – S19)", or "not showing consideration for the boundaries (경계, 선) of others (S20)". P18 notes that whereas *mwusihata* is similar to *yeyuy epsta* in that both involve inconsiderate behaviour (배려하지 않는 행동), *mwusihata* is a more extreme form of impoliteness.

Some participants referenced that this model of respect was perhaps somewhat different from stereotypical notions of politeness in the Korean context. S03 noted that since Korea was originally an "elders first" (연장자 우선) culture that *concwung* might be equated by some people with something that was performed to elders, but that she did not view this as being the case since it's also not acceptable to display *mwusihata* towards status subordinates:

(5) S03 (21:24)


'but if so, wouldn't that mean that we can be disrespectful (*mwusihata*) towards status inferiors?

3 이런 건 아니잖아요. 'I don't think that's the case.'

Similarly, participant S01 referenced the rather hackneyed saying that elders need to be respected because they have lived longer and therefore amassed more knowledge, but contrasted it with the observation that juniors may at times have more knowledge than elders and be more mature (성숙하다) and grown up (어른스럽다). Whereas some participants equated more hierarchical aspects of politeness with Korea's Confucian tradition, it's notable that S01 positioned the idea of *concwung* as respecting the individuality of others as a Christian concept:

(6) S01 (27:59) 1 이제 성경 같은 거에 본다면 'If you look in the bible'

	- 'that's what I think [*concwung*] is about'

Somewhat in contrast to this notion of respect as a horizontal concept, some of the students saw a hierarchical side to *mwusihata*, with five out of nine students viewing this as a vertical concept. Although these hierarchical understandings of *mwusihata* at times specifically referenced age- or rank-based social hierarchies (e.g., P13 observed that *mwusihata* would occur when you don't follow the instructions of status superiors such as professors), more commonly these vertical notions of *mwusihata* were less closely mapped onto traditional notions of hierarchy. Rather, as in the following extract from S11, students noted that *mwusihata* involved placing people in a position below you (낮은 위치, 하등) in quite general or abstract terms. In fact, S11 links this more to someone's ability rather than necessarily their status.

(7) S11 (14:50)


Based on this example, *mwusihata* appears to involve a lack of respect or recognition (인정) for someone's value as a human being in general, irrespective of how high or low their social position might be.

The striking finding in this section is that the concept of *concwung* 'respect' seems to be rather divorced from notions of deference or "elevating" superiors. Given that there are multiple Korean words that might correspond to the English word 'respect', readers who are familiar with Korean politeness metaconcepts may wonder whether this horizontal account of respect applies only to *concwung*. In recognition of this, we also asked two participants from each group to provide definitions for *conkyeng* (존경; 尊敬) 'reverence, esteem'. All four participants saw *conkyeng* as being a non-reciprocal concept, but one which targeted famous people or other role models on the basis of their deeds or talents rather than on their age or status. As stated by P16, whereas we should show *concwung* to everyone, this is not the case for *conkyeng*:

(8) P16 (28:16)

1 모든 인간은 존중 받아야 되는 어떤 그런 권리가 어차피 있는 거잖아요

'After all, all humans have the right to receive respect (*concwung*), don't they?' 2 그렇다고 해서 우리가 꼭 모두 이를 존경해야 될 일은 없는 것 같아요

'But on the other hand, this doesn't mean that we have to revere (*conkyeng*) everyone.'

There may be other words for respect such as *contay* 'deference', which has been widely theorized in previous literature (see Section 2.2 above). However, these were never mentioned by the participants. As expanded below, our explanation is that the use of honorifics and other deferential behaviour towards elders is more commonly conceptualized as "elevating" (높이다) elders rather than necessarily as an act of respect.

#### **4.1.3** *Chincelhata* **'Be Courteous' and** *Pwulcincelhat***a 'Be Discourteous'**

There was almost uniform concurrence across our participants that *chincelhata* 'be courteous' and *pwulcincelhata* 'be discourteous' were horizontal concepts. There was just one student who defined *chincelhata* along vertical lines, but all other participants saw both concepts as horizontal.

Across both the students and the professors there was a high level of consensus that *chincelhata* is a mode of politeness that is aimed at people you don't have a personal relationship with, and which is most closely associated with working in customer service industries. Participants associated *chincelhata* with a smiling expressions (웃는 표정 – S19), a 'sweet smiles' (상냥한 미소 – S13) or bright facial expressions (밝은 표정– P06), particularly those produced by sales assistants. Similarly, when participants described *pwulchincelhata*, they would evoke episodes from service industry encounters. Since *chincelhata* does not require an established relationship, P09 noted that judging someone as *chincelhata* can sound somewhat distancing. You could not describe someone you are close to as *chincelhata*, but rather as 자상하다 *casanghata* 'thoughtful and considerate'.

Some participants, particularly among the professors, described *chincelhata* as an extra, optional layer of politeness, which went beyond what was normatively required or what was demanded by the interlocutor. P15 explained that *chincelhata* involves "going one step beyond the "line" of expected behaviour (지켜야 할 선 보다도 더 한 걸음 앞서 나가서)". *Chincelhata* does not just protect the feelings of others, but "actually boosts their emotions and makes them feel better (감정을 오히려 북돋아 주는 부분인 것 같아요)". For example, whereas it might be enough to answer someone's request for directions verbally, *chincelhata* might involve physically escorting someone to their destination (P12). Or instead of just saying thank you to a delivery driver, you might offer them a glass of water (P15). Ultimately, *chincelhata* involves treating others with a high level of consideration (배려, see Sub-section 4.1.1 above), plus an additional layer of kindness (상냥함 – P18).

Although *chincelhata* was associated with smiling and friendliness, it was also viewed by participants as being inauthentic and affectatious, particularly when it occurred in the service industry. *Chincelhata* can be performed even if you don't really respect the other party, explained S11:

(9) S11 (15:45)

1 친절은 굳이 존중하지 않아도

'with *chincel,* even if you don't respect (*concwung*) [the other person]'

$$
\begin{array}{ccc}
\text{2} & \neg \exists \, \exists \, \exists \, \exists \, \exists \, \exists \, \exists \, \neg \forall \, \exists \, \lozenge \lozenge \land \\
\end{array}
$$

	- 'you just use a warm and kind embodied attitude (*thayto*)'

4 그런 태도로 대해주는 것이 친절이라고 생각합니다 'treating people with that attitude is *chincel*, I think'

This example illustrates an important difference between *chincelhata* and *concwunghata* – the latter relies on consistency between feelings and actions, whereas the former does not. Since *cincelhata* is an affectation, there are cases where people overdo it – what P10 refers to as 'overdone *cincel*' (지나친 친절).

Just as *chincelhata* involves an extra layer of politeness, *pwulcincelhata* 'be discourteous' represents a particularly acute form of discourtesy. P15 described *pwulcincelhata* as a being more extreme than *yeyuy epsta* 'lack civility' since it involves a higher level of *mwusihata* 'being disrespectful', thus bringing together all three of the impoliteness-related metalexemes examined in this chapter. She illustrates the difference in the following example: whereas *yeyuy epsta*  might involve passing over a document in a rough manner, *pwulcincelhata* would involve practically throwing the document at you (note the ideophones in emphasis below):

```
(10) P15 (43:10)
```
1 공공기관에서 뭐 서류를 처리해서 'if you're at a government office and they process your documents' 2 저한테 이제 서류를 주는데, 탁! 'and then when they're giving them to you, *thak* [in an abrupt or aggressive fashion]!' […] 3 한 손으로 탁! 주고 휙 나가버렸어요. 'they give it to you, *thak*, and then just leave abruptly, *hwik* [with an abrupt movement]' 4 그게 예의 없음이고요 'that is *yeyuy epsta*' 5 저한테 이렇게 거의 던지듯이 휙 던져놓고 'but if they almost throw it at me, *hwik*' 6 빠르게 나가버렸어요 'and then run out' 7 그러면 불친절

'that would be *pwulchincel*'

We see here and in other previous examples that im/politeness-related concepts are often mapped onto actions and episodes rather than necessarily to verbal behaviour. In other words, rather than illustrating the concepts of *yeyuy epsta* and *pwulcincelhata* through reference to rude things that people say, P15 exemplifies the difference through reference to different layers of rude physical behaviours. The centrality of multimodal aspects of politeness to metapragmatic knowledge is something that we explored in more detail in our previous paper, Brown et al. (2022).

#### *4.2 Other Metalexemes*

Over the course of the interviews, participants used various other lexemes to refer to politeness and related concepts, beyond those that we asked them to define. Some of these lexemes have already appeared in the previous sub-section, and we believe that a subset of them is worth exploring separately since (1) they were deemed by us as being highlighted fairly often by interviewees and (2) they have not been described in previous accounts of politeness, particularly in the Korean context. We consider four such concepts in this section: 편하다 *phyenhata* 'comfortable', 태도 *thayto* 'embodied attitude', 배려 *paylye* 'consideration' and 인정 *inceng* 'recognition'. Table 6 shows how many participants in each group mentioned these metaconcepts. Seven of the 20 participants mentioned all four of the concepts, whereas a further seven mentioned three of them. All participants mentioned at least one of these concepts.


**Table 6** Number of participants mentioning each metaconcept

We now provide more details about *phyenhata* 'comfortable' (Section 4.2.1), *thayto* 'embodied attitude' (Section 4.2.2), and *paylye* 'consideration' and 인정 *inceng* 'recognition' (both covered in Section 4.2.3)

#### **4.2.1** *Phyenhata* **'Comfortable'**

The term *phyenhata* 'comfortable' and the closely related lexical item *phyenanhata* 'feeling comfortable and peaceful' were mentioned by 18 of our 20 participants (9 students, 9 professors), often in relation to the focal metaconcepts *yeyuy paluta* 'possess correct civility' and *concwunghata* 'be respectful'.

The term primarily appeared in reference to interactions with close friends and/or in casual or informal contexts, whereas interactions with non-intimates status superiors were uncomfortable (불편하다). In "comfortable" interactions, you can act without care for formality (격식), politeness (예의, 공손) (S11, S20, P16) or fear that you will cause offense to the other party (S14). You can speak naturally (자연스럽게 – P10), freely (자유롭게 – P17), and frankly (솔직하게 S19), and you can also show your true personality (S05). Non-honorific speech (반말) can be used (P18), as well as slang (S07, P10), and you also don't need to pay attention to your posture and non-verbal behaviour (S20). In contrast, an uncomfortable situation is one in which you have to follow a formal role (형식적인 역할 – P4), and when you're unable to be perfectly honest. S19 noted that "if you can be 90% honest with friends, in front of professors you can only be 30% honest" and explained that differences in opinions have to be glossed over when talking with superiors. P4 noted that an uncomfortable person is someone who is authoritative (권위적이다) or official (사무적이다).

In addition, *phyenhata* 'comfortable' is something that can be created through showing appropriate respect (존중) and consideration (배려) to others, not just to friends but to superiors as well. In fact, S03 explained that politeness (예의) involved making elders feel comfortable (윗사람을 조끔 더 편안하게 해주는 걸) through the use of more careful and considerate behaviours. Failing to do this (and just doing what is "comfortable" for you) results in impoliteness (P16). Meanwhile, the professors also noted that they could make relationships and interactions with students more "comfortable" by making efforts to lighten the atmosphere, such as by giving students tea and snacks when they visit during office hours (P15).

#### **4.2.2** *Thayto* **'Embodied Attitude'**

*Thayto* was mentioned by 14 participants, including five students and nine professors. This term is commonly translated into English as 'attitude', 'manner' or demeanour', but we find that it has an explicitly embodied aspect not always present in the English counterparts. The term frequently occurred in relation to *yeyuy paluta* 'possess correct civility'.

Indeed, we observed that participants made an explicit division between polite words (말) and polite embodied attitude (태도), and placed more emphasis on the latter. In the following P09, describes how *yeyuy* 'civility' stems from *thayto* rather than the mere use of vocabulary (어휘):

(11) P09 (26:06)

1 예의는 태도에서 나오는 거지 '*yeyuy* comes from *thayto*' 2 어휘에서 나오는 거는 아닌 거 같아서 'I don't think it comes from vocabulary'

Likewise, S20 noted that the difference between non-honorific speech and honorific speech depends on correct *thayto* rather than correct honorifics:

(12) S20 (27:30)

1 막 저 소파 위에서 팔 기대어서 누워 거의 반 누워있으면서 "아! 그러셨습니까?" "저렇습니까?"

'If you're just stretched out on the sofa almost lying down and you say "is that so?" or "is it really" [using honorifics]'

3 존댓말이라고 생각을 안 해요.

'I don't think that's actually honorific language'

4 태도도 있다고 봐요. 'we can't forget about *thayto*'

As for the content of *thayto*, in addition to posture and body position noted in the example above, participants also made reference to avoiding crossing the arms or legs in front of superiors (e.g., S05, S14, S20, P02), using the open hand to point instead of the index finger (S11), reducing the number of gestures (S01, P12, P16), using more fixed facial expressions (S01), pretending to laugh when required (S07), clasping your hands in front of your body (P06, P17, P18) and appearing more docile and less animated in general (S01, P15). In sum, the frequent mentions of *thayto* in the data reveal the intrinsically multimodal view of politeness held by our participants.

#### **4.2.3** *Paylye* **'Consideration' and** *Inceng* **'Recognition'**

*Paylye* 'consideration' appeared in the data from 14 participants (6 students, 8 professors), whereas eleven of our participants (5 students and 6 professors) referred to the concept of *inceng* 'recognition'. We deal with these two concepts together since they seem to be tightly connected.

Both *paylye* and *inceng* are closely related to the concept of *concwung* 'being respectful' described above in Section 4.13. On the most basic level, *concwung* relies on *inceng*

'recognition' of someone's individuality (S03, S19, P17) and/or for their social position, no matter what that position may be. P15 mentioned that he took particular efforts to perform *insa* 'greetings' to everyone he saw in his daily life including those of lower social position in order to show recognition of them. This outward expression of *inceng* and *concwung* is where *paylye* emerges. In the following extract, P09 describes how *concwung* and *inceng* are located in 심성 'mind' and 마음씨 'heart', whereas *paylye* resides in *thayto* 'embodied attitude' and action (행동).

(13) P09 (30:19)

	- 'if *concwung* resides in the mind'
	- 'then *paylye* seems to reside in *thayto*'

'when *inceng* first occurs in your head, then that can translate into *thayto* and action'

According to P09, in order to show *paylye* for others through one's actions, first of all you had to recognise and respect them.

#### **5 Discussion and Conclusion**

In this paper, we asked two groups of Korean speakers from different generations to provide definitions of six politeness metaconcepts, and also explored four other terms that appeared organically in the data.

As shown in Table 7, based on our sample, participants mapped the six concepts onto discrete aspects of Korean politeness. *Yeyuy paluta* 'possess correct civility' and its lexical opposite *yeyuy epsta* were described in terms of the exercise of care, constraint and consideration. The concept of *yeyuy* involves regulating your behaviours and showing consideration in order to make others feel comfortable and avoid hurting their feelings. As for *concwunghata* 'be respectful' and *mwusihata* 'be disrespectful', this dichotomy involved displaying genuine concern and interest in others, and showing recognition of someone's social position. Finally, *chincelhata* 'be courteous' and *pwulchincelhata* 'be discourteous' were seen as modes of (im)politeness that were specific to encounters with non-acquaintances and/or in the service industry, and as additional or more extreme forms of (in)civility. In contrast to *concwunghata* 'respect' which was equated with genuine concern for others, *chincelhata* was viewed as inauthentic and affectatious.

Whereas previous accounts of politeness metaconcepts in Korean have tended to emphasize hierarchical, that is, vertical, aspects of politeness (e.g., Lee & Ramsey, 2001; Yoon, 2004) and treat these as "separate systems" from non-hierarchical and/or strategic aspects of politeness (e.g., Hwang, 1990; Sohn, 1986), our results suggest a more complex picture. The six terms that we investigated are certainly variegated according to relative social standing, but not in a way that maps directly onto a hierarchical understanding of politeness. Across the six different metaconcepts that we researched, it was only *yeyuy paluta* 'possess correct civility' that was seen by the majority of our respondents as being a hierarchical concept, and this result was driven by the students. *Concwunghat*a 'be respectful' and *mwusihata* 'be disrespectful' were viewed by most participants as horizontal, whereas *(pwul)chincelhata* 'be (dis)courteous' is sensitive to social distance rather than power since, according to our participants, these terms were only used to refer to strangers.


It was particularly noteworthy that the concept of *concwung* 'respect' was viewed primarily as showing regard for someone as a human being (인격체) and treating others as one would like to be treated oneself rather than the "elder respect" emphasized in previous studies (L. Brown, 2011a; Yoon, 2004). Instead of using the term "respect", the need to use honorific language and deferential nonverbal behaviours to elders was described as "raising" (높이다) the interlocutor while lowering (낮추다) the self, and as showing "recognition" (인정) for

someone's more advanced social standing. Some participants explicitly challenged the notion that elders needed to be respected more just because they had lived longer.

Participants acknowledged that interactions with elders came with specific sets of expectations, and also with distinct emotional qualities. Whereas interacting with friends was comfortable (*phyenhata*), interactions with elders were effortful and careful, and required endurance. However, the underlying principles that determined behaviour across different interactions were viewed as being the same. Namely, irrespective of the identity of the interlocutor, it is important to show recognition (인정) of their social position, and consideration (배려) for their emotional wellbeing. Participants saw it as socially unacceptable to ignore or disrespect (무시하다) someone as a human being, or cause them emotional pain (기분 상하다), irrespective of how low their social position might be. It was notable that the professors, who have a high social standing, talked frequently of the need to show consideration towards students, and to be careful with behaviour towards younger generations. Although this finding may be somewhat specific for the university context where we undertook our research, this is a fairly significant finding for the study of East Asian politeness. The need for elders to treat their juniors with care is a politeness-related concern that has scarcely been described in previous literature.

Whereas politeness research has viewed "face" as an important construct underlying politeness (see Haugh & Bargiela-Chiappini, 2010) and which is applicable to Korean via the notion of *cheymyen* (De Mente, 1998; Lim & Choi, 1996; Oak & Martin, 2000), this concept was practically absent in our data (it was mentioned in passing by just one participant). Rather than face, participants viewed emotional attunement as an important underlying motive for politeness. Achieving politeness lay in showing care for the emotions of others, and in the related concept of comfort (*phyenhata*). In intimate interactions it is important to maintain this feeling of "comfort" in a mutual way, whereas in interactions with elders the burden is placed on the status inferior to maintain the "comfort" of elders acting deferentially towards them and suppressing the kinds of "comfortable" words or actions that would be used with intimates. These results raise the question of whether *cheymyen*, a term that only appeared once in our data, actually has (or rather, *still* has) much currency as a politeness concept in Korean culture. They also raise questions for the wider practice in the politeness research field of theorizing on supposedly culture-specific emic concepts of politeness and face, without grounding such work in empirical data.

Our participant's understanding of *(pwul)chincelhata* 'be (dis)courteous' also sheds light on some previously unstudied aspects of politeness in Korean. In contrast to the focus on vertical politeness in previous studies, it was noteworthy that *(pwul)chincelhata* 'be (dis)courteous' was defined in terms of social distance: this is a term that is only ever used when referring to strangers. *Chincelhata* was also described by participants as being an extra layer of politeness that goes beyond what is normally required. In this way, our participants showed that metapragmatic knowledge of politeness may involve understanding of politeness in terms of different levels that may either meet normative expectations, or even surpass them. This distinction between a basic expected layer of behaviour and an extra layer of behaviour that goes beyond what is simply appropriate is reminiscent of some previous theoretical claims in the politeness research literature, particularly Watt's (2003) distinction between "polite" and "politic".

In addition, participants noted that *chincelhata* was distinct from other aspects of politeness, particularly *concwunghat*a 'be respectful', due to its affectatious and inauthentic nature. Our participants thus held a distinction between sincere politeness (*concwunghata* as well as *yeyuy paluta*) and insincere politeness (*chincelhata*). The potential for some aspects of politeness to be fake and deceptive occurs in previous studies of metapragmatics using interview data, notably Blum-Kulka (2005, p. 257) where Israeli interviewees described politeness simultaneously both in positive terms and "as something external, hypocritical and unnatural." Similarly, Greek and Turkish informants in Bayraktaroğlu and Sifianou (2001, p. 7) distinguished politeness of manners" (where "real intentions" may be hidden) from "politeness of the soul" ("true politeness"). Future research will be needed to capture this distinction in more detail. It raises interesting questions about how "fake politeness" is used and perceived during interaction. Although our participants tended to see *chincelhata* in a more negative light due to its lack of authenticity, there could also conceivably be contexts where participants prefer the impersonal nature of "fake politeness" over the negotiation of personal emotions that might be required with "genuine politeness."

Quite importantly, what our results show is that multimodality is essential to the very concept of politeness itself, at least in the Korean context. To talk of politeness is to talk not just of verbal strategies, but of an embodied experience that inhabits physical contexts and spaces. Whereas previous descriptions of politeness in Korean and beyond have tended to focus on morphosyntactic and lexical aspects of (im)politeness, our interviews showed that participants viewed (im)politeness as an intrinsically multimodal concept. Participants saw politeness as emerging from the content of speech, the sound of the voice and, most notably, from nonverbal behaviours. Politeness was described as an 'embodied attitude' (*thayto*) that was manifested in posture, body position, gesture and facial expression. In fact, a swathe of recent studies have argued for the importance of multimodal aspects of politeness (see L. Brown & Prieto, 2017), including in Korean (e.g., L. Brown & Winter, 2019; Winter & Grawunder, 2012). What is remarkable from our perspective was that participants' metalinguistic and metacultural knowledge about politeness quite naturally included that very multimodality that researchers have only recently started to theorise (see our previous paper, L. Brown et al., 2022).

Although interviewees from both of our participant groups (students and professors) appeared to understand politeness in rather similar ways, there were also important differences. Most strikingly, students expressed more of a hierarchical understanding of politeness and showed more awareness of the need to upkeep politeness with elders. They were also more cynical towards the concept of politeness, with some students seeing it as a tool used by elders to judge younger generations. Professors, on the other hand, appeared less concerned with hierarchy and instead spoke of the need to exercise caution towards students and younger people. These contrasts seem to primarily reflect the different current ages and social positions of our participants. Put simply, the students are more sensitive to age and rank since adhering to such hierarchies is a more salient part of their everyday lived experiences. As university students they are constantly reminded of their relatively "low" place in the hierarchy of the university through daily interactions with those of higher status, not just professors and teaching staff but also other students who are their 'seniors' (선배). Although students might not necessarily agree with the hierarchical application of politeness in the university, at the same time they are aware of the need to adhere to these norms to cultivate a refined self-image for their own self-preservation and, ultimately, to strive for a higher status.

The professors, on the other hand, with their relatively high social positions, were less affected by such hierarchies (indeed, participant P02, a male professor in his 60s, noted in the interview that he doesn't really have any status superiors these days). Instead, the professors are more concerned with the need to treat students and other social inferiors reasonably. Their high

awareness of this need appears to reflect the fact that elders or status superiors drawing out their power and abusing their authority is increasingly seen in a negative light in South Korean society (see Kim-Renaud, 2001), and indeed has become recognised as a social issue dubbed as *kapcil*  'abuse of power' (Cho & Yoon, 2017). The need to treat students with care was further magnified in recent years by the spread of the #MeToo movement in South Korea, which brought to light instances of sexual harassment by university professors and others in positions of authority (see Hasunuma & Shin, 2019). The question of what enacting politeness means for elders and status superiors in Korean and East Asian societies is a topic that calls out for further research, given that most research to date has focussed on modes of politeness enacted by juniors and status subordinates. It would be particularly interesting to find out how this awareness of the need to treat juniors reasonably extends to other contexts in Korean society beyond the university and professor-student interactions.

To conclude, our study has shown that the ways in which Korean students and professors talk about politeness metaconcepts in an interview setting differ in vital ways from previous claims about politeness, including studies on Korean and other Asian languages. For our participants, politeness was seen as an embodied attitude (*thayto*) that was applied in order to uphold emotional attunement during interaction. Respecting others (*concwung)* via acknowledging (*inceng*) their emotional needs and social position is needed in all interactions, even if the specifics of how this works differ depending on the identity of the interlocutor. These understandings of politeness were shown to be tightly linked to the social context of the interviews (i.e., university setting) and the relative positions of the student and professor participants, underlining the need for more empirical studies into politeness metaconcepts in Korean and other languages.

#### **Acknowledgment**

This research was supported by the Laboratory Program for Korean Studies through the Ministry of Education of the Republic of Korea and Korean Studies Promotion Service of the Academy of Korean Studies (AKS-2016-LAB-2250003). We would like to thank Grace Eunhae Oh, Bodo Winter, Hyunji Kim, Heeju Song, Yiseul Won, Kumseon Choi, Lucy Park and Jaehoon Yeon.

#### **References**


#### **Appendix: Interview Questions (English Translations)**

The interviews were semi-structured and the following was used just as a guide.

#### *Part #1: General Questions*

1. What is the most important thing for you when interacting with other people?

2. What do you have to be careful to do when interacting with older people?


6. How does the way you use nonverbal behavior change according to the context? *If participant mentions terms of their own:* 


#### *Part #2: Questions about Concepts*

*We asked questions about the following Korean politeness-related terms: yeyuy paluta 'polite', yeyuy epsta 'impolite', concwunghata 'respect', mwusihata 'disrespect', chincelhata 'courteous' and pwulchincelhata 'discourteous'.* 


#### *Part #3: Cross-perceptions of Different Groups*


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## "할미 마음이 아파요"**: Korean Honorific Speech Level Markers as Contextualization Cues in Family Instant Messages**

#### **Hanwool Choe<sup>1</sup>**

**Abstract** Bringing together Goffman's (1981) concept of footing, Gumperz's (1982) notion of contextualization cues, and Tannen's (1994, 2007) ambiguity and polysemy of power and solidarity, I investigate how the management of power and solidarity and the construction of family-related identities are signaled through Korean honorific speech level markers in family instant messages. For this study, I examine naturally occurring instant messages of three Korean families(-in-law) via KakaoTalk, a free instant messaging application. The analyzed message exchanges happen between mothers-in-law and daughters-in-law, between a married couple of the same age, and between grandmothers and (virtually present) grandchildren. The family members use the deferential and polite speech level markers as 1) an egalitarian marker; 2) a face-saving marker; 3) a footing marker; and 4) an affective marker. Illuminating language use in contemporary family discourse via instant messages, the chapter demonstrates how the use of the honorific speech level markers neither always adheres to politeness nor is necessarily regulated by traditional social factors such as age, roles, and status. I therefore demonstrate how the speech level markers actually work in everyday family talk online.

**Keywords** Honorifics Speech level markers Honorific speech level markers Footing Power and solidarity Identity KakaoTalk Contextualization cues Family instant messages Digital discourse

<sup>1</sup> H. Choe ()

School of English, The University of Hong Kong, Hong Kong e-mail: hanwoolc@hku.hk

#### **1 Introduction**

The current chapter is a part of my larger project about family group chats among Korean families(-in-law). Although the use of Korean language was not my initial focus, it came into view through close reading of data and interviews with participants. Across the family chatrooms considered in the project, I noted that Korean honorific speech level markers – deferential and polite – are not necessarily used as politeness markers. Rather, they serve to signal how family members manage power and solidarity, while also constructing and performing their familyrelated identities.

In this chapter, I therefore present how the honorific speech level markers neither always adhere to politeness nor are necessarily regulated by traditional social factors such as age, roles, and status, especially in the context of family instant messages. The usage of honorific speech level markers is not fixed, but discursively deployed. I argue that honorific speech level markers are rather viable as a discursive strategy to create contextual meanings and manage family relationships and identities. The chapter demonstrates how the speech level markers actually work in everyday family talk online.

For this study, I use instant messages via KakaoTalk. KakaoTalk, also known colloquially as KaTalk, is a free instant messaging application popular in South Korea. According to the Pew Research Center's Spring 2018 Global Attitudes Survey, cited in Silver (2019), 100% of Korean adults surveyed reported that they own mobile phones and 95% of them have smartphones (N=1,007). The Economist (2019) also notes that 94 % of South Koreans use KakaoTalk. These numbers clearly show how much KaKaoTalk is central to daily communication in South Korea.

KakaoTalk has become a popular research site for analyzing a variety of aspects of Korean language use and communication in online contexts. Those studies include the exploration of listenership among Korean female friends and family instant messages that I present in Choe (2018, 2020), respectively; Shi and Jang's (2017) comparative study on writing styles between Korean generations in their 20s and 60s; Hur's (2017) examination of college students' gender-specific patterns of language; Kang and M. Kim's (2017) analysis on usage of Korean discourse markers, and Kang's (2018) honorific final endings. Some scholars study KakaoTalk discourse to explore translanguaging (e.g., H. Lee and Jang 2021); English language learning (N. Kim 2016); and Korean language learning (M. Kim and L. Brown 2014).

Keeping in line with those studies, the chapter looks into family talk via KakaoTalk. More specifically, it analyzes how Korean family(-in-law) members, in different roles and across generations, use honorific speech level markers (deferential and polite) in their family chatrooms to accomplish various communicative purposes, especially pertaining to the management of power and solidarity and the construction of family-related identities. The chapter therefore highlights the use of honorific speech level markers, which moves beyond politeness, in digital family discourse contexts.

In previous studies on Korean family discourse, scholars consider different types of Korean families including immigrant families, transnational families, and interracial/interethnic families (e.g., S. Choi 2000; E. Cho 2005; J. Choi, Y. Kim, and D. Lee 2012; Song 2012, 2019; A. Kim, J. Lee, and W. Lee 2015). However, many of them are primarily based on spoken interactions between parents and their young children (mostly prepubescent children), especially in relation to language acquisition, language socialization, and multilingualism. A few discourse studies analyze Korean family discourse in the context of mealtime (e.g., E. Cho 2005; H. Kim

2006) and regarding identity construction such as Korean immigrants (D. Kim 2013) and immigrant mothers (J. Choi, Y. Kim, and D. Lee 2012). Furthermore, many Korean family discourse studies have been conducted under the framework of Conversation Analysis (CA) – e.g., K. Yoon (2010); S. Suh (2020); and Pyun and K. Yoon (2022). Very little is understood about Korean family discourse through the lens of Interactional Sociolinguistics (IS) (Gumperz 1982). To bridge the gap with the current literature on Korean family discourse, the chapter, by having IS as the backbone of data analysis, examines family instant messages where relatively understudied familial relationships that include not only marital and parental bonds (e.g., husband-wife and [older] parent-adult child) but also in-laws and grandparent-grandchild relationships are observed.

In addition, while western family groups have received a great deal of research attention in family discourse studies, Korean family (or Asian family in general) discourse is relatively underrepresented. Related, the chapter addresses non-English language and discourse in online family contexts; while it has been more than a decade since Danet and Herring (2007) pointed out the scholarly necessity of examining languages other than English in digital discourse, there is still much to be explored in this context – exceptions include Spanish in Fernández-Amaya (2021) and Arabic in Al Rashdi (2015).

The organization of the chapter is as follows. First, I describe the Korean honorific system, with a primary focus on a variety of speech level markers and give an overview of prior studies on situated meanings of the Korean honorifics. Next, I outline theoretical frameworks that include Goffman's (1981) notion of footing, or alignment, Gumperz's (1982) contextualization cues, or (para-)linguistic and prosodic devices that signal what speakers intend to say, and Tannen's (1994, 2007) theory about the ambiguity and polysemy of power and solidarity. After describing the presented data, I analyze four discursive functions of honorific speech level markers (deferential and polite): 1) an egalitarian marker; 2) a face-saving marker; 3) a footing marker; and 4) an affective marker, and how they, serving as contextualization cues, contribute to family identity work and power and solidarity management in family group chats. Finally, I conclude the chapter by highlighting how the notion of contextualization cues can be applied to understanding relatively underexamined aspects of Korean honorific speech level markers in family talk online. Showing that the honorific speech level markers are not merely politeness markers, my analysis further enriches our understanding of how the honorific speech level markers not only make and signal contextual meanings but also express and negotiate family relationships pertaining to power and solidarity dynamics.

#### **2 Korean Honorific System: Korean Speech Levels and Honorification**

Honorification is one of the distinct linguistic features of the Korean language, and one that I have found family members in my study strategically use in interaction. Korean honorifics explicitly express a speaker's respect for an addressee (addressee honorification) and for a referent (referent honorification) (see Sung 2007; S. Yoon 2010). Addressee honorifics determine a form of sentence ending (e.g., speech level markers), whereas referent honorifics are produced with honorific suffixes and words. When the referent and the addressee are the same, the use of the referent honorific suffix -시(-*si*) usually honors the addressee.

Given that Korean follows the SOV structure (Subject-Object-Verb), the sentence ending, attached to a predicate, is considered the most important honorific feature in Korean. Among sentence final particles in Korean, speech level markers indicate different degrees or levels of

politeness and/or formality that the speaker shows to the addressee and are often interchangeably used with the term "speech style markers" (see M. Park 2012).

Scholars propose various classifications of speech level markers (see J. Hwang 1990) and different numbers of clause types in Korean ranging from five to eleven (see Pak 2008). In Table 1 below, I borrow the classification of six speech levels from S. Yoon (2010), adapted from Sohn (1999), as it clearly differentiates honorific speech levels from non-honorific ones. I also classify clauses used in my data into five types, outlined in Mun (2013), as the five-level classification is generally accepted (see Table 1).


**Table 1** Korean speech levels by clause types

As presented in Table 1, there are six speech levels in the Korean language. They include honorific speech level markers – deferential and polite – and non-honorific speech level markers – blunt, familiar, intimate, and plain. The relationship between speakers and hearers, in general, determines which speech level ought to be used. The honorific speech level markers are usually used by the speaker when talking to those whose social position is higher or in a formal situation, and the deferential is more formal than the polite. The non-honorific speech level markers are used by those socially equal, by the speaker when talking to those lower in the social hierarchy, and/or in an informal situation. Commonly used speech levels in everyday interaction are deferential, polite, intimate, and plain whereas the use of the blunt and familiar markers is less common in contemporary Korean discourse – see M. Kim (2015) and S. Yoon (2015) for the Korean honorification system. Korean clauses can be categorized into five types: declarative, interrogative, imperative, exhortative, and exclamative. As mentioned above and as shown in the transcripts in data analysis, sentence final particles mark clause types in the Korean language. The six speech levels appear across the five clause types in different forms.

Traditionally, the use of Korean honorifics is related to the formality of a situation and is generally determined by social factors such as age, socioeconomic status, and gender (see Sohn 1999 and Sung 2007 for details). As Agha (1998:153) explains, however, honorific speech can serve to accomplish "control and domination, irony, innuendo, and masked aggression, as well as other types of socially meaningful behaviors that native ideologies of honor or respect do not describe." For instance, J. Hwang (1990) identifies four Korean utterances that are deferential and polite, deferential but impolite, non-deferential but polite, and non-deferential but impolite. He argues that different speech level markers, coupled with sentence types such as requests and commands and a speaker's relationship to addressees, can encode different intended messages, in interaction (48).

There has been growing scholarly attention to Korean media discourse to study various contextual meanings of Korean honorific speech level markers used in it. S. Yoon (2010), in his study on Korean media discourse, demonstrates how the polite speech level marker can index "soft affective stance" (98) expressing friendliness, intimacy, or closeness. L. Brown (2013) presents how Korean honorifics are used to indicate sarcasm in Korean TV dramas. In addition, situated meanings of Korean honorifics are often addressed in relation to the speech level shifting phenomena within and between honorific and non-honorific levels (e.g., Eun and Strauss 2004; S. Yoon 2010, 2015; L. Brown 2015; Jo 2018). Other scholars study the strategic use of Korean honorific first-person pronoun, 저(*ce*) ('I' in English) indicating lowering oneself, in the context of political discourse in the media (J. Kim 2018; Chen and J. Lee 2021). To elaborate further, J. Kim (2018) examines that the Former South Korean president, Moon Jae In, indexes humility and loyalty in his use of *ce* in his presidential speeches. Meanwhile, in Chen and J. Lee (2021), South Koreans metapragmatically perceive that Kim Jong Un's use of *ce* during the South-North Korea summit contributes to managing North Korea's threatening image, on the basis of *ce*'s stereotypical meaning of lowering oneself and elevating others.

Previous studies effectively show the strategic usage of honorific speech level markers in institutional contexts where participants negotiate different interactional dynamics. It is thus worth noting how contemporary Korean family members, via instant messaging, employ honorific speech level markers in everyday lives and what they express, especially when the markers do not bear upon politeness-related meanings. Therefore, in this chapter, I present how honorific speech level markers play a part in managing power and solidarity dynamics in family relations, thus creating family identity.

#### **3 Theoretical Frameworks**

#### *3.1 Footing and Contextualization Cues*

Interactional Sociolinguistics (IS) is an approach to qualitative discourse analysis, developed by John Gumperz, to interpret "what participants intend to convey in everyday communicative practice" (Gumperz 2015:309). Among key concepts in IS are Goffman's notions of frames (1974) and footing (1981) and Gumperz's (1982) contextualization cues. According to Goffman (1981), each interlocutor shows certain orientations toward the current interaction, which is what he calls footing, or alignments. Goffman (1981) notes that changing footing means "a change in the alignment we take up to ourselves and the others present as expressed in the way we manage the production or reception of an utterance" (128). Footing changes thus show the dynamics of what is presently happening in on-going interaction, or a "frame" in Goffman's (1974) terms. What kinds of footings are taken upon by interlocutors can be signaled through what Gumperz (1982) calls "contextualization cues" that include lexical items, syntactic structure, and paralinguistic features such as tone, pitch, laughter, and non-verbal actions, and interlocutors glean meanings of current interaction through those contextualization cues. Contextualization cues indicate what a given message intends and thus how it should be interpreted in interaction. Through footing and contextualization cues, we are able to better understand that discourse is context-bound and thus that meaning resides not only in the words spoken but in every aspect of how they are spoken. Scholars have demonstrated what resources and strategies people make use of to manage and accomplish footings in various contexts such as a pediatric encounter in Tannen and Wallat (1993), family interaction in Gordon (2009), email exchanges in

Georgakopoulou (2011), and everyday conversations among friends (Sierra 2023). Such an analytical inquiry is also found in a range of Korean discourses including a foreign language classroom (Park 2016); livestreamed mukbang, or a livestream of eating (Choe 2019, 2020, 2021); TV shows (S. Yoon 2015; K. Kim and Suh 2021); code-switching between English and Korean (Shin 2010); and narratives of *kirogi* mothers, or mothers living abroad with children for their education, while fathers living alone in Korea for financial support of the family (H. Lee 2010).

#### *3.2 Ambiguity and Polysemy of Power and Solidarity*

Power and solidarity feature prominently in the context of (im)politeness (see Spencer-Oatey and Žegarac 2017). Especially, related to (in)direct speech, many consider how power and solidarity play a role in performing (im)politeness. For instance, P. Brown and Levinson (1987) note facework – whether an act is face-threatening or face-saving – is bound up with how a speaker says what they intend to say. Tamaoka, Yamaguchi, Miyaoka, and Kiyama (2010), building upon P. Brown and Levinson (1987), claim that the use of (in)directness is also genderrelated. In Korean discourse, (im)politeness, in light of power and solidarity, is constructed and accomplished through the use of honorification (e.g., Sohn 1981; Hijirida and Sohn 1986; Kim-Renaud 2001; Leech 2014; S. Yoon 2015) and terms of address (e.g., S. Hwang 1991; K. Lee and Y. Cho 2013). In such language use, a range of sociocultural factors including role, status, and age is considered, thus discursively creating and managing power and solidarity dynamics among interlocutors.

As scholars note, power and solidarity cannot be simply understood in a dichotomous way. Through the lens of IS, Tannen (1994, 2007) proposes "the ambiguity and polysemy of power (hierarchy) and solidarity (connection)" in interaction. By ambiguity, she refers to meaning either power or solidarity whereas by polysemy, she refers to meaning both power and solidarity. Tannen notes that in conventional belief, power and solidarity are dichotomous in that hierarchical relationships preclude closeness and vice versa. But in real-life discourse, Tannen argues that they, in fact, are paradoxically and simultaneously exerted. In other words, power and solidarity are not mutually exclusive; each entails the other, thus creating ambiguity and polysemy. Tannen (2013) also notes such "dual, paradoxically related, dimensions of power and connection" (491) often emerge in gender-related patterns in family discourses such as with mothers (e.g., Gordon 2002, 2007; Tannen 2006) as well as with fathers (e.g., Gordon, Tannen, and Sacknovitz 2007; Marinova 2007). However, very few interactional sociolinguistic studies consider other types of family discourse such as between in-laws as well as between grandparents and grandchildren, concerning the complexity of power and solidarity relations – recent exceptions include Nguyen (2020) and Hirasawa (2023), both of which are based on faceto-face spoken interaction.

Tannen's paradox of power and solidarity has been observed in Korean language use as well. For instance, M. Kim (2015) investigates the entwinement of power and solidarity through the use of the solidarity term, 자기 (*caki,* you in English) in a circle of Korean married women friends. M. Kim notes that the informal, affective second person pronoun, *caki*, creates and indicates solidarity and closeness, but at the same time, the use of *caki* indicates age hierarchy as it is only possible from older friends to younger ones, unless it is used between those of a same or similar age. M. Kim and Strauss (2018) claim that the widespread use of *caki* even in formal contexts, in place of the formal second person pronoun, 자네 (*caney*), suggests that solidarity is

the new power. In line with this body of literature studying the co-occurrence of power and solidarity in the usage of Korean pronouns, I consider how Korean honorific speech level markers show such ambiguous and polysemous dynamics in interaction and what discursive functions they perform in family instant messages.

#### **4 Data Collection**

In this chapter, I analyze naturally occurring KakaoTalk instant messages – among families of the married couple, Jia and Taewoo, among Sara's family-in-law, and between the married couple, Phillip and Eunbyul. The analyzed message exchanges happen between mothers-in-law and daughters-in-law, between a married couple of the same age, and between grandmothers and (virtually present) grandchildren. By virtually present grandchildren, I refer to those who are not present in family chatrooms (i.e., they are not members of the family chatrooms), but are constantly mentioned with their daily photos and videos shared by other members of the family chatrooms.

I recruited study participants in person and by posting a recruitment post on Facebook. With informed consent, approved by my then affiliation, the participants were asked to download and share with me their family KakaoTalk conversation(s) by email. When downloaded, messages are converted to a text file and each message appears with its time sent and its sender's name. All family participants agreed to have their instant messages published, except shared photos and videos which I described in language in transcripts. Although I asked them to submit at least one month of their instant messages, I let them decide the exact length of and time range of their interaction. Jia's family and family-in-law offered me a total of 19 months-long instant messages (December 2015-January 2017 and September 2017-January 2018); Sara's family-inlaw did 28 months (September 2014-January 2017); and Phillip and Eunbyul did 5 months (August 2017-December 2017). Any identifiable information of the participants including names, jobs, and locations are adjusted to protect their privacy, except for the cities that the participants live in.

My data collection and analysis are retrospective. I collected and analyzed instant messages that had already happened. My retrospective approach helps to minimize any possible concerns caused by "the observer's paradox," or the idea that the presence of a researcher or recording equipment for data collection does not allow linguists to access to "how people speak when they are not being observed" (Labov 1972:97). It was made possible by the most distinct feature of KakaoTalk, among other features, that all exchanged messages, including the name of the sender and date and time sent, are automatically stored in the database of all involved participants' KakaoTalk accounts. The participants can access and download the entire instant message interaction anytime as long as they stay in the chatroom where those instant messages are exchanged. To learn the context of instant messages having occurred and the participants' insights, I also conducted what Tannen (2005) calls playback interviews via email, phone call, and in person. During the interviews, the participants were asked to read their own instant messages and tell me their thoughts and background contexts at that time.

#### **5 Data Analysis**

In the context of Korean family discourse, it is no wonder that adult children(-in-law) employ honorific speech level markers, when interacting with their parents(-in-law) because the adult children(-in-law) are younger by age and lower in the family hierarchy than their parents(-inlaw). However, my collected data show different usage of the honorific speech level markers among family members, regardless of their age and family role-hierarchy. The family members strategically use the honorific speech level markers, as observed in instant messages between the same-aged married couple, from grandmothers to (virtually present) grandchildren, and from mothers-in-law to daughters-in-law. I note that the honorific speech level markers do not always convey politeness-related meanings. In my data, the speech level markers serve as 1) an egalitarian marker, 2) a face-saving marker, 3) a footing marker, and 4) an affective marker, to accomplish various interactional goals, especially in relation to the management of power and solidarity and thus the construction of familial identities. I argue that the honorific speech level markers, functioning as "contextualization cues" (Gumperz 1982), signal how family members linguistically orient to the current interaction. Please note that the honorific speech level markers appear in bold in transcripts.

#### *5.1 Egalitarian Marker*

The following example is instant messages between Phillip and Eunbyul, a same-aged married couple. At the time the interaction happened, it was the wedding day of Phillip's cousin. Although the couple was invited to the wedding to be held in Seoul, Phillip attended the wedding on behalf of Eunbyul and himself. It was because the couple was in a long-distance relationship at that time in order to gain some help from Eunbyul's parents to take care of their newly born daughter. Eunbyul resided with the baby at her parents' in Busan, while Phillip stayed in Seoul.


ventilation-also if possible let-should not-and-IMV.**POL**-CUTE 'Also the level of particulate matter is pretty bad So try not to open the window'

Discourse analysis studies on family interaction, such as by Tannen (2004) and Gordon (2009), identify patterns of language use that construct and display family identities, which is what Søndergaard (1991) and Gordon (2009) call familylect (c.f., Van Mensel's 2018 multilingual familylect). The couple's use of the polite speech level marker, I note, can be counted as what I call couplect, or a type of language between a married couple, which creates "reciprocal pairs" (Gordon 2009:61) to construct a couple-centered interactional activity (or frame in Goffman's 1974 terms) in family interaction. As addressed in Section 2, Koreans use honorific speech level markers to older or hierarchically higher acquaintances or when they just meet. It is rare for interlocutors close to each other to use those markers, especially when they are of the same age. However, as you see the speech level markers in bold in the transcript above (i.e., deferential in line 1 and polite in lines 2 and 3), the couple constantly uses the honorific speech level markers to each other and the couple's use of them is worth considering, given that they are of the same age, especially in the context of Korea's age-centered culture and language use. In the playback interviews, the couple mentioned to me that the use of the honorific speech level markers is their own interactional habit to display respectfulness toward the other as a partner. Their doing so is strategic, as it contributes to supporting the construction of the discourse of egalitarian marriage and their familial identities as marriage partners in it, thus creating a couple-centered frame in their instant messages.

Interestingly, in lines 2 ('put on a coat over the suit') and 3 ('not to open the window'), constructed is the IMV-POL structure, which is noteworthy in a sense that Eunbyul's use of the polite speech level marker constitutes her dual spousal identities, in relation to power and solidarity, through which she not only exerts her power to take care of her husband and household affairs, but also enacts as a caring spouse. This example will be further analyzed in the next section where I present honorific speech level markers as a face-saving marker.

As mentioned earlier, one of the most salient features of the Korean language is its highly developed honorific system (see Sohn 1999; S. Yoon 2015). Sentence endings including speech level markers as well as clause type markers make the most important contribution to the honorific meaning of the whole sentence (S. Yoon 2015:99). In other words, in Korea where language use is greatly determined by sociocultural factors such as age, role hierarchy, and the degree of closeness between a speaker and an addressee (or a recipient), the use of honorifics is usually designed to display respect and deference between interlocutors. However, through the untraditional practice of using honorific markers, the couple signals their equality and sameness in their family hierarchy and interaction, thus performing their spousal identities. This linguistic phenomenon, I claim, shows how honorific speech level markers, as contextualization cues, serve as an egalitarian marker.

Using the non-honorific speech level marker could suffice enough to indicate their egalitarian relationship as it is usually used between close ones, especially when they are of same age. However, it does not capture the couple's linguistic effort to make it more salient to not only construct and show equality but also perform their own family identities as a married couple. Importantly, the couple's using honorifics means more than reciprocity; it goes beyond one person using an honorific and the other doing the same in return. When I asked the couple about their motivation for using honorifics during the playback interview, Phillip and Eunbyul

mentioned to me that they have engaged in this practice since they started dating in order to display respect to one another. The couple's intentional use of the honorific speech level markers therefore not only creates linguistic reciprocity but also discursively frames the couple's marriage as an equal relationship and constructing identities.

#### *5.2 Face-Saving Marker*

The following example presents together two separate instant messages sent by Sara's mother-inlaw to her family chatroom where Sara, her husband, Insung, and her brother-in-law, Inseok (Insung's younger brother) participate. Kihong is the 4-year-old son of Insung and Sara at the time the data was collected. Sara's mother-in-law often asks for photos of her grandson, Kihong, in the chatroom. Regardless of whether the mother-in-law specifies an addressee or not, when issuing a request for daily photos and videos of Kihong, it is always Sara, her daughter-in-law, who fulfills, or is expected to do, her requests. Her request messages are very different from other instant messages that she sends to the family chatroom. Her request messages are usually sent with politeness markers that include a polite speech level marker (*-a/eyo*) and apologetic expressions that include "미안" (*mian*: 'sorry' in English). Please note that the singularity/plurality of the word, photo, is intentionally omitted in English translations because it is not clear in the mother-in-law's requests.

(2)


Recall the previous example where Eunbyul tells her husband to put on a coat and not to open the window. When she sends such instant messages, she uses the IMV-POL structure, which, I note, also indicates the polite speech level marker as a face-saving marker, thus creating the ambiguity and polysemy of power and solidarity. In her polite speech level marker-included imperative messages, Eunbyul minimizes her face-threatening act, while also maneuvering power to look after Phillip and household chores. But simultaneously, given the couple's intentional use of the honorific speech level markers, her use of the polite speech level markers in those IMV-POL structured instant messages can also be interpreted as solidarity as a part of the couple's couplect. This observation suggests the versatile usage of honorific speech markers in context.

In a similar vein, we can analyze how Sara's mother-in-law uses the polite speech level marker, especially when asking for photos and videos of her grandson, Kihong. By taking upon the footing of the requestor, she reduces her face-threatening acts through the use of the polite form. That is, the polite speech level marker in the mother-in-law's messages signals face-saving acts that can function similarly to indirect speech acts in which a speaker mitigates their own self-assertion.

The requests of Sara's mother-in-law are ambiguous and polysemous in terms of power and solidarity, especially when interpreted in light of family-role identities. The highest-ranked person in the family hierarchy is Sara's mother-in-law, who is also the mother of Insung and Inseok and the grandmother of Kihong. She exerts power and exercises her place in the hierarchy to make requests, which is always to be accepted by her daughter-in-law. But, at the same time, she also mitigates the face-threat by strategically using the polite speech level marker (POL) in examples 1 ("기홍 정면 사진 부탁해요" translated as 'please send me photo of Kihong facing front') and 2 ("기홍 사진 보내줘요" translated as 'Daughter-in-law please send me photo of Kihong') and by adding an apologetic expression in example 1 ("미안하지만" translated as 'I am sorry but'). Whether the requests are linguistically mitigated or not, all are a display of power, but they are also a display of solidarity: They show interest in and connection to Kihong and his parents.

#### *5.3 Footing Marker*

In this excerpt, examined are instant messages between Doyoon and her mother-in-law. Their interaction occurs in Jia's family-in-law chatroom – Jia and Doyoon are sisters-in-law. In this chatroom, Doyoon, her husband, her parents-in-law, her sister-in-law (Jia), and her brother-inlaw (Jia's husband) attend. Doyoon and her husband have three children. As they are working parents, Doyoon's parents-in-law babysit their children. Right before the following interaction, the grandparents were babysitting Haneul (5 years old at the time), the second child of Doyoon. Haneul was not feeling well, as he had contracted hand-foot-mouth disease. The grandparents took him to the park in an art gallery nearby their place, to lift his mood.

(3)



I notice that the mother-in-law implements honorific speech level markers, especially when reporting on how the children spent the day during babysitting. In other cases, she uses the non-honorific speech level markers in the family chatroom. In theory, the mother-in-law does not have to use such honorific speech level markers since she is older and higher status in the family chatroom. However, by doing so, the mother-in-law linguistically signals her situational footing to a babysitter. Her distinct use of the speech level markers, I suggest, can thus be conceptualized as constituting a 'babysitter register' toward parents of the child, along with a discursive act of reporting. Given that the traditional use of the deferential marker creates formality, in this example, the mother-in-law, through the use of the deferential marker, discursively as well as symbolically puts down her power as a family senior. It also means that this babysitter register, which is proper in the typical babysitter-parent relationship (where parents are those who hold power over childcare and hire the babysitter), serves as a contextualization cue to indicate her taking upon a situational alignment as a babysitter.

The mother-in-law's use of the honorific speech level markers therefore contributes to framing the current interaction as a babysitting reporting frame (from the family elder's perspective), rather than the one between in-laws. Although the mother-in-law strategically takes on and displays responsibility as a babysitter, interestingly, her daughter-in-law, Doyoon, who is the mother of Haneul, rather frames her babysitting as a face-threatening act, as an imposition she has made. In theory, Doyoon, as the mother of Haneul, has the right to enact power over her babysitter. Instead, in line 3 ('Oh my god ㅠㅠ Thank you ㅠㅠ'), she expresses gratitude to her mother-in-law with crying eyes ("ㅠㅠ"). Those crying eyes, conveying her apologetic behavior, suggest that Doyoon mitigates her face-threatening act. By doing so, she preserves and conforms to the expected power dynamics in the parent(-in-law)-child(-in-law) relationship, rather than bringing in the parent of child-babysitter relationship. The mother-in-law's use of the honorific speech level marker stands out in comparison with Doyoon's. This, I argue, highlights how meanings of the honorific speech level markers are contextually constructed, emerging from situation circumstances and interactions between interlocutors.

#### *5.4 Affective Marker*

In the following excerpt, Kihong's grandmother uses the polite speech level marker in response to a photo of Kihong that was sent by Sara, who is her daughter-in-law as well as Kihong's mother. In the photo, Kihong whines and stretches his hand to grab a phone in Sara's hand.

(4)


Across the family chatrooms in my study, grandparents employ the polite speech level marker when responding to daily photos and videos of their grandchildren who are not physically present in family chatrooms. The polite speech level marker (POL) is used by both Sara (line 3) and her mother-in-law (line 2), but the purpose of their usage is different. In her message, Sara uses it to display respect and politeness toward her mother-in-law because the mother-in-law occupies a higher level in the family hierarchy and also is older than Sara. Sara's use of polite speech level marker is socio-culturally expected.

Sara's mother-in-law (Kihong's grandmother) also employs the polite speech level marker in response to the photo, sent by her daughter-in-law, in line 2 ("우리 손주가 왜 울상이고 할미 마음이 아파요," translated as 'what makes our grandson somber It breaks *halmi*'s heart'). Considering Sara's reason to use of the polite speech level marker, her mother-in-law does not have to use it when interacting with Sara. However, Sara's mother-in-law makes use of the polite speech level marker in her message. She also utilizes *halmi* in the same message. 할미 (*halmi*) is an informal variant of the word, 할머니 (*halmeni*) that refers to 'grandmother' in English. According to the Standard Korean Language Dictionary, published by the National Institute of Korean Language, *halmi* is used when elderly women refer to themselves when talking to their grandchildren, who are the direct addressees in interaction, to lower and humble themselves.

Both the polite speech level marker and *halmi* suggest that the recipient of Sara's mother-in-law's message is her grandson, Kihong, not Sara. Thus, her message is constructed as if the grandmother were directly talking to her grandchild, although Kihong is not present in the chatroom in a way his grandmother is. Referring to herself as *halmi*, Kihong's grandmother, as the highest-ranked family member, linguistically lowers herself for her grandson, who is the lowest-ranked in the family. She uses the honorific speech level markers, as contextualization cues, to discursively strengthen solidarity and thus show her affective alignment toward Kihong. In this sense, the word, *halmi,* suffices to display her affective footing work as well as perform her identity as a grandmother. However, with the addition of the polite speech level marker, the grandmother's emotive alignment toward Kihong becomes much stronger, while also reinforcing her empathetic understanding of tearful Kihong in the photo. In other words, these linguistic devices not only create imaginary interaction with Kihong, but also accomplish her identity work of do(t)ing grandmother.

There is more to be said about how such linguistic maneuvers show grandmotherly love and care. While conducting playback interviews, I asked (grand)parent participants in the study and other parents and grandparents in Korea about what motivates family adults to use honorifics including the honorific speech level markers, when speaking to family children. The (grand)parents said to me that they intentionally use it in spoken interaction for the educational purposes of teaching manner and respect in language use (see Sung 2007 for Korean mothers' uses of the polite speech level marker to teach kids manners). But the analyzed examples are not the cases when (grand)parents have direct (face-to-face) interaction with family children for such an educational reason. Also, both the participants and other older Koreans mentioned to me that using honorifics to family (grand)children is because they deserve respect and love. One of them said to me that as children are human beings, adults owe them the same respect and kindness offered in any other relationship. These comments suggest that the use of honorific speech level markers is a powerful linguistic device for family elders to decrease hierarchical distance to build up solidarity with their grandchildren. But at the same time, it also can be interpreted as a power maneuver in that only family elders can exert it. If younger and lower-ranked family members use non-honorifics toward older family members for the exact same interactional goals, it would likely be perceived as rather rude and disrespectful.

Importantly, the function of the affective marker may appear to be undistinguished from that of the egalitarian marker (section 5.1) as both signal, to some extent, solidarity-building in family relationship and interactions. The difference between them, however, lies in the relationship between the interlocutors. In section 5.1, the same age married couple uses the honorific speech level markers not to lower one's hierarchical status to the other's, but to strengthen their sameness and equality as partners. Meanwhile, in the imaginary interaction between the grandmother and the visually present grandson, the grandmother's linguistically putting herself down intends to show affection for her grandchild, rather than their sameness in the family hierarchy. If the current example is interpreted as the case for the egalitarian marker usage, the honorific speech level marker should perform the same function when her grandchild uses it, when talking to his grandmother – however, in such a case, the speech level markers are normally recognized as a politeness marker.

#### **6 Conclusion**

Building upon previous studies on Korean speech level markers, primarily based on Korean media discourse, the chapter demonstrates the usage of the honorific speech level markers in contemporary Korean family discourse via instant messages. By analyzing discursive functions of the honorific speech level markers in family instant messages, I have shown how the honorific speech level markers are used as contextualization cues and thus inextricably tied up with the context where they are used, performing different discursive functions that go beyond politeness: 1) an egalitarian marker, 2) a face-saving marker, 3) a footing marker, and 4) an affective marker.

Drawing from Gumperz's (1982) notion of contextualization cues, the chapter has shown how such four types, as contextualization cues, signal that family members are manuevering power and solidarity, thus performing their family-related identities, in various family contexts made in family chatrooms. In the couple's chatroom, the honorific speech level markers are contextualization cues that highlight how the couple linguistically constructs and values equality in their interaction, thus reinforcing solidarity. In Sara's mother-in-law messages to Sara, the

honorific speech level markers, as contextualization cues, suggest how the mother-in-law skillfully maneuvers her hierarchical power in relation to her daughter-in-law to minimize her face-threatening act, when requesting photos and videos of her grandson (i.e., Sara's son). The speech level markers also function as a footing marker as a contextualization cue. The mother-inlaw, through the polite speech level marker, signals her taking upon the footing as a babysitter who takes care of her grandchildren while her children(-in-law) work full time. Lastly, the polite speech level markers are contextualization cues that display the grandmother's affective alignment toward her grandson. In each situation, the honorific speech level markers are strategically employed by family members whose hierarchical status is higher, when they talk to those lower in the family hierarchy (e.g., mothers-in-law to daughters-in-law and grandmothers to grandchildren) or by those in the same hierarchy level (between the married couple). This presents how the honorific speech level markers as contextualization cues illuminate that their interactional functions can be contextualized, which is a far cry from their traditionally recognized functions as politeness markers.

Oriented toward the theme of the edited volume, specifically, the chapter closely engages in how (im)politeness is not inherently pregiven in linguistic properties of Korean honorific speech level markers but is rather a discursive and contextual construct. My analysis shows how meanings of the honorific speech level markers are discursively and situationally negotiated and constructed to meet various communicative purposes. This demonstrates how Korean honorifics, in the context of family talk online, do not exclusively and always adhere to social factors such as age, gender, and role and status. Especially when it comes to managing the complexity and nuance of power and solidarity dynamics, Korean speech level markers are deemed to be identity-constituting as well as context-specific.

What has struck me most is the ways in which the older women in the family chatrooms strategically and appropriately use the honorific speech level markers and therefore perform their family-related identities. In presented examples, one of them employs the polite speech level marker when responding to daily photos and videos of her grandson in order to linguistically construct an affective alignment toward him. But, in the babysitting situation, the honorific speech level marker is adopted as a babysitter register so that the mother-in-law (or the grandmother) enacts as a babysitter of her grandchild, especially when reporting to the working parent of the child (i.e., her daughter-in-law) on what has happened to the child during the day. In addition, the older woman uses the honorific speech level markers to save her face when making a photo/video request. These examples clearly show how, in the chatrooms where communication is primarily written with the absence of paralanguage, the honorific speech level markers, as contextualization cues, play a key role in making specific and clear nuanced meanings that the senders intend to convey in their instant messages.

Moreover, the married couple's language use, shown through their instant messages and interviews, highlights how the egalitarian marriage relationship is linguistically accomplished. Also, the wife, by using the IMV-POL structure (see section 5.1), strategically maneuvers power and solidarity to perform her spousal identity, through which she displays the enactment of doing being a caring wife who takes care of her husband and household affairs, but at the same time she enacts power to do so, while also saving her face. It is, according to Tannen (2007:34), the ambiguity and polysemy of power and solidarity that "a linguistic strategy intended as a connection maneuver functions simultaneously as a power maneuver." This dual aspect of the honorific speech level markers also means that discursive functions of the honorific speech level markers identified in this chapter are not mutually exclusive, while also emphasizing the

importance of viewing the honorific speech level markers as discursively fluid and varied, rather than constrained to the context of politeness.

Lastly, the analysis undertaken here has extended our knowledge of how language use links together online interactions and offline interactions, in terms of performing everyday family lives and identities (c.f., the blurring of online and offline interactions in Bolander and Locher 2020). In that regard, future research needs to examine more closely how the family members considered in this chapter use Korean honorific speech level markers in face-to-face spoken family interaction. Also, further work is needed to fully understand the usage of the honorific speech level markers in other types of family group chats such as between (adult) siblings as well as between a married couple with age differences. These two specific future studies will further delve into a range of discursive functions of Korean honorific speech level markers, not limited to (im)politeness, in different types of contemporary Korean family interactions.

#### **Note**

1.


#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Addressee Honorifics as an Interactional Resource for Socialization in Korean Adult-Child Interaction**

#### **Eun Young Bae, Gahye Song, and Seunggon Jeong<sup>1</sup>**

**Abstract** This study examines how addressee honorifics, recognized as an index of deference and respect, are utilized as an interactional resource for socialization in Korean adult-child interaction. This study analyzes 62 instances of adults' use of Korean addressee honorific utterances to children collected from video recordings of five different families using multimodal discourse analytic and language socialization frameworks. The study identifies three major environments where addressee honorification serves as a tool for socializing children into sociomoral values of Korean society: a) giving compliments, b) showing gratitude, and c) issuing directives. In these interactional environments, addressee honorification is used alongside semiotic resources to evaluate children's behaviors and draw attention to the action and content of honorific utterances. The analysis of status-incongruent and creative or performative uses of adults' addressee honorifics further demonstrates that honorifics, which are sometimes employed for politeness and deference toward addressees, do also serve as resources for fostering social awareness, social responsiveness, and courtesy in children. This may contribute to children becoming competent members of Korean society who think, feel, and act in accordance with Korean cultural norms and expectations.

**Keywords** Addressee honorifics Socialization Multimodality Evaluation Attention Social awareness Social responsiveness Korean adult-child interaction

G. Song

S. Jeong

<sup>1</sup> <sup>1</sup>E. Y. Bae ()

Department of Korean Language Education, Seoul National University, Seoul, South Korea Korean Language Education Research Institute, Seoul National University, Seoul, South Korea e-mail: ebae23@snu.ac.kr

The World Languages Center, Northeastern University, Boston, Massachusetts e-mail: gahye.m.song@gmail.com

Department of Korean Language Education, Seoul National University, Seoul, South Korea e-mail: upgroundla@gmail.com

#### **1 Introduction**

In this study, we investigate how honorifics, considered a type of grammatical resource for displaying deference and respect (e.g., Agha, 1993; Duranti, 1992; Hwang, 1990; Lee, 1996), are used in adult-child interaction in Korean. As will be shown, honorification is used by adults as a means to socialize children into competent members of a society who think, feel, and act according to Korean cultural norms, expectations, and preferences. Korean is considered to have one of the most systematic honorifics systems among the world's known languages (Sohn, 1999), and its complex structure and usage have been documented extensively in the literature of both Korean linguistics and honorification (e.g., Brown, 2015; Choo, 2005; Hwang,1990; Ju, 1998; Kim-Park, 1995; Lee, 1996; Lee & Brown, 2022; Martin, 1964; Strauss and Eun, 2005; Park, 2007). In this study, we take both multimodal discourse analytic and language socialization frameworks and focus on a use of honorifics by adults toward children, which deviates from typical honorific usage. We will show that adults employ the addressee honorifics, accompanied by other verbal and non-verbal semiotic resources (e.g., prosody, gestures, body movements, etc.) in performing a range of actions such as complimenting, showing gratitude, and giving directives. We will argue that honorifics mark those utterances as salient and noteworthy, drawing children's attention to what is being said (e.g., *komap-supnita* 'Thank you' produced with an honorific sentence ender and head bowing), and to the socio-cultural and socio-moral information being conveyed by it (e.g., 'Your action—i.e., sharing your possession with others is worthy of respect and deference'). In addition, honorific utterances toward children, by virtue of using them, place children in a social role deserving of respect or with certain socio-moral responsibilities, guiding and nudging them to behave accordingly. In this sense, adults' honorifics are evaluative, with its use being motivated by adults' evaluation of children's past and/or current behavior and of desired, appropriate future actions. Therefore, honorifics are used selectively and strategically by adults as a medium to foster social sensitivity and a sense of morality within children.

#### **2 Background**

With its focus on the role of Korean addressee honorifics and accompanying non-verbal semiotic resources in the socialization of children, the study builds upon the literature of both honorifics and language socialization. As for the literature on honorifics, this study is in line with studies which take an indexicality approach in understanding socio-pragmatic functions of honorifics. Within language socialization research, the study is closely related to those on the use of honorifics in adult-child interaction.

#### *2.1 Honorifics System in Korean*

Honorifics are traditionally considered linguistic markers of social relationship between participants (e.g., Levinson, 1983). By using honorific expressions to/for a person of a higher rank, speakers convey their regard or deference to the person. Comrie (1976) classifies honorifics into three main types based on the entities honorifics deal with: 1) Referent honorifics are used to indicate the speaker's relationship with things or persons being referred to; 2) Addressee honorifics express the speaker's relationship with the addressee; and 3) Bystander honorifics conveys information on the speaker's relationship with bystanders or overhearers. Not all honorific languages have all three types. For example, Lhasa Tibetan has only referent honorifics (Agha, 1993). Brown and Levinson (1987) suggest that while referent honorifics seem to be common among the known honorific languages, addressee honorifics seem to be much rare. Levinson (1983) further mentions that bystander honorifics are most uncommon among the three types of honorific expressions.

Korean has two types of honorifics – referent honorifics and addressee honorifics. Referent honorifics can be further divided into subject, object, and oblique honorifics (Sohn, 1999, p. 414). Honorifics in Korean can linguistically be realized in a variety of ways. Korean exploits personal pronouns, address-reference terms, nouns, predicates, particles, and affixes to form honorific expressions (Sohn, 1999, pp.408-414). One crucial means to linguistically implement addressee honorification is to exploit sentence enders, or 'speech levels' (Choo, 2005). According to Sohn (1999), Korean has six speech levels, whose suffixal realization is dependent on sentence types, as shown in Table 1 below.


 **Table 1** Sentence enders according to sentence types and speech levels

(Modified from Sohn, 1999, p. 413)

Additionally, in Korean, the addressee honorification interplays with referent honorification. More specifically, when the subject of a sentence is the same as the addressee of the sentence, the subject honorific suffix *–si* also functions as a device for addressee honorification. Lee (1996) proposes that in such cases, the subject honorific suffix *–si* reinforces the deference paid to the addressee.

<sup>2</sup> -E indicates either -a or -e.

#### *2.2 Choice of Honorifics and Non-Honorifics*

Many researchers have attempted to identify the factors that influence a speaker's choice of honorific/non-honorific speech levels. In general, four factors are commonly mentioned: social hierarchy (e.g., age, kinship, gender, occupational rank, social status, etc.), intimacy, politeness, and formality. For instance, Hwang's (1990) explanation of the use of honorific speech levels centers around the differentials in social hierarchy between the speaker and the addressee. Hwang suggests that the deferential speech level is the most deferential while the plain speech level is the least. Kim-Park (1995) and Cho (2005) point out both social hierarchy and intimacy as major factors affecting a speaker's choice of honorific/non-honorific speech levels. In contrast, Martin (1964) distinguishes honorific speech levels from non-honorific speech levels in terms of politeness, with the deferential speech level being the most polite and the plain speech level being the least. Choo (2005) and Strauss and Eun (2005), however, differentiate honorific speech levels from non-honorific ones based on the degree of formality, the deferential speech level being the most formal and the plain speech level being the least formal.

Even though social hierarchy, intimacy, politeness, and formality explain in many cases the motivations for selecting honorifics or non-honorifics, there are many cases where these factors do not explain a speaker's choice of speech level. One prime example is adults' use of addressee honorifics to young children. Adults' use of addressee honorifics is puzzling in that the hierarchy existing between adults and young children, in terms of age, kinship, and social status, predicts that adults would use non-honorific speech levels to young children. In addition, the relationship between adults and young children tends to be intimate and personal, and the nature of their interaction tends to be casual and private rather than public and formal. Thus, adults' use of addressee honorifics to young children cannot easily be explained in terms of the aforementioned factors.

In response to early studies that take the view of language as a representation of the preexisting world and assume that honorifics directly encode objective social statuses among speakers (e.g., Levinson, 1983; Brown & Levinson, 1987; Hori, 1986; Jain, 1969; Milner, 1961; Wang, 1990), some researchers began to problematize the assumed reflexive relationship between honorifics and social hierarchy, orienting to the context-(re)creating capacity of language (e.g., Agha, 1993; Anderson, 1993; Clancy, 1985; Duranti, 1992; Lee & Brown, 2022). These studies documented various instances where the use of honorifics index various socio-pragmatic meanings, which are not necessarily associated with the social statuses of participants per se.

In particular, building on Shils' (1982) notion of social status and deference entitlement, Agha (1993) shows that in Lhasa Tibetan, an honorific register does not necessarily signal social status of participants. Rather, it signals deference entitlement, which specifies who (i.e., the origo of deference) pays deference to whom (i.e., the focus of deference) during the course of interaction. In one instance, a father uses an honorific form in referring to his daughter while talking to his granddaughter. Agha proposes that in that particular context, the father is taking on the perspective of the granddaughter, the addressee, by establishing the granddaughter as the origo of deference and his daughter as a focus of deference. In a similar vein, Duranti (1992) points out that Samoan Respect Words (RWs) are not a mere reflection of social status as an a priori fact. Rather, they function as an index of particular roles of and the relationships among the interlocutors that are foregrounded and oriented to at the moment of interaction. He adds that RWs can be understood as keys (Hymes, 1972) or contextualization cues (Gumperz, 1977) in that they also evoke "a certain set of expectations, attitudes, and inferential processes" (p.88)

linked to the roles, relationships, and activity established through the use of RWs. Duranti notes an instance where two speakers talk about the same chiefly title of a higher rank. Whereas one speaker uses RWs, the other speaker does not use RWs. Duranti mentions that the non-use of RWs by the latter displays a positive attitude toward the chief title rather than showing disrespect.

Studies that document the status-incongruent use of honorifics therefore clearly show the importance of examining honorifics as an indexical sign. According to Silverstein (1976) (as cited in Cook, 1999), indexes are classifiable into two categories: presupposing indexical signs and creative or performative indexical signs. Indexical signs are presupposing when utilized to indicate some contextual variables that are already recognized, whereas creative or performative signs make explicit some contextual variables as their referents (Cook, 1999). Cook (1999) explains that honorifics can be both presupposing and creative or performative. Honorifics are context-presupposing when they are used to reflect pre-existing social structure, and contextcreative or performative when used to created context-specific social meanings such as perspective-taking shown in Agha (1993).

For example, Lee and Brown (2022) examine the uses of the deferential -*supnita*, the polite -*yo*, and the non-honorific -*e/a* enders during customer-vendor interaction in the marketplace, where both vendors and customers are shown to switch among these three speech levels during the same stretch of interaction with the same conversational partners. In explaining the shift among speech levels, Lee and Brown focus on indexical meanings of each speech level and how such meanings are utilized for each participant's interactional needs in-situ. For instance, -*supnita* has at its core an indexical meaning of "formal presentational stance" (Brown, 2015), which is utilized in vendor-customer interaction when participants shift from conversational talk to functional and transactional one. The non-honorific *panmal* indexes a lack of social distance at its core. Thus, when used reciprocally and in a friendly way, *panmal* will appear intimate, whereas when used non-reciprocally, it will appear authoritative, as can be seen in its use in utterances to take the upper hand in haggling. Thus, the use of speech levels is to be understood in terms of social and interactional meanings they index, rather than pre-existing factors such as age or social statuses. Lee and Brown also show that a shift in speech level is often accompanied by corresponding shifts in eye gaze, bodily orientation, and/or head position, demonstrating that the indexical meanings of speech levels are created in a multimodal fashion.

As such, this study takes a similar approach in explaining adults' use of honorification towards children, examining the indexical meaning of honorification and how co-occurring nonverbal behavior, as well as prosody, reinforces such meaning.

#### *2.3 Honorifics in Child-Adult Interaction and Language Socialization*

Language socialization research recognizes the indexical capacity of language in socializing children into cultural norms, preferences, and expectations (Ochs & Schieffelin, 1984; Ochs, 1990). It proposes that during the process of acquiring language by interacting with adults, children come to see, learn, and manipulate the organization of social roles, relationships, norms, and expectations that are reflected on or created by language use (Ochs & Schieffelin, 1984). Some researchers of language socialization have shown interest in the use of honorifics in adultchild interaction (e.g., Clancy, 1985; Cook, 1997; Ju, 1998; Park, 2007). These language socialization researchers have attempted to uncover various social meanings produced creatively or performatively through honorifics and how they contribute to socializing children into certain cultural norms and values. For instance, Cook (1997) investigates Japanese mothers' use of the

addressee honorific form *masu* to their children. She argues Japanese mothers use the *masu* form as a way of indexing their public, social personae as a mother who has the responsibility to take care of children and teach them social roles, norms, and expectations, thereby claiming their authority over their children. She argues that Japanese mothers' use of the *masu* form enacts "the mode of self for public presentation" (p. 697) and contributes to socializing children into *omote*  (front) (i.e., the object of public attention).

For Korean, while numerous studies have examined adult-child interaction or family interaction, not many studies have looked into adults' use of addressee honorifics towards children from the perspective of language socialization. Among studies taking this perspective, the works of Ju (1998) and Park (2007) are noteworthy. Ju (1998) investigates the use of a subject honorific suffix *–si*, which also functions as a device for addressee honorification, in an elementary school setting, where she notes that teachers sometimes use *–si* in imperatives to their students despite their higher status. Ju argues that *–si* is a grammatical device that creates social distance between the teachers and the students and establishes their institutionally relevant identities. Using *–si* in imperatives, the teachers define the context of interaction as formal and present themselves as authoritative figures while positioning the students as public performers who should follow teachers' instructions. Ju concludes that teachers' use of *-si* constructs the imperatives as an imposition rather than a request and becomes a tool to socialize children into classroom norms and expectations. Conversely, the status-incongruent, indexical use of honorifics can sometimes lower the imposition of utterances. In Park's (2007) study on the addressee honorific sentence ender*–yo* during family interaction, parents and grandparents sometimes use *–yo* to children in imperatives and interrogatives. Park argues that the use of honorific sentence ender *–yo* in directives and requests contributes to reducing their facethreatening force. She further argues that, by being exposed to the use of the honorific sentence ender *–yo* as a politeness marker, children learn the social meaning of politeness behind its form.

The necessity for studies on Korean honorifics from the language socialization perspective, coupled with the seemingly contradicting functions of Korean honorifics (i.e., an index of authority and imposition vs. an index of politeness) documented in the literature, warrant further investigation of adults' use of Korean honorifics to children from this particular perspective.

#### **3 Data and Method**

Data for this study was collected from five Korean middle-class families, three of which were living in South Korea and two in the United States. The data collection took place in 2009 and 2011. One of the researchers video-recorded mundane interactions involving the participants of this study, resulting in a total of 245 minutes of video recordings. Detailed information about the participants and the recording sites can be found in Table 2 below.


**Table 2** Information on participants and video-recording sites


The adults typically employed non-honorific sentence enders when interacting with the children. However, we identified 62 instances in the dataset where the adults used addressee honorifics when talking to the children (e.g., *komap-supnita* 'Thank (you)', *cal-hay-ss-eyo* 'Good job!'). Note that these instances do not include cases where adults (1) read text or sang lyrics written in addressee honorific forms (e.g., *sangil chwukha-ha-pnita*' Happy birthday (to you)'), (2) prompted children to say a particular utterance (e.g., *"cwu-sey-yo" hay-ya-ci* '(You) should say, "Please give (it to me)"'), or (3) spoke on behalf of the children as if they were the children themselves (e.g., *imo mwe ccik-eyo?* 'Auntie, what are you video-recording?'). The reasons for excluding such cases are as follows. In the former cases (i.e., (1)), the addressee honorific features used in the pre-established text or lyrics are not determined by the choice of the adults. In the latter cases (i.e., (2) and (3)), the adults assume the perspective of the children speaking to adults. Therefore, the addressee honorifics are intended for the relevant adults rather than the children themselves.

These 62 instances were then analyzed within multimodal discourse analytic and language socialization frameworks. Multimodal discourse analysis examines the construction of meaning through various modes of semiotic resources, including language but not limited to it (e.g., prosody, non-verbal behaviors such as facial expressions, gestures, and postures, physical objects in the interactional scene, etc.) (Jones, 2023). Language socialization investigates a close connection between language acquisition and the socialization process by focusing on the socialization through language (i.e., how language is used as a resource of socialization) and socialization into language (i.e., how children learn to use language) (Ochs & Schieffelin, 1986, 2008). Using these two frameworks, we examined the 62 instances by analyzing how adult use of addressee honorifics, in conjunction with other semiotic resources, functions as an interactional resource for socializing the children into the socio-moral values of Korean society. We found three major interactional environments which clearly illustrate children's socialization. Each environment will be analyzed in detail in the Analysis section to follow, along with specific data excerpts, which were transcribed using Conversation Analysis (CA) transcription conventions developed by Jefferson (Atkinson & Heritage, 1984, ix-xvi; Ochs, Schegloff, & Thompson, 1996, pp.461-465; see Appendix A).

#### **4 Analysis**

In this section, we present three major interactional environments where adults shift speech registers and employ addressee honorification toward children. The three environments are when: adults 1) give compliments, 2) express gratitude, and 3) issue directives to children.

#### *4.1 Giving Compliments*

One interactional environment where the addressee honorifics is frequently employed by adults is when adults compliment children. In our first example, two fraternal twin sisters, Minji (older sister) and Suji (younger sister), and two adults, their mother and the mother's friend Hyorin, are sitting on the floor. While the children are playing together, the adults are watching them and chatting. Prior to the segment below, Suji attempted to take away a box of wipes that Minji was playing with, but Minji did not let her. Suji started crying out loud, and their mother lifted and

carried Suji to appease her. Meanwhile, Minji brought Suji another box of wipes so she could play with it. In lines 1 and 2, Suji and Minji's mother notices Minji's behavior and verbalizes Minji's intention in the form of reported speech along with much laughter. In lines 3 and 5, Hyorin, who has been an onlooker during this time, responds to the mother's remark by displaying an affiliative stance as shown in her slight laughter, evaluation, and interpretation of Minji's behavior. Then, in line 7 she asks the mother whether the situation has been resolved. The mother says *i(h)ke kac(h)-ta cwe-ss(h)-e h* '(Minji) brought this (to Suji)' with laughter dispersed within the talk, indicating the trouble has been settled with Minji's giving another wipes box to Suji (line 10).



$$\begin{array}{lll} 10 & \text{Mom:} & \text{=} \circlearrowleft (\text{h}) \rhdlearrowleft & \text{\mathbb{Z}} \text{\mathbb{H}} (\text{h}) \text{\mathbb{Z}} \Vdash & \text{\mathbb{Z}} (\text{h}) \rhdrow \\ & & \text{=} \text{i(h)kce} & \text{kac(h)-ta} & \text{cwc-ss(h)-c h} \\ & & \text{this thing bring-and give-PST-DC:} \text{INT} & \\ & & \text{\textquotedblleft (Mijn) bronight this (to Suji).} \end{array}$$


14 Mom: hehheh

15 Hyorin: 아이:구 ai:kwu wow 'Wow!' *((patting on Minji's left shoulder twice as shown in Figure 1.2))*


While the mother is still laughing, Hyorin shows her appreciation, delight, and fondness for Minji's behavior with exclamations *ai:kwu:.ai:kwu* 'Wow! Wow!' and accompanying gestures (lines 13 and 15). With the first *ai:kwu:*, Hyorin gently caresses Minji's head with both hands (Figure 1.1), and with the second one she pats Minji's left shoulder twice with her right hand (Figure 1.2). Although Minji is not gazing at Hyorin, Minji can still experience Hyorin's positive attitude through these auditory and tactile inputs. Note that Hyorin's construction of Minji's action as delightful, adorable, and favorable is not an automatic or predetermined way of looking at Minji's behavior. The mother's laughter in lines 1, 2, 4, 6, 8, 10, 12, 14 and 18, as well as Hyorin's laughter and evaluation in line 3, shows how Minji's behavior can alternately be seen as something funny.

Following these exclamations, Hyorin issues a complimentary statement at Suji by saying *enni nolus-ul ttokttokh-i ha-si-ney-yo* '(You) are doing (your) part as older sister very well!' (line

15). Here, Hyorin employs addressee honorification, using the subject honorific suffix -*si*<sup>3</sup> and the sentence ender -*yo*. We focus on Hyorin's compliment featuring honorification here as an important site of socialization.

First, a sudden shift in register in the compliment and the verbal and physical actions prefacing it give saliency to this utterance. Since the unmarked register for an adult speaking to a child is the non-honorific, Hyorin by default uses non-honorifics as well when talking to Minji. Shifting to the honorifics in one utterance therefore makes it salient among a stream of nonhonorific speech. When, as in this example, such shift in register is accompanied by non-verbal action, the utterance's distinctness will be even more pronounced. Hyorin's use of honorification therefore is likely to draw the child's attention to the utterance, and, more importantly, to its message.

In her compliment, Hyorin is displaying an evaluative stance towards Minji's behavior. Considering that the basic meaning of honorifics is the expression of deference and respect, Hyorin is expressing respect to Minji in addition to complimenting her. Looking closely at the lexical choice, the kin term *enni* 'older sister' and the noun *nolus* 'part, role' specify why Minji deserves respect. Because Minji played her role as an older sister well—and the role being showing care to her younger sister and meeting her needs—her action is recognized and commended with reference to the moral norms associated with social roles in the Korean society. In other words, *enni nolus* 'older sister's role or part' offers the child a frame of interpretation through which her action gains social and moral meanings. Combined with the expression of respect and deference, the compliment signals to the child that being kind and bringing the box to her younger sister is an older sister's role. Moreover, successfully performing the role of a caring older sister is something commendable, respectable, and perhaps what is expected of her. In this sense, Hyorin's compliment can be considered as an instance of socializing a child through language use.

The next example shows a similar case of using honorification in compliments for evaluating children's conduct. The exchange below occurred as a small group of people were engaged in leave-taking after a church service. The exchange revolves around adults' conjoined efforts to have Minsu perform socially appropriate leave-taking for the pastor. In line 1, Minsu's grandmother is prompting Minsu to say good-bye to the pastor with a polite bow called *yangswu insa,* 'a polite greeting/leave-taking that accompanies a deep bow with hands on the abdomen area,' or *paykkop insa* 'belly button bow' in kid's term. Prior to line 1, Minsu's grandmother has already configured Minsu's body for the polite bow by overlaying her hands on Minsu's hands and positioning his hands on his abdomen area. Then, in line 1, she demonstrates it for Minsu in a slightly exaggerated manner. She bows at almost 180 degrees and says *annyenghi kyey-sey-yo* 'Good-bye, *lit*. Stay in peace.' Afterwards, she explicitly directs Minsu to say good-bye to the pastor with a bow by saying *insa-lul hay pwa '*Try saying good-bye (to him)' (line 3)*.* As can be expected in Minsu's expression of annoyance in line 2, Minsu does not comply with his grandmother's directive.

<sup>3</sup> Hyorin's remark in line 10 is an illustration of the interplay between addressee honorification and subject honorification. Here, the omitted subject of a sentence (i.e., Minji) is the same as the addressee of the sentence. Thus, the subject honorific suffix *–si* also functions as a device for addressee honorification.

#### **Excerpt 2. Good-bye**



03 Grandma: =인사를 해 [ 봐:, ] =insa-lul hay [pwa:,] greeting/farewell-AC do try 'Try saying good-bye (to the pastor) with a bow.'



08 Researcher: [아 : ] [ : ] [유 : ] [: : :] [ a : ] [ : ] [ yu : ] [: : :] wow

'Wow!'

*((with a smiley voice))* 

09 Pastor: [ 아 ] [ : 유 ] [ a ] [ : yu. ] wow

'Wow!'

#### 10 Grandma: [이쁘게 ] [도 하 [네¿ ]] [ippu-key] [-to ha-[nye¿]] pretty-AD-also do-APP

'(You) did (it) so nicely!'

#### 11 Intern Pastor: [ h ] h

12 [h h h h h ] *((While laughing, the intern pastor first looks at pastor's wife, who is smiling at Minsu as shown in Figure 2.4, and then she looks at Minsu.))*

13 Pastor: [아이 잘하네, ] 민수, [ai calha-ney,] Minswu, wow do well-APP Minsu (Name) 'Wow, (you) did a great job, Minsu!' *((looking at Minsu and using a gentle voice))*

14 Researcher: 아[우 ] a[wu ] wow 'Wow!' Addressee Honorifics as an Interactional Resource for Socialization… 69


*((The pastor holds Minsu's left hand as shown in Figure 2.5.))*




Minsu's grandmother is not the only one who is trying to have Minsu perform a proper good-bye to the pastor. Throughout lines 4 to 7, adults around Minsu—his grandmother, an intern pastor, and the pastor's wife—encourage the boy to bow and say good-bye to the pastor, a scene illustrative of the importance of leave-taking etiquette in Korean culture. Finally, on the last syllable of the intern pastor's command in line 5 (i.e., *hay*), Minsu starts bowing to the pastor by lowering his neck and back, which continues throughout the utterance by the pastor's wife in line 7 (Figures 2.2 and 2.3). The adults pleasantly respond to Minsu's socially appropriate behavior (lines 8 to 14). The pastor, the recipient of Minsu's bow, begins his response with an exclamation *a:yu* 'wow' (line 9), echoing another adult's show of delight in line 8, and continues to say *ai calha-ney, minswu* 'Wow, (you) did a great job, Minsu!' (line 13). The grandmother, the intern pastor, and the pastor's wife all join in expressing their delight through complimenting, i.e., the grandmother's *ippukey-to ha-nye* '(You) did (it) so nicely!' (line 10), or nonverbal cues such as smiling and laughing (lines 11-12, Figure 2.4).

The pastor now prepares to leave the church sanctuary with Minsu. In line 17, he turns towards the exit of the sanctuary, holds Minsu's left hand, and says *kaca:: ili wa pwa* 'Let's go. Come here' with a gentle voice (Figure 2.5). Then, in line 21, the pastor re-issues a compliment on Minsu's bowing. He says *cal hay-ss-eyo wuli minswu* '(you) did a great job, Minsu.' Note that the pastor uses the addressee honorific sentence ender -*yo*.

As in the previous example, this utterance with honorification stands out among the surrounding talk, all of which contributes to drawing Minsu's attention to the utterance. First, using honorification makes the utterance salient considering that the default register for

addressing Minsu, as can be seen throughout the excerpt, is non-honorifics. Using honorification, however, is not the only feature that makes the pastor's utterance distinct. Its placement in the ongoing sequence further adds to its markedness. The pastor's suggestion of leaving the place in line 17 was advancing the progressivity of the interaction towards its closure. However, in line 21, the pastor is re-doing his earlier compliment, halting the progressivity of the interaction. Given that there is a general preference for progressivity in interaction (Stivers & Robinson, 2006), the pastor's compliment in line 21 is sequentially marked. In other words, the grammatical and sequential markedness signals the noteworthiness of the turn, prompting the child to heed what is being said and what message is being communicated.

The utterance is highly evaluative, in terms of both action (i.e., complimenting) and the meaning created by honorification. Again, by using honorification, the pastor is showing respect to Minsu and rendering Minsu's proper bowing an act that is worthy of respect and deference even by adults. Furthermore, the utterance is designed with prosodic and lexical features that create a particularly affective tone: The honorific sentence ender -*yo* is elongated in a gentle voice which conveys his favorable stance toward the boy's behavior; the pastor adds the inclusive pronoun *wuri* 'our' to his addressing term for the boy, i.e., *wuri Minsu* 'our Minsu,' thereby displaying his approval and fondness towards the child. In other words, the pastor is positively evaluating Minsu's bowing and signaling that it is socially and morally commendable.

Children's knowledge of how to conduct socially proper leave-taking (as well as greeting) for adults is of paramount importance in the Korean culture. The importance is evidenced in this example as well by the constant pursuit of Minsu's bowing by his grandmother, the intern pastor, and the pastor's wife. In fact, children learn this social etiquette in preschools, elementary schools, and even in some middle schools in South Korea. Korean even has the fixed compliment expression *insaseng-i paruta* 'A person has good greeting/leave-taking manners.' In light of this, the pastor's marked compliment in an addressee honorific form is an importance socialization tool. The pastor is helping the 16-months-old boy see that a deep bow is an expression of respect and acknowledgement towards adults and holds a positive socio-moral value in Korean society.

In this section, we saw how adults' compliments designed with marked honorification target children's morally commendable behavior. Compliments presented here have dual messages, that of complimenting and that of assigning a positive moral value to a child's action, the latter of which is realized by the marked use of honorification. These compliments guide children in interpreting and evaluating their own actions according to the moral norms and frame of understanding of Korean society. In that sense, we can regard these compliments as examples of socializing children through the use of language.

#### *4.2 Showing Gratitude*

Another environment where addressee honorification is used as a socialization tool is when adults express gratitude to children. The two examples in this section (i.e., Excerpts 3 and 4) will show how adults say *komap-supnita* 'Thank you' to children, along with bowing after children hand them an object. By noticing and responding with 'Thank you,' adults ascribe the action of offering or sharing to the children's sometimes ambiguous move of transferring an object and evoke the benefactor-beneficiary relationship between the children and themselves. At the same time, by showing respect and deference to the benefactor, they also display their positive evaluative stance toward such an action, guiding the child to see how sharing can be valued and appreciated in Korean society.

In Excerpt 3, the two children Suji (far right in the pictures) and Yoonchan (in the middle), and the adult Hyorin (in the far left) are sitting on the floor and playing with toys together. In line 1, Suji utters a series of random sounds *eykkepwa?* in a rising intonation while she gazes and holds up a toy milk carton with her left hand towards Hyorin (Figure 3.1). In response, Hyorin, who refers herself as *imo* 'auntie,' issues two consecutive understanding checks (line 2). She first says, *wuyu?* 'milk?' clarifying what Suji meant by the expression *eykkepwa*. She then asks, *imo cwu-nun-ke-ya* 'Are (you) giving (it) to Auntie?,' checking Suji's understanding of her holding up the toy towards her. Her cupping gesture visually complements the clarifying questions (lines 2 and 3, Figure 3.2). Hyorin so far has not employed any addressee honorific forms.

#### **Excerpt 3. Milk Carton**

01 Suji: 에꺼봐? eykkepwa? *((directing the toy milk carton to Hyorin as shown in Figure 3.1))*


03 (2.0)

 *((Still cupping her hands around the milk carton, Hyorin waits for Suji's next move as seen in Figure 3.2. Then, Suji gives the milk carton to Hyorin as seen in Figure 3.3.))*

**Fig. 3.4 Fig. 3.5**

Hyorin, however, shifts from non-honorific to honorific form in line 4 when she shows appreciation, saying *komap-supnita* 'Thank you' to Suji after receiving the toy from her. She employs the deferential addressee honorific verb suffix *-supnita* and bows her head as a sign of deference and respect (Figures 3.4 and 3.5). Note that Hyorin's use of honorification as well as the bowing is marked because both are typically used to show respect towards someone in a higher social position than the speaker. The markedness of the turn is likely to draw the child's attention to the utterance.

Hyorin's expression of gratitude accompanied by bowing is an interactionally complex move that creates a teachable moment from Suji's seemingly incomprehensible (to adults) action. First, by expressing gratitude, Hyorin is ascribing the action of offering to Suji's verbal and non-verbal behavior in line 1. This momentarily reconfigures the interactional roles of Suji and Hyorin from playmates to a benefactor and a beneficiary. At the same time, with the use of honorification, Hyorin's gratitude towards her benefactor, Suji, is expressed with a marked display of respect and deference, as if to show Suji that an action of offering deserves great

respect. Thus, Hyorin's utterance here is imbued with her interpretation and evaluation of Suji's action and designed to guide the child to seeing the desirability and moral value of offering and sharing. Finally, the turn can also be considered from the aspect of the adult's modeling for the child how to show gratitude to someone who shared or offered something. In many aspects, then, this marked utterance *komap-supnita* 'Thank you' can be viewed as a discursive instrument for socializing a child into the moral value of sharing and taking care of other's (potential) needs in the Korean society.

Excerpt 4 below further exemplifies the case. Yumi and her mother have been playing together with the Fisher Price Running Home playset. Yumi is behind the playset, and the mother in front. Prior to the exchange shown in Excerpt 4, the mother had opened and closed the window of the playset, to which Yumi had attempted to pass to her mother a small red ball she was holding in her right hand. This appears to have failed, as the mother had not realized what her daughter was doing. As such, Yumi then tries a different route: the door. She tries to open the door and makes a jerking sound, which leads the mother to re-position herself near the door. In line 1, the mother produces the sounds °*ch ch ch ch*° with a soft voice and thereby initiates a play. Yumi, however, continues to pursue the opening of the door (line 2) and finally succeeds in opening it wide (line 3, Figure 4.1). The mother then produces a celebratory response *ha::!* with a soft voice (line 3).

#### **Excerpt 4. Red Ball**


01 Mom: °ㅊ ㅊ ㅊ ㅊ° °ch ch ch ch° *((Yumi is trying opening the door))*

02 (0.5) *((Yumi is opening the door))*

03 Mom: °하::! ° °ha::! ° *((Yumi finally opens the door wide as shown in Figure 4.1.))*

04 (0.6) *((Yumi holds the red ball, which is out of view due to the Fisher Price Running Home playset, towards her mom as shown in Figure 4.2.))*

06 → 고맙습니다:::, komap-**supnita**:::, Thank-**DC:DEF (HON**) '**Thank (you)!'** *((Mom bows her head as shown in Figure 4.4.))*

In lines 4 through 6, the interaction takes on a new dimension. In line 4, Yumi holds the ball towards her mother's line of sight (Figure 4.2) and, in line 5, the mother responds to her daughter's action with the exclamatory inhaling sound *he!,* which displays her pleasant surprise. Then, the mother reaches her hand towards the ball and grabs it (Figure 4.3). In line 6, the mother says *komap-supnita* 'Thank you,' using the deferential declarative sentence ender  *suptina* and bowing to her daughter in a respectful manner (Figure 4.4). The turn is distinctively marked, as both honorification and bowing are features not normally employed by adults when speaking to children.

The mother's grabbing the ball and offering an appreciative remark frames the child's simple action of transferring an object as an act of offering intended for the mother's benefit. This establishes a benefactor-beneficiary relationship between Yumi and the mom. Expressing respect and deference to her benefactor, Yumi, as her response to Yumi's offering, the mother is also demonstrating how sharing with others is appreciated and valued. In this sense, the mother's *komap-supnita* 'Thank you' accompanied with bowing is a socialization device that guides Yumi to seeing the moral significance of sharing one's possessions for other's benefits in the Korean society. Additionally, Yumi can also learn how to appropriately show gratitude when she is on the receiving end of offering and sharing.

#### *4.3 Issuing Directives*

Our data show that adults sometimes use honorification when giving directives to children. As "utterances designed to get someone to do something," (Goodwin, 2006, p. 515) directives may take syntactic forms associated with offers, requests, orders, prohibitions, proposals, and so on (e.g., Ervin-Tripp, 1976). In this section, we show two examples where honorification is used in request and proposal, respectively. As will be shown, marked uses of honorification co-occur with a set of discursive and non-verbal resources to create a subtle interactional pressure on the children to cooperate with the adult.

In the first example, Homin, his mother, and his grandmother are sitting together on the floor. The grandmother is trying to feed Homin ground pear with a spoon. Homin, however, is not interested in the pear and grabs the spoon in front of him (line 1, Figure 5.1). Ensuing is a gentle tug-of-war between Homin, who will not let go of the spoon, and the grandmother, who is trying to take the spoon out of his hands.

#### **Excerpt 5. Spoon**

01 Homin: (.)

 *((Homin grabs the spoon with his right hand as seen in Figure 5.1.))*

**Fig. 5.1**


#### **Fig. 5.2**





#### 22 (3.5) ((*Grandma goes to the kitchen with the spoon and pear.*))

Throughout lines 5 to 20, the grandmother makes multiple attempts to have Homin let go of the spoon. As Homin increases his grip of the spoon even further by holding it with both hands (line 3, Figure 5.2), the grandmother tells him to release the spoon, repeating *nwa* 'Let go' quickly and with pitch changes (lines 5 and 6). After seeing that this does not work, she changes her hands position for a tighter grip (Figure 5.3). She then issues prohibition, *an tway* 'No,' multiple times in a quick tempo (lines 9 and 11). The prosodic features here create choppiness in rhythm and a sense of urgency in taking the spoon from him.<sup>4</sup> After the pear spills onto Homin's hands (line 13), the grandmother tries to take the spoon again, saying *incey nwa nwa* 'Now let go! Let go!' (line 18). Note that so far, both directives, *nwa* 'Let go' and *an tway* 'No,' have been produced with a non-honorific ending. Homin is still holding the spoon (line 19), and the grandmother tries to take the spoon yet again. This time, however, she changes her strategy. She says *cwu-sey-yo::* 'Give (it to me), please' (line 20), using the subject honorific suffix *-si* as well as the addressee honorific sentence ender *-yo* and speaking noticeably slow. Homin finally lets go of the spoon (line 20, Figures 5.4 and 5.5) while the grandmother elongates the last syllable of *-yo::.* Then, the grandmother compliments him, saying *ayu ippe* 'Wow, good boy!' (line 21).

A close examination of the grandmother's utterance *cwu-sey-yo* 'Give (it to me), please' reveals that she is employing a set of discursive strategies in this turn in pursuit of Homin's compliance. First, the prosodic shift from quick, choppy speech to a noticeably slow and mild delivery differentiates this turn from its surrounding talk, potentially drawing Homin's attention to what is being said. A shift in register to honorifics also makes the turn distinct.<sup>5</sup>

Addressee honorification employed in this turn, while adding to the turn's saliency, also introduces a new interactional dynamic that encourages the child's compliance. As noted, statusincongruent use of honorifics can be examined as an indexical sign (e.g., Agha, 1993; Cook, 1997; Duranti, 1992). Indeed, the honorification here is best understood as a creative or performative index to a high, respectable social status. By the virtue of being addressed in honorifics, Homin momentarily takes a social persona with a high, respectable social status, who is expected to act accordingly. In other words, the grandmother is putting a subtle interactional pressure on Homin to behave like someone worthy of respect (i.e., by cooperating with the adult and letting go of the spoon). The shift in the kinds of directive issued, from an order ("Let go!") and prohibition ("No!") to a gentle request ("Give (it to me), please!"), further contributes to the construction of the child's new social persona while lessening the grandmother's authority.

Eliciting a desired action from the child is not the only function of honorification here. In the sense that honorification was used in the context of the adult's trying to 'correct' the child's behavior, it is also evaluative and can be considered as an instrument for socialization. That is, the grandmother is conferring Homin a respected social persona and nudging him to act

<sup>4</sup> We see that the concern for the two adults, the grandmother and the mother, is to avoid the pear spilling out of the spoon and making a mess. In line 3, the mother cups her hands to catch the pear in case it falls. In line 12, the grandmother is trying to eat the pear herself before it falls but she fails, and the pear finally slides out of the spoon and lands on Homin's hand.

<sup>5</sup> Securing a child's attention to a directive and the task at hand has been shown to often lead to the child's compliance in family interaction (Goodwin, 2006). Here, the grandmother is using the prosodic and register shifts to make her directive salient and thus more likely to draw attention.

accordingly in response to his persistent non-compliance. This may signal that Homin's behavior so far lacks such respectable qualities. In other words, by using honorification in this context, the grandmother is evaluating Homin's behavior and guiding him to see that complying with the adult's directives is what is desired and appropriate. The message is further reinforced through the grandmother's complimenting Homin in response to his letting go of the spoon in line 21.

The next example shows a similar use of addressee honorification, this time from the outset of an activity. In Excerpt 6, the grandmother is getting ready for a snack time for another grandchild of hers, Suji. Suji is seated in a high chair. The high chair has a tray that can be stored in the back of the chair when not used. To use the tray, two bars on both sides of the tray must be put into the slots under the arm rests so the tray can be securely placed on top of the arm rests. Suji is currently seated and looking out at a 2 o'clock position. Each of her arms are placed on the arm rests. The grandmother is starting to install the tray, bringing out the tray (Figure 6.1). In order to put the tray into the slots, she needs to have Suji remove her arms from the arm rests. Note that the grandmother normally uses non-honorific forms when speaking to Suji, as they are the unmarked, default choice for an adult talking to a child.

**Excerpt 6. Tray**

01 (2.0) *((Grandma takes out the tray from the back of the high chair.))*

02 → Grandma: 자: 끼웁시다, ca: kkiwu-psita, alright. stick-**PR:DEF (HON) 'Alright, let's stick (it) in!'** *((Grandma is rearranging the tray in its horizontal* 

*position. Suji's hands are on arm rests as shown in Figure 6.1.))*

03 (0.5) *((Grandma is bringing the tray to the front.))*


*((Suji lifts both arms as shown in Figure 6.2.))*

Rearranging the tray in its horizontal position, the grandmother says *ca* 'Alright,' a discourse marker that often prefaces proposals or commands (Yim, 2005) (line 2). A proposal comes next, as the grandmother says *kkiwu-psita* 'let's stick it in!' Here, the grandmother is using the honorific propositive suffix -*upsita* instead of the non-honorific -*ca*. At this point, Suji's arms are blocking the slot (Figure 6.1). The grandmother continues, adding the reason for her proposal, saying *ikes an-kkiwu-myen yeki mos-anc-key tway iss-eyo* 'If (we) do not stick this in, you cannot sit here (by its nature)' (line 4). Again, she uses an addressee honorific sentence ender -*yo*. Suji starts lifting her arms almost at the same time with the grandmother's account (line 4, Figure 6.2).

As in the previous example, honorification occurred in the context of eliciting a child's cooperation for the task at hand. The interactional work being done here—drawing the child's attention and invoking a social figure who deserves respect and act respectably—is similar as well. First, in the beginning of the excerpt, the grandmother is using lexical and non-verbal resources to draw the child's attention. The discourse marker *ca* 'alright' strongly projects an activity that directly involves the recipient, such as proposal or command, thereby attracting the child's attention to the upcoming talk. The physical action of bringing the tray right in front of the child also directs the child's attention to what is going on. The marked use of honorification in the proposal *kkiwu-psita* 'let's stick (it) in' makes the utterance salient.

Second, addressee honorification is used indexically as a way of guiding the child to compliance. Addressee honorification endows Suji with a respectable social persona and creates an implicit pressure for her to behave accordingly and cooperate with the grandmother. The construction of Suji's new, albeit, temporary identity is further bolstered by the grandmother's choosing proposal (i.e., 'Let's stick (it) in.') over command (e.g., 'Lift your arms') to get Suji to do the job needed, as well as her adding an account for proposing, also produced in the honorific form (line 4). Proposing indexes a less power disparity between the grandmother and Suji than commanding would do, thereby highlighting Suji's new role as a respected figure. Similarly, the grandmother's offering an account for the proposal sustains Suji's elevated status—Suji is treated as needing to be persuaded rather than to be told, as if she possesses power and agency to reason and reject.

Therefore, honorification is used strategically as a discursive tool for gently urging the child to comply with the adult's directive. Even though the grandmother's directive in this example is not evaluative in the same way the other grandmother's directive was in Excerpt 5 (i.e., it was not produced in response to the child's non-compliance), it still implicitly associate being a respectable person with complying with adults' directives. In this sense, addressee honorification employed by both grandmothers in Excerpts 5 and 6 can be considered as an instrument for socializing Homin and Suji into the moral norms expected of children.

#### **5 Discussion**

This study examined adults' use of honorification toward children in terms of indexical meanings of addressee honorifics and how they are strategically mobilized as a resource for socializing children. Our analysis show that adults' use honorification toward children has the quality of evaluating children's behavior. More specifically, the deferential sentence ender *supnita* and the polite sentence ender -*yo* (sometimes accompanying the subject honorific suffix  *si*) appear to index respectability or admirability. Thus, when adults use honorification in complimenting, showing gratitude, and issuing directives, they display their evaluative stance, at times acknowledging the presence of admirable and respectable quality, at other times revealing the lack thereof. Being repeatedly exposed to such feedback from adults, children will eventually internalize the norms and frames of understanding applicable to their actions and become a competent member of the Korean society who can act, feel, and think in a culturally patterned way.

The use of addressee honorification, however, is not an isolated phenomenon. We have shown that when adults shift to addressee honorification, this shift is consistently accompanied by prosodic and/or non-verbal behaviors and even sequential cues. These different modes of semiotic resources reinforce the respectability and admirability conveyed by addressee honorifics. They also serve to highlight the honorific utterances as something special and deserving of attention. In other words, adults' evaluative utterances tend to be "spotlighted" multimodally, through the marked use of honorification and other types of co-occurring semiotic resources, facilitating children's learning from adults' speech addressed to them.

The evaluative and attention-grabbing functions of adult's addressee honorifics, coupled with various co-occurring semiotic resources, can be further examined by referring to Husserl's notions of modification and attitude as understood by Duranti (2009). Duranti mentions how children relate, orient, and attend to the world through thinking, acting, and feeling can be guided and shaped through interaction with adults. In explaining the relationship between child and the world, as well as the changes in that relationship, Duranti introduces Husserl's notion of "natural attitude" and "theoretical attitude." Duranti understands the conception of the "natural attitude" as "the practical, moral, and aesthetic stance that we ordinarily take toward the surrounding world" (p. 213), and the "theoretical attitude" as adopting a reflective, evaluative stance towards "natural attitude" by stepping out of it. Duranti explains that the shift from natural attitude to theoretical attitude is what Husserl calls "phenomenological modification."

In light of the aforementioned notions of Husserl, we propose that adults' use of addressee honorifics towards children scaffolds the children to take a theoretical attitude toward the activity they are engaged in. By issuing honorific utterances to children, adults introduce a frame of understanding in which particular social roles, norms, and expectations become relevant and children's behaviors are judged and evaluated (e.g., a) *enni nolus-ul ttokttok-hi ha-si-ney-yo*, '(You) are doing (your) part as an older sister very well.': An older sister should take care of her younger sister's needs, and you fulfil your role as an older sister very well; b) *cwu-sey-yo ::* 'Give (it to me), please.:' You should comply with my command to let go of the spoon, but you're refusing to do so. Please follow my instruction as a respectful individual). Furthermore, the marked design of such honorific utterances, combined with other semiotic resources, directs children's attention to the locally meaningful social roles, norms, and expectations. By doing so, it provides children with an opportunity to reflect on their behaviors within the proposed frame of interpretation. In this sense, adults' use of addressee honorifics can be regarded as a linguistic resource that mediates children's phenomenological modification. This, in turn, may contribute to the development of children's social awareness and social responsiveness, which are key elements of moral personhood (Ochs and Izquierdo, 2009, p.391).

The findings on the status-incongruent and creative or performative use of adults' addressee honorifics thus demonstrate that the grammatical features of deference and respect can be utilized as a socialization device to cultivate a respectable and admirable member of society. In other words, addressee honorifics, which are sometimes employed to display the speaker's politeness and deference towards the addressee, serve as tools for fostering politeness, courtesy, and social responsiveness in the next generation.

#### **6 Conclusion**

We, so far, have investigated the socio-pragmatic functions of addressee honorifics used by adults when talking to children in an attempt to deepen our understanding of status-incongruent and creative or performative use of addressee honorification. Using multimodal discourse analysis and language socialization approaches, we have examined 62 instances of such a phenomenon from video-recordings of mundane interactions from five different Korean families (245 minutes). The results of analysis have shown that adults can use addressee honorifics towards children, combined with other various semiotic resources, to give compliments, show gratitude, and issue directives. In these interactional environments, adults utilize addressee honorifics as a means to socialize children into the socio-moral values of Korean society. This is achieved through evaluating children's behaviors as worthy of respect or admiration, or lacking such qualities, and drawing children's attention to both the action and content of their honorific utterances. Gentle voice quality and bodily behaviors such as caressing, patting children, and bowing reinforce the use of addressee honorifics as an index of respectability and admirability. Tempo adjustment in speech delivery, object manipulation for accomplishing relevant actions such as moving a tray, and halts in sequential progress contribute to supporting the markedness and saliency of adults' use of addressee honorific utterances to children.

We have proposed that the evaluative and attention-grabbing functions of addressee honorifics, reinforced and supported by other types of semiotic resources, can be considered as a linguistic resource that mediates children's phenomenological modification. They scaffold children's reflection on their behaviors against the backdrop of the social norms, expectations, and preferences of Korean society. Furthermore, they may contribute to the development of

children's social awareness and social responsiveness. The use of adults' addressee honorifics towards children as an interactional resource for children's socialization into Korean socio-moral values thus show that honorifics are not only a linguistic resource for displaying politeness, but also a resource for fostering a courteous person.

We would like to conclude this study by suggesting future research agendas concerning status-incongruent and creative or performative use of adults' addressee honorification towards children. In our study, we mainly focused on the juxtaposition of honorifics with non-honorifics and their noticeable contrast, rather than delving into differences between the sentence ender *supnita* of the deferential speech level and the sentence ender -*yo* of the polite speech level. Since there are studies that have identified distinct indexical meanings of the deferential *supnita* and the polite -*yo* (e.g., Brown, 2015; Lee & Brown, 2022), it would be a fruitful endeavor to further examine whether adults utilize these two different honorific levels distinctively as a resource for socialization when speaking to children. Additionally, it would be interesting to explore cases where the indexical meanings of addressee honorifics (i.e., respectability and admirability) are incompatible with the meanings created by other semiotic resources (e.g., disrespectful prosody, staring gaze, etc.). Examining the discrepant cases may shed light on more intricate interplay between the addressee honorifics and other types of semiotic resources in socializing children into a society's norms and expectations.

#### **References**


Brown, L., & Winter, B. (2019). Multimodal indexicality in Korean: "Doing deference" and "performing intimacy" through nonverbal behavior. *Journal of Politeness Research, 15*(1), 25–54.

Clancy, P. M. (1985). The acquisition of Japanese. In D. I. Slobin (Ed.), *The crosslinguistic study of language acquisition Vol. 1: The data* (pp. 373-524). Hillsdale, NJ: Erlbaum.


#### **Appendix A: Transcription Conventions**



#### **Appendix B: Abbreviations**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Solidarity through Negotiated Interactional Identities in Korean**

#### **Mee-Jeong Park<sup>1</sup>**

**Abstract** This paper shows how Korean speakers use different strategies to increase solidarity among newly acquainted interlocutors in performing common tasks by co-constructing through the negotiation process of their interactional identities and adjusting themselves to the right level of intimacy and/or politeness within the given interaction. According to Swann (1987, 2008), "identity negotiation" refers to the processes where interactants try to find a balance between their interactional and identity-related goals, keeping a conflict-free relation between their interpersonal and intrapersonal interactions.

 The ways in which Korean speakers negotiate their situational and interactional identity will be illustrated using excerpts taken from TV talk shows, reality shows, or dramas where different participants achieve what is considered an adequate level of intimacy with their conversational partners within the given tasks as the show participants. In interactions where Korean speakers meet for the first time, it is very common to see how they exchange personal information. Among them, interlocutors' age is very often exchanged at the very early stage of their encounter. In many reality shows and talk shows on Korean TV, participants often start their first-time encounter by asking about their age and work-related backgrounds. Interlocutors achieve an increased level of intimacy by assigning new interactional identities to themselves, that of (a) friends (=same age), (b) siblings (=different age), or (c) senior/junior (work-related). Oftentimes, this process is streamlined by adjusting their speech style and/or address terms that match their newly constructed identities in order to successfully perform their common tasks.

**Keywords** Korean Speech Style Korean Politeness Identity Negotiation Speech Style Shift Panmal (half-talk)

<sup>1</sup> M-J. Park ()

 Department of East Asian Languages and Literatures, University of Hawaiʻi at Mānoa, Honolulu, Hawaiʻi, USA.

e-mail: meejeong@hawaii.edu

#### **1 Introduction**

Korean is well known for its intricate system of honorifics and various speech styles, each reflecting different levels of politeness between speakers. When any two Korean speakers interact with each other, they make a choice on a particular speech style that reflects the power and solidarity relationship between the two interlocutors (Sohn, 1981). Scholars have proposed up to six distinct speech styles, from the most polite to the least, such as Deferential, Polite, Blunt, Familiar, Intimate, and Plain style, from which only four styles are frequently used (Sohn, 2001; Song, 2005; Brown, 2011). Regardless of the total number of speech levels suggested by different authors, they are categorized based on different features conveyed within these styles, such as ±polite, ±formal, ±distant, and ±honorific, among others.

In an interaction between two newly acquainted adult speakers of Korean, it is most common that they use a mixture of Polite and Deferential styles to each other. However, when the relationship between the two becomes more intimate, they either decide to drop the polite speech style and use the intimate (non-polite) form of speech style under mutual agreement, or simply try to change their speech style slowly over time. In the latter case, speakers go through a transitional stage where they switch from polite to non-polite style, during which time, speakers use different strategies to make the transition as smooth as possible by minimizing any FTA (face-threatening acts, Brown and Levinson, 1987). Depending on their relationship, the two interlocutors either progress toward a completely intimate relationship in terms of their speech style or make their polite and rather distant stage a permanent one. This is especially the case among aged speakers, whereas teenagers to roughly 30s will be relatively more at ease to change their speech level in a relatively short time depending on the nature of their relationship.

Korean speakers frequently switch between speech styles within a single interaction, using a mix of styles when addressing the same interlocutor. Numerous studies have explored how speakers strategically shift between polite and non-polite speech styles in various contexts, such as classrooms, interviews, debates, and more, to achieve specific interactional goals (Lee, 2000; Eun & Strauss, 2004; Strauss & Eun, 2005; Kim & Suh, 2007; Park, 2014; Kiaer et al., 2019). Shifting from non-polite to polite speech style typically indicates a shift in formality. However, the transition from polite to non-polite speech style requires careful planning and mutual agreement, a process even native speakers find challenging and sensitive.

As mentioned earlier, there are several different versions of Korean speech styles, but this study adopts the version provided by Sohn (1999), which is included in the table below. Each of the following suffixes is attached to the verb stem to denote different sentence types as well as different degrees of politeness.


#### **Table 1** Korean Speech Style by Sohn (1999)

According to Kiaer et al. (2019), in Korea, numerous news reports have documented serious conflicts, including cases of physical violence, between two adult passersby triggered by the inappropriate use of *panmal 2* 'half-talk' by one of them. When used appropriately between suitable interlocutors, *panmal* is considered to enhance the level of solidarity between them. However, it can be offensive when used with speakers who do not share an appropriate level of solidarity. Nevertheless, even native speakers often find it challenging to determine the perfect stage in their relationship to shift to *panmal*. Some speakers gradually transition from a polite to a non-polite speech style when they become close enough to use *panmal* as a sign of intimacy and informality, making this transition permanent. Obviously, the choice of speech style in Korean is not straightforward but instead involves a complex process in which interlocutors continually monitor the nature of their interaction in terms of formality, distance, deference, and other factors (Kim-Renaud, 2001; Song, 2005; Sohn, 2001; Choo, 2006; Brown, 2011).

This study focuses on how newly acquainted Korean interlocutors negotiate their level of solidarity to avoid conflicts that can arise from failing to reach a mutually agreeable level of intimacy while performing common tasks. They do so by co-constructing their interactional identities and adjusting themselves to the appropriate level of intimacy and politeness for the given interaction.

Interlocutors achieve an increased level of intimacy by attributing new interactional identities to themselves. This process is facilitated by adjusting their address terms and/or speech style to align with their newly constructed identity. This adjustment enables them to effectively carry out various common tasks, including work-related tasks and personal activities, among others. The ways in which Korean speakers negotiate their interactional identity will be illustrated through excerpts taken from TV programs, including talk shows, reality shows, and dramas. The data reveals three commonly observed types of negotiated identities: (a) the identity of *chinkwu* 'friends' when the two newly acquainted interlocutors are of the same age; (b) the identity of siblings, including *hyeng/oppa* (older brother) or *nwuna/enni* (older sister) when the two interlocutors are of different ages; (c) *senpay* 'senior' or *hwupay* 'junior' if they belong to the same professional field or have attended or go to the same school.

<sup>2</sup> *Panmal* in Korean literally means 'half talk,' which includes the Intimate and Plain speech styles, or non-polite versions, provided in Sohn's (1999) classification in the table above.

#### **2 Identity Negotiation**

According to Swann (1987, 2008), "identity negotiation" refers to the processes in which interactants attempt to strike a balance between their interactional and identity-related goals, aiming to maintain a conflict-free relationship between their interpersonal and intrapersonal interactions. According to this theory, there is a competing process in which two interlocutors, the "perceiver" and the "target," may initially hold different perspectives about each other's identities but will eventually reconcile over time as their level of "congruence" increases. Therefore, Swann (1987) asserts that sometimes, the resolution of "conflicting agendas" occurs during the identity negotiation process.

In interactions where Korean speakers meet for the first time, it is very common to observe how they exchange personal information. Among these details, interlocutors' ages are frequently exchanged at the initial stage of their encounter, either through direct inquiries about their age or indirect expressions of curiosity regarding their interactants' ages (e.g., mentioning their Chinese horoscope birth year or college entry year). In many Korean reality TV shows and talk shows, participants often begin their first-time encounters by inquiring about each other's ages, especially if their interactants appear to be of a similar age. However, they may refrain from asking about age if their interactants visibly appear to be 10 or more years older than themselves. In real-life situations, the nature and formality of their interaction (e.g., business versus personal settings) would influence the appropriateness of such open inquiries about each other's age.

According to the data, when two interlocutors exchange personal information, including their age, and wish to or need to continue their relationship for any reason, they tend to adjust their relationship based on this age information. This process leads to identity negotiation, resulting in an increased level of solidarity and facilitating a continued and more comfortable relationship between the interlocutors. The ease of this negotiation varies, being relatively straightforward in some interactions and more complex in others.

#### **3 Data**

It would have been desirable to capture real conversations between two newly acquainted adult speakers in real-life situations to support the claim made in this paper. However, collecting such interactional data is technically impossible. Instead, this study relies on data from Korean TV programs, including four episodes from two TV dramas and nearly 30 episodes from eight different talk shows, included in the table below. For the talk shows, only the first 10 minutes of each episode, starting from the latest episodes were used to examine how the guests interacted with each other. In most cases, guests greet each other and inquire about each other's age during the opening part of the shows. In some episodes, when the guests are in their mid-40s or older, they do not discuss each other's age. Sometimes, when their ages are already well-known, neither the emcees nor the guests mention their age. Therefore, I found only 30 instances where age-based identity negotiations were discussed in the shows among the over 100 episodes I searched for.


**Table 2** List of the Korean TV programs used in this study

Based on the episodes of these talk shows, guests in their teens to late 30s are more inclined to attempt to negotiate their interactional identities based on their age. However, individuals over the age of mid-40s and beyond rarely inquire about each other's age, and even if they do, their age information does not typically lead to an identity negotiation aimed at increasing solidarity. This study focuses on tendencies rather than conducting a quantitative analysis of instances in which interlocutors attempt to negotiate their identities.

This study analyzes the data using Discourse Analysis and transcribes the dialogues using Yale Romanization. For excerpts that illustrate speech style shifts, a four-tier transcription is employed, including the Korean text, romanization, word-by-word English gloss, and English translation. For excerpts where linguistic structure is not as relevant, a two-tier transcription was used including Korean text and English translation.

#### **4 Negotiation of Interactional Identities**

As mentioned earlier, the ways in which Korean speakers negotiate their age-based interactional identities are classified into three groups: (a) the identity of *chinkwu* 'friends' when the two newly acquainted interlocutors are of the same age; (b) the identity of siblings, including *hyeng/oppa* (older brother) or *nwuna/enni* (older sister) when the two interlocutors are of different ages but have less than a 10-year difference, roughly; (c) *senpay* 'senior' or *hwupay* 'junior' if they belong to the same professional field or have attended or go to the same school but with different starting years. Each of these three types will be illustrated using the excerpts taken from TV programs.

#### *4.1 Negotiated identity #1: Friends = same age*

In the following excerpt, the five emcees of the show are conducting a special interview with two guests they call "master." They are very famous movie stars in their late 40s. Both actors are among the two most famous actors in Korea. One of the emcees, DH in his 30s, approaches one of the two masters, WS, and asks about his age.

#### **Excerpt #1 [All the Butlers, ep. 232]**


4 → Subtitle: "두 사부님은 동갑내기 친구" "*The two Masters are friends of the same age*."

The term *chinkwu*, which means 'friend,' is, of course, used in the normal sense for those who have known each other for a long time. However, this term is often used in conjunction with the term *tongkap*, which means 'the same sexagenary cycle,' to refer to someone of the same age or born in the same year. The two Masters appearing in the show are long-time friends who have been working in the movie business together, working under the same entertainment management company, neighbors in the same apartment, and so on. Still, the term *chinkwu* in this context specifically refers to their same age. In fact, the term *chinkwu* 'friend' is easily used between two newly acquainted people, but this practice is not as common among elderly individuals in their 60s and above, roughly.

In the following excerpt taken from the reality/talk show, the guest singer Crush is sharing some stories with the emcees about his friendship with the highly acclaimed South Korean professional soccer player Son Heung-min, who is currently playing for the Premier League club Tottenham Hotspur in the UK. Crush is explaining how he became friends with Son Heung-min.

#### **Excerpt #2 [My Little Old Boy, ep. 311]**


*Uh, six years ago, I had..*


The emcee DY initiates the conversation by inquiring about how Crush and Son Heungmin became friends in lines 1 and 3. Given that Crush may not be considered one of the top singers in South Korea, while Son is a globally renowned soccer player, DY uses the expression "unexpectedly close to Son." Crush then proceeds to offer background information regarding the beginning of their friendship in lines 4 through 12. In line 6, Crush explains that they met in Europe when Son attended the concert Crush organized, and in lines 10 and 12, they grew close because they discovered they were of the same age and they still keep in touch. In several excerpts taken from TV programs, it is very common to see people emphasizing that sharing the same age, among other factors, plays a key role when two newly acquainted individuals become close in a relatively short amount of time. This is likely the reason why people label individuals as "friends" when they are of the same age, even before they develop a close bond.

In the two previous excerpts, the show's guests talk about their relationships with their long-time friends. In the following excerpt, the two interlocutors are not friends. However, they

know each other because they are both top celebrities in the Korean entertainment field, and they are meeting in person for the first time. In this excerpt, the main emcee of the show, JS, is interviewing one of the greatest Korean musicians of the 90s, Seo Taiji (TJ), who has not been very active in recent years. TJ used to be known as "the President of Culture" for many years and is recognized as one of the hidden-in-the-veil celebrities. Consequently, his personal life has remained relatively unexposed to the public.

#### **Excerpt #3 [Happy Together: Special Episode – Seo Taiji (2014)]**

This excerpt is taken from the opening portion of a two-hour-long TV talk show, with the entire episode dedicated to an interview with TJ. In an effort to make his life more open to the public, TJ participated in a one-on-one special interview with JS, one of the top Korean emcees.



In the opening part of the interview, JS mentions the fact that they were both born in the year 1972, as shown in lines 1-3. TJ acknowledges this in line 4, and JS reconfirms that they both belong to the animal zodiac sign "rat," as indicated in lines 6-8. In Korea, the majority of people calculate their age based on the Lunar Calendar<sup>3</sup> , making the animal zodiac sign crucial when determining people's age. Once it was confirmed that they were born in the same year and share the same zodiac sign, TJ mentions that they are indeed of the same age and, consequently, friends in line 9. He repeats the word "friend" three times in a row to emphasize that they share the same age and, thus, are friends.

The fact that they both share the same age, along with TJ's use of the term "friend," suddenly offers an opportunity for the two to strengthen their relationship and become closer and more intimate in less than an hour since they first met in person. Furthermore, this situation opens up the possibility for JS and TJ to engage in *panmal* with each other without worrying about being rude to one another. In line 10, JS asks TJ if it is okay to use *panmal* with him. When TJ responds positively, JS rephrases his question in line 12, asking for TJ's permission to use *panmal* briefly. Once TJ agrees with a 'yes' to his question, JS addresses TJ using the plain vocative case particle -*(y)a*, and TJ responds in kind, also using the same particle -*a* in lines 14-15. They both share hearty laughter right after lines 14-15, likely because they could engage in *panmal* with one of the top Korean celebrities without any feelings of guilt and because of the bashful sensation of developing an intimate connection in such a short amount of time.

Since this is a one-to-one interview, they both use a polite speech style throughout the interview. However, this brief use of an intimate speech style in lines 14-15 demonstrates how they associate the *panmal* style with the identity of a 'friend,' someone of the same age. This may not be the catalyst for a lifelong friendship and ongoing use of an intimate speech style with each other, but it does seem to be an acceptable reason for initiating the transition from polite speech to *panmal* without concern about appearing rude. So, JS and TJ adapt their speech style to their relationship, which transitions between that of host and guest, and that of friends, at different stages during the interview as shown in the figure below.

**Fig. 1** A summary of the interactional identity and speech style shift in Excerpt #3

The process of negotiating their age-based identities may appear easy and proceed smoothly in some cases. However, it is not always amicable and straightforward, as interlocutors may have disagreements during the negotiation process. In the following excerpt, four emcees discuss the age issue with their guest, Pak Seri, the most well-known former Korean professional female golfer, who was inducted into the World Golf Hall of Fame in 2007.

#### **Excerpt #4 [Problem Child in House, ep. 159]**

1 HD 저랑 갑이에요.

<sup>3</sup> South Korea's traditional age-counting custom considers every person 1 year old at birth and adds another year when the calendar hits January 1st. While the new law is the country's latest attempt to retire that method and standardize international ages based on the passing of birthdays, it is not immediately clear what will actually change. From AP News (June, 2023)

(pointing at Seri with his finger) We're of the same age.


#### 17 All 하하하하 Hahahah

HD, one of the male emcees, asserts that he and Seri are of the same age during the opening stage of the talk show, affirming that he is aware that Seri was born in 1977. However, one of the female emcees, EI, points out that HD was born in 1978, which is not the same year Seri was born. Another male emcee, KH, attempts to intervene, advising HD not to use *ppalun* 'early' calculation. In Korea, when people negotiate their age-based identities, they often calculate their age difference based on the month they were born in comparison to someone of the same age. Those who were born in January often consider themselves friends with those who were born in the previous year, as they share the same animal zodiac sign. In other words, for those born in January, it is still counted as December of the previous year in the Lunar calendar.

In line 7, Seri disagrees with HD's claim that they are of the same age. She asks HD if he belongs to the snake zodiac sign. Instead of answering Seri's question, HD asks her what difference it makes to have a different animal sign, as indicated in line 10. Seri continues to insist that they are not of the same age because he may not have the same zodiac sign as hers, the snake. To verify this, KS asks HD which month he was born, to which he responds, "February." When Seri asks HD the same question again in line 15 about his zodiac sign, HD says he would like to be an early snake sign. Then everyone laughs. In this excerpt, HD desires to be regarded as Seri's friend and attempts to negotiate his age-based identity with her. He makes several attempts, but Seri remains steadfast in her stance regarding age calculation. This form of age calculation is frequently observed among male speakers as they negotiate their relationships. Often, it appears that certain individuals are unwilling to bestow the title of "friend" upon their conversation partners unless they are, in fact, of the same age with the same zodiac sign.

Continuing from the previous excerpt, in line 18 below, SK allows Seri to determine how her relationship with HD should be defined. Seri's ultimate conclusion is "younger brother," which implies that they cannot be classified as "friends" based on their age. Consequently, there is an audible collective sigh of "ahhh" in line 22, signifying a sense of disappointment. Seri's rejection of labeling her relationship with HD as "friends" is primarily based on two reasons. Firstly, she repeatedly emphasized the importance of age calculation, asserting that only those born in the same year and zodiac sign can be deemed friends. Secondly, as mentioned in line 23, she is uncomfortable labeling a newly acquainted person as a "friend" solely because they share the same birth year. She seems to believe that such a designation requires a long-time relationship, and KH supports Seri's view, in line 24.


How can we be friends?


23 → SR 왜냐면 솔직히 친구들도 같은 동갑 77 이라고 해도 우리가 같은 연배지만 처음 보는데 갑자기 '야, 친구야' 하기 좀 어렵지 않아요? To be honest, even for friends who were born in '77, **although we are of the same age, how can I suddenly call them "hey, friend"?** Isn't it a bit difficult?

24 KH 저는 이해해요. I do understand.

As observed in the above excerpt, not all negotiations of age-based identities are successful or straightforward, and this can be attributed to various factors. One such factor is the diverse range of personalities and beliefs held by individuals, leading to a multitude of possible interaction outcomes. Nevertheless, there remains a prevailing tendency among the majority of Korean people to seek to enhance their solidarity or connection with their conversational partners through this negotiation process, regardless of whether it ultimately succeeds or not.

#### *4.2 Negotiated identity #2: Siblings = Different age*

As seen from the last part of excerpt #4 above, the use of sibling terms with close friends or even acquaintances between interlocutors of different ages is very common among Korean speakers. In Korean, the sibling terms form a dyad based on the speaker's gender and elderliness (whether they are older or younger than the addressee), as shown in the following table.


**Table 3** List of the Korean address terms for siblings for male and female speakers

When two interlocutors negotiate their age-based identities and discover that they are not of the same age, the next option is to opt for the sibling terms. In this case, they need to determine who is older and who is younger so that they can decide what kind of relationship they should maintain, if they choose to continue their relationship for any reason. As shown in line 18 in the excerpt above, Seri determines that HD should be labeled as a "younger brother" as he was born in 1978, the year following her birth year. Additional excerpts are presented in the next section to further examine instances where participants attempt to negotiate their level of intimacy while carrying out their tasks on these TV shows.

In the following excerpt, the female host discusses her past part-time job experience when she was just starting her career as an actress. One of her most memorable experiences was working in a sales position at a mobile phone store, where she earned the title of "Top Salesperson." She begins by sharing her strategies for selling mobile phones with the emcees and other guests. As she unfolds her story, she engages in roleplay with a male guest and elaborates on the strategies that contributed to her success as a top mobile phone salesperson. During her explanation, she highlights two key strategies when interacting with customers: (a) Rapidly explaining the device's features to create an impression of expertise, making it difficult for the customers to grasp all the details while convincing them that she knows the best deal; (b) Building a close rapport with the customers by using *panmal*, which is justified by her appearance that looks older than the customers. The excerpt below illustrates her second strategy.

#### **Excerpt #5 [Radio Star, ep. 513]**


*That is... a bit intimate.. meaning, it shows intimacy.* 


In the excerpt above, line 3 features a follow-up of some omitted interactions where the female guest, F, demonstrates to the male guest speaker, M, how a rapid sales process can increase the likelihood of closing a sale. When M raises a question about a different item, in line 4, expressing dissatisfaction with the initial choice, F alters her approach, as evidenced in the remaining lines. She now attempts to establish an intimate connection with the customer by using the term *tongsayng*, which means a younger brother or sister, in line 5, to justify her use of a more casualspeech style in addressing the customer, in line 6. This, in turn, strengthens their relationship, transforming the salesperson into a nurturing *nwuna* and the customer into a cherished *tongsayng*. Consequently, F becomes a dependable figure who treats the customer as her own younger sibling, offering guidance and advice with sincerity and integrity, as noted in lines 8 and 10. The customer need not worry about anything but should simply follow her instructions and heed her

recommendations to secure the best possible mobile phone deal. In line 17, her ultimate strategy is to pretend to offer a "rare and exclusive discount," a secret so closely guarded that not even the store manager is supposed to be aware of it.

As demonstrated in their role-play, the salesperson's most effective strategy for gaining the customer's trust was to cultivate a sense of intimacy between them, thereby enhancing the customer's trust and achieving the desired level of closeness. To attain this, she proposed a new relationship dynamic between them: that of an older sister and younger brother. Her shift from a polite speech style to *panmal*, along with the man's acceptance of this sudden shift, ultimately solidified this newly established identity. In this way, she successfully maintained a balance between her interactional and identity-related goals of intimacy, which ultimately contributed to her becoming a top salesperson.

In the above excerpt, the saleswoman and the customer, both of whom appear in the roleplay, negotiated their age-based interactional identities with the understanding that they would not see each other again, indicating a very short-term relationship. In the following excerpt, on the other hand, the four interlocutors are about to embark on a new project in a reality TV show. They anticipate living together for the next few weeks and frequently crossing paths, as they are all actors working in the same field. This excerpt is taken from the introductory part of the first episode, where two close actors, JW and JH, in their early 40s and late 30s, arrive in a restaurant room to have dinner with two younger actors, MW and JK, in their early 30s and late 20s. All four of them meet for dinner before flying to New Zealand for their road trip shoot.

#### **Excerpt #6 [Bros on Foot, ep.1]**


anyway travel go-and roll over-must become-CIRCUM *Since we will travel and suffer together anyway*


*Let's give it our best.*

When JW encounters two younger actors, MH and JK, he asks MH how old he is using the title suffix "ssi," which indicates that JW is not using a casual address term, in line 1. When MH responds that he's 32 years old, JW comments that he's not that young. Although JW does not employ an honorific speech style when addressing MH, he at least refrains from being too casual. Then, in lines 6-8, JH suggests to the two younger actors, MH and JK, that they should speak comfortably using *panmal* to each other, in line 9. MH responds with a nod and affirms with a glance at JK, and MH and JK mutually agree to do their best and use casual talk between them.

In the next scene, the four actors arrive at the airport in New Zealand and search for the car that the producers have rented for them in the parking lot. They locate an SUV and proceed to their destination, marking the beginning of their road trip journey in New Zealand. Considering the context, this scene seems to take place on the day following their initial dinner.

#### **Excerpt #7 [Bros on Foot, ep.1]**

1 JW 진구야, 형이 할까? 운전? Cinkwu-ya, hyeng-i ha-lkka wuncen? JK-VOC/INT Hyeng-NM do-shall-Int/INT driving


<sup>11</sup> JW 가 보자.

ka po-ca. Go see-Prop/PLN *Okay, let's get going.*

12 JK 네 ney. Yes/POL *Yes.*


In this excerpt, JW adopts a casual speech style when addressing the two younger actors. He asks JK if he's okay with driving a car with the driver's seat on the right side, which is opposite to the cars in Korea. JW uses JK's first name and the informal vocative suffix *-ya* to address him, employing *panmal* when asking his question in line 1. In line 2, JW also uses the informal personal pronoun *ne* 'you' to refer to JK. In response, JK uses a polite speech style in line 3. Furthermore, when MH asks JW where they should go first, he also uses the polite form, addressing JW as *hyeng-nim*. MH adds the suffix -*nim* to the sibling term *hyeng* to convey his politeness toward JW. In line 5, JW responds using *panmal* when addressing MH.

An interesting point to note is that both JW and JH continue to use *panmal* when addressing the two younger actors, MH and JK, since they had agreed to use a casual speech style during their dinner before their flight to New Zealand. However, MH and JK consistently use the polite speech style when addressing both JW and JH. Furthermore, JK, being the youngest among the four, uses the polite style when interacting with all three other actors, as evident in lines 7, 10, 12, and 15. Conversely, MH, the second youngest in the group, employs the polite form when addressing the two elder actors but switches to *panmal* when interacting with JK, the youngest, as seen in lines 8 and 13.

Although JH suggested that they all use *panmal* when talking to each other during their first dinner, the practical application of this suggestion meant that the older individuals should use *panmal* when speaking to the younger ones, but not the other way around. Similarly, when JH proposed that the two younger actors, MH and JK, communicate with each other using *panmal*, only MH, the older of the two, ended up using *panmal*, while JK continued to use the polite style. Unlike the age-based negotiation between two "friends," in this case, the shift to *panmal* was not intended to be mutual, or at the very least, it was an agreement for the elders to speak informally

with the younger interlocutors. The younger interlocutors don't appear to mind if the elders use *panmal* when addressing them. Perhaps it takes more time for younger individuals to become comfortable using *panmal* when addressing their elders, or they may never fully transition their speech style when interacting with the same people, especially when the age difference exceeds what one might consider a borderline, which can vary from person to person.

**Fig. 2** A summary of the interactional identity and speech style shift in Excerpt #7

As mentioned earlier, not all age-based identity negotiations are successful in achieving the outcome desired by the interlocutors. In the following excerpt, taken from a zombie TV drama, a situation arises where everyone is fleeing from zombies, and some of them are taking refuge on top of their roofs. A detective who had been investigating a crime case in the neighborhood, where the initial zombie outbreak occurred, is seeking shelter on the rooftop of a nearby house. On that rooftop, the detective encounters a YouTuber with an injured leg who had been attempting to film the zombie attacks, and the detective provides first-aid treatment to the injured man.

#### **Excerpt #8 [Our School Now, ep. 8**]



After helping the injured man, the detective opts not to be addressed as "detective" and, instead, requests the man to refer to him as *hyeng*. The man readily accepts this age-based identity negotiation and begins calling him *hyeng-nim*, with the honorific title suffix -*nim*, in line 4. Subsequently, the detective inquires if they can converse casually, which literally means "comfortably," in line 5, and the man also agrees to this. However, as soon as they both agree to these negotiated age-based identities, the detective starts using *panmal* with offensive language and insults the man by calling him stupid for risking his life by coming to the town to film a zombie chase around the neighborhood, as seen in lines 7 through 12. In an immediate response, the man apologizes to the detective shifting back to the title "detective" with the suffix -*nim*, as shown in line 13. The detective offers to be addressed as *hyeng*, once again, but the man politely declines in line 15. In this case, the man wants to reconsider the identity negotiation as the detective becomes too casual and starts insulting him. In line 15, the man's utterance *anipnita, hyengsa-nim* 'No, it's fine, detective' seems to imply that he does not deserve to be accorded the title of "*hyeng-tongsayng*" with the detective. The man readily acknowledges that he engaged in a very foolish act by risking his own life while attempting to shoot the zombies attacking people, and therefore, he does not deserve such an intimate relationship with the detective.

#### *4.3 Negotiated identity #3: Seniority = Different Work Experience*

Excerpts in the previous sections demonstrated how Korean speakers use the term "friend" to refer to someone of the same age and sibling terms to refer to someone of a different age, even if they had no prior personal acquaintance. In this section, interlocutors negotiate their interactional identities beyond their age difference, specifically in terms of their seniority within the workrelated or school-related fields. Seniority in their field of work and/or the year they entered college plays a crucial role in assigning their constructed identities, either as *senpay* 'senior,' or *hwupay* 'junior,' or in some cases, as colleagues or peers. The following excerpts illustrate how participants switch between polite and intimate relationships, depending on their age and/or seniority.

In the following excerpt, several members of a popular Korean K-pop girl group are discussing their backgrounds as guests on today's show. Among them, two of the most senior girls, SN and SJ, were under the spotlight as they were the two most well-known stars. Before this conversation started, the participants were discussing these two individuals who were of the same age. The excerpt begins with a discussion of who greeted whom first when they encountered each other just before today's shooting, starting in line 1. Both SN and SJ responded that they greeted each other simultaneously, implying that they were already aware of each other's age and, therefore, treated each other as friends. Then, MC2 mentions that SJ debuted earlier than SN, but SJ does not seem eager to confirm this, instead just laughing in line 6.

#### **Excerpt #9 [Happy Together – August, 2016]**



12 MC1 왕언니한테 다시 인사 드려 Greet to our BIG-sister (King sister).

13 (all girl-group members stand up and greet bowing at 90 degrees)

14 sub [왕언니께 대하여 경례] [Salutation to the King-big-sister!]

In line 6, MC3 mentions the name of the girl group SJ debuted with, and SJ laughs once more in line 7. SJ confirms this fact with her laughter, though she is not particularly eager to admit it herself. Then, SN asserts that SJ is more senior, and therefore, she should address her using the term *senbay-nim* 'senior-HON,' and she reaffirms this by actually addressing her as *senbay-nim* in lines 8-9. In line 8, SN uses the subject honorific suffix -*si*- and uses the polite speech style to emphasize that SJ is indeed senior to her. Conversely, in line 11, SJ responds with *aniya* 'no' in *panmal* and places her index finger in front of her lips, gesturing for SN to be quiet. This gesture suggests that she may be shy about being called 'senior' by SN, who is much more popular than herself.

In line 12, MC1 instructs all the other girls to stand up and greet SJ as *wang-enni* 'kingbig-sister,' as a way of acknowledging her seniority among all the female guests. Everyone, except the emcees and SJ, stands up and bows at a 90-degree angle to greet SJ. The subtitle in line 14 reads: "Salutation to the King-big-sister!" This term, *wang-enni*, playfully signifies that she holds the highest seniority among all the girls present on the talk show, adding a fun element to their interaction. Everyone laughs and continues their conversation.

As evident from the above excerpt, two friends of the same age are once again subjected to an inquiry to determine who holds seniority, particularly when both belong to the same profession: members of K-Pop girl groups. The moment SN and SJ discovered that they were of the same age, they gained the privilege of using *panmal* with each other without it being considered rude, which is a sign of increased intimacy. However, shortly afterward, they realized that their relationship as "friends" was no longer applicable. It remains unclear whether they will continue using the newly discovered identities of senior and junior. However, during the talk show, they engaged in a brief role-play with these newly assigned identities. Based on other excerpts from talk shows, some individuals continue their relationships with these new identities, while others do not, especially if they do not have the opportunity to meet again after the show. SN did not consistently address SJ with the term *senbay-nim* 'senior-HON,' throughout the show. Nevertheless, it is evident that they will always remember who holds seniority in the entertainment industry.

**Fig. 3** A summary of the interactional identity and speech style shifts in Excerpt #9

The next excerpt is taken from a TV drama, where GW and WJ meet as the last two players in the final round of a boxing championship match. GW emerges victorious, earning the title of champion. On his way home, he encounters WJ. Despite not feeling entirely at ease with each other, they decide to have lunch together and sit at a restaurant table. They both assume they are of the same age and, as a result, use *panmal* with each other from the moment they enter the restaurant. After a few minutes of exchanging questions and answers, GW asks about WJ's age.

#### **Excerpt #10 [Bloodhound ep. 1]**


#### Solidarity through Negotiated Interactional Identities in Korean 115


(omitted)




In line 1, GW asks about WJ's age and learns that WJ is two years older than him. Then, GW promptly apologizes to WJ for not using the polite speech style and addresses him as *hyeng*, in lines 3-4. WJ is visibly upset that GW was using *panmal* with him without first confirming their age difference. From lines 3 to 11, WJ continues to employ *panmal,* while GW shifts to the polite speech style, addressing WJ as *hyeng-nim* with the honorific suffix -*nim*.

However, in line 12, WJ asks GW if he has served in the army. While GW responds that he had mentioned his military service before he began boxing, they discover that they both served in the Marine Corps. WJ is delighted to learn that GW also served in the Marine, leading him to inquire about GW's class in line 19. When GW provides his class number, 1207th, WJ quickly stands up and refers to GW as "sir" in line 21. WJ's exaggerated reaction is common among Korean men who have served in the same branch of the army, as military connections hold special social and emotional significance for them.

From this point onward, WJ not only stands up to salute GW but also begins using the polite speech style. He insists that he will not sit down until GW returns his salute, as seen in line 23. WJ appears to be serious about their army-related relationship, but GW is somewhat hesitant to openly acknowledge their military connection and adjust their interactional identities based on their seniority in the army. Therefore, in line 26, GW suggests that they return to an age-based relationship, expressing his willingness to address WJ as *hyeng* in a casual way. Eventually, in line 29, WJ agrees to revert to a *hyeng-tongsayng* relationship. Finally, they both use *panmal* with each other, as they initially began their conversation, but this time as an older and younger brother instead of as the winner and loser of a boxing match. Different stages of their negotiations of the interactional identities are summarized in the table below.


**Table 4** A summary of the interactional identity and speech style shifts in Excerpt #10

As illustrated in the excerpt and the summary table above, the journey that WJ and GW underwent to finalize their negotiation of interactional identities appears very complex and meticulously calculated, even somewhat excessively stingy. Although this negotiation pattern is derived from a TV drama, it is quite common to encounter this type of negotiation in talk shows and reality programs. Such intricate negotiations are more prevalent among male individuals in their 20s to 40s, including some female athletes who appear on TV programs. However, it is not unusual to witness this kind of dispute-like negotiation in real-life situations as well.

**Fig. 4** A summary of the interactional identity and speech style shifts in Excerpt #10

#### *4.4 Summary*

In summary, the process of negotiating interactional identities among Korean speakers may exhibit variations among interactants, influenced by factors such as age, personality, social identity in relation to others, and more. While the specific details of how this negotiation process unfolds may differ significantly, two core factors consistently impact these processes: power and solidarity. An elevated level of intimacy, achieved through the construction of a new identity in addition to their pre-existing ones (whether work-related, school-related, or stemming from any previously established relationships), appears to be the primary factor initiating the negotiation process. Conversely, a hierarchical structure based on age and/or social status significantly shapes the outcomes when it comes to reshaping the interactional identities of the interlocutors engaged in this negotiation process. The shift in speech style during the process of negotiating their interactional identities, transitioning between the polite and formal style and *panmal* or casual style, reflects an increased level of intimacy. Simultaneously, it underscores the hierarchical ordering between the two interlocutors, primarily based on factors such as age and social status (whether work-related, school-related, or profession-related). Often, this shift in speech style aligns with corresponding changes in their address terms, mirroring the evolving nature of their negotiated interactional relationship.

The negotiation process itself can be quite sensitive and challenging, involving a series of offerings, acceptances, rejections, and renegotiations until both parties arrive at a mutual understanding, marking the conclusion of the negotiation and beginning of a new interactional based on the new identities. However, this intricate process is made more efficient when both parties share the common interactional goal of enhancing solidarity. This goal elevates their relationship beyond mere interlocutors to that of friends, siblings, or members of the same professional or institutional communities.

#### **5 Conclusion**

This study aims to dissect the strategies employed by Korean speakers in the evolution of their newly formed or recently established relationships towards a more intimate and structured framework, with a particular focus on two key determinants: power and solidarity. Furthermore, this research delves into the intricate process by which interlocutors navigate and co-construct their interactional identities, drawing from their pre-existing social identities, such as age and

social seniority. This process is meticulously shaped by the modulation of their speech styles and choice of address terms, all directed toward the achievement of diverse interactional objectives. These objectives encompass the cultivation of increased solidarity through the establishment of novel identities, mutually agreed upon during the course of the speech event.

However, given the intricate nature of Korean speech styles concerning their structure and usage, a more tangible overview of specific phenomena may be attained through the collection and analysis of substantial real-life speech interaction data. Amassing a large-scale dataset for quantitative analysis could prove invaluable in comprehending the nuances and variations inherent in the identity negotiation process, offering valuable insights for future research endeavors.

#### **References**


251–272.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **"What does hyung mean please?": Moments of Teaching and Learning about Korean (Im)politeness on an Online Streaming Platform of Korean TV Drama**

#### **Miriam A. Locher and Thomas C. Messerli<sup>1</sup>**

**Abstract** In this paper we combine an interest in the pragmatics of fiction with interpersonal pragmatics by exploring how Korean (im)politeness norms surface and are negotiated in fictional TV drama. Our data is derived from the streaming platform Viki.com, which allows viewers to comment on the episodes they stream. Building on previous work by Locher (2020), we first report on the pervasive occurrence of scenes containing 'moments of relational work' in Korean TV drama and then explore how viewers comment on this very relational work. While our quantitative results show that viewers do indeed pick up on (im)politeness negotiations (in linguistic and embodied, multimodal form), this finding is relativized by the many other functions that the comments also have. Nevertheless, we are able to show question-answer sequences about relational work and 'moments of teaching and learning' about Korean (im)politeness in this online fan community.

**Keywords** Korean Politeness Korean Address Terms Lay Translations Communal TV Watching and Commenting Fan Subtitles Fan Discussions

<sup>1</sup> M. A. Locher ()

T. C. Messerli

Department of Languages and Literatures, University of Basel, Basel, Switzerland e-mail: Miriam.Locher@unibas.ch

Department of Languages and Literatures, University of Basel, Basel, Switzerland e-mail: Thomas.Messerli@unibas.ch

#### **1 Introduction**

As part of the Korean wave, the viewership of Korean TV drama has been internationalized. This development has been facilitated through the Internet, which has made these cultural artefacts available online on streaming platforms. This chapter works with data from the streaming platform *Viki.com*, which acquires the rights for series and provides the technical affordances for fans to create their own subtitles and to interact with each other in written comments. One of the challenges for people who are interested in Korean culture in general and learners of Korean in particular (Brown 2011: 1) is to grasp the politeness levels that are grammaticalized in adjective and verb inflections, honorific morphemes, specialized vocabulary and in a complex system of address terms as well as embodied actions such as bowing or lowering one's eyes. Translating these combined indexicals is often impossible due to typological non-compatibility (e.g., there are no matching verb/adjective morphemes in English that index honorification), so that translators often aim at translating the general 'tone' of a text (House 2018; Kiaer 2018). In this chapter, we are interested in how the international viewership on *Viki* makes sense of (im)politeness negotiations witnessed in the Korean dramas. Adopting the target audience's perspective rather than the Korean original audiotrack and its Korean dialogue as our starting point, we explore how the English subtitles in combination with the visual input and the nonlinguistic audio information from the Korean original soundtrack provide viewers opportunities of learning about (im)politeness in Korean society.

The chapter builds on our previous work on translating and commenting practices observed in *Viki* (Locher 2020; Locher & Messerli 2020, 2023; Messerli & Locher 2021, 2023, 2024), but sheds particular light on the appropriation and discussion of aspects of Korean (im)politeness. We argue that fan discussions can function as starting points for awareness building of the complexities of the Korean (im)politeness system as well as reflections on viewers' expectations about their own cultural (im)politeness norms. In Section 2, we will first clarify our theoretical approach derived from interpersonal pragmatics and (im)politeness studies and also report on aspects of our previous research. In Section 3, we will explain the datasets for this study and elaborate on our methodological decisions before presenting our results in Section 4. Section 5 concludes our work with an outlook for further research.

#### **2 Literature Background**

#### *2.1 Positioning the Topic and Theoretical Concepts*

Teaching and learning about (im)politeness norms is part and parcel of growing up in a particular society at a particular time. Acquiring knowledge on ideologies and norms about what is considered appropriate behavior in specific contexts goes hand in hand with acquiring knowledge on other norms that shape society, such as the importance of age, seniority, class, gender, race, to name just a few aspects. For Korean society, historian Yuh (2020) reports that an individual's status vis-à-vis another is indeed shaped by a number of combined factors that include, among others, (1) birth, family, marriage, (2) hometown, region, (3) education, (4) profession, (5) and financial wealth. From a linguistics vantage point, age difference, social distance and seniority are added to this complex situation and interact in complex ways (see, e.g., Brown 2011). Since it is important to understand how an addressee is positioned vis-à-vis the

speaker in order to choose appropriate linguistic indexicals (from address terms to verb/adjective endings; see below), interactants often strive to establish salient factors early on in a conversation, such as the conversational partner's age – a practice that is not shared in cultures where age is either considered a private matter or a less important factor for positioning. In Korean society, perhaps more so than in others, these acts of positioning are thus often done explicitly by asking background questions or by challenging, demanding and confirming particular uses of address terms and honorifics.

To illustrate how such practices find their way into televised fiction, consider the following example in the extract from the TV drama *While You Were Sleeping* (2017), in which two young male characters (Han Woo-tak and Jung Jae-chan) meet again by chance after having encountered each other briefly once before during a car accident. After staring each other up and down for several seconds, they start talking to each other with Han Woo-tak clearly wanting to get to know Jung Jae-chan. He grasps his hand (which Jung Jae-chan accepts reluctantly and flinchingly) and asks Jung Jae-chan about his profession (triggered by looking at Jung Jae-chan's name tag) and then invites him for a meal to thank him for the help he received during the car accident. After these turns (not shown here), they start walking next to each other towards the restaurant and start negotiating how they should address each other, and thus how they position each other and shape their relationship.


(1) *While You Were Sleeping*, Ep. 3, 00:54:23, subtitles in Korean and English from Viki (comments in square brackets are not in the original subtitles display)

Informing an interlocutor about one's year of birth is an invitation to start negotiating their relationship. In subtitle 1, Han Woo-tak reveals his age to Jung Jae-chan by sharing his Chinese zodiac (year of the Dragon).<sup>1</sup> Jung Jae-chan matches the information by stating that he too is born in the year of the Dragon. In subtitle 2, Han Woo-tak takes this information up and, on the

<sup>1</sup> The Chinese zodiac constitutes a twelve year cycle and allows interactants to gauge the age of an interlocutor.

basis of sharing the same birth year<sup>2</sup> , calls on the convention that it might be an option to speak to each other in panmal, i.e. without honorifics and informally. While uttering his suggestion he takes Jung Jae-chan's agreement for granted by already dropping the verb endings and thus using panmal. However, Jung Jae-chan outrightly and bluntly rejects this invitation to claim common ground in subtitle 3. This rejection is acknowledged and accepted by Han Woo-tak, as indicated by him slightly hissing and snipping his fingers. In the continuation of this drama, these two characters re-enact similar dialogues, with Jung Jae-chan resisting Han Woo-tak's attempts to change their positioning. Scenes like in extract (1) are abundant in Korean drama (Locher 2020) and this data is an example of findings reported in research that argues for the dynamic negotiation of politeness levels in Korean (rather than a 'by rote' application and computation of factors such as similar age) (e.g., Brown 2011, this volume).

This negotiability of appropriate politeness levels is particularly complex in the Korean language since there are many different indexicals available (Brown 2011, 2013, 2015, 2022; Brown & Winter 2019; Brown, Winter, Idemaru & Grawunder 2014; Choo 2006; Kiaer 2018; Kim 2015; King 2006; Koh 2006; Rhee 2019). There is particular vocabulary used when addressing people of higher status (e.g. using 댁 *tayk* instead of 집 *cip* for 'house'), there is a plethora of particular address terms used to indicate relationships and hierarchical status (e.g. 선배 *senpay* / 후배 *hwupay*) to indicate seniority in a relationship), there are morphemes added to verb and adjective endings that index different politeness levels and honorific morphemes that can be combined with them. The verb/adjective endings are particularly important as their choice is mandatory and therefore any sentence contains positioning information (Brown 2011: 1; Rhee & Koo 2017: 101). The linguistic positioning options are thus plentiful and are also combined with multimodal indexicals, such as bowing, facial expressions and (the lowering of) eye gaze (Brown & Winter 2019). The choices made have consequences for positioning oneself vis-à-vis the other and thus for identity construction in every utterance.

Research on identity construction that highlights the dynamics of acts of positioning (see, e.g., Bucholtz & Hall 2005; Davies & Harré 1990; Locher 2008) expects the possibility that many different ideologies surface in interaction, and that these are drawn on, exploited, reinforced but also potentially changed. In the scene from extract (1), ideologies about age and profession but also closeness and distance surfaced. Moments of (im)politeness, as expressed through linguistic and multi-modal means, are where such ideologies often converge. It has by now a long tradition to combine research on identity construction with the negotiation of face concerns and (im)politeness studies (Garcés-Conejos Blitvich 2013; Locher 2008). For this study, we adopt the perspective of interpersonal pragmatics (Locher & Graham 2010), i.e. focusing on the relational aspect of language and communication (while acknowledging other dimensions such as the informational at the same time). We also work with the concept of 'relational work', which describes all the linguistic and multimodal work that people invest to shape, challenge, contest and confirm relationships in interaction (Locher & Watts 2005, 2008). This term is a technical term that allows us to describe face-enhancing, face-maintaining and face-challenging behavior from an etic perspective and to explore comments and behaviors by interactants from an emic perspective. Adopting the term 'relational work' within the field of (im)politeness studies highlights that (im)politeness too is dynamic, despite the fact that it draws on existing societal ideologies and is thus always embedded in its socio-historical context. This

<sup>2</sup> In the dialogue as well as in the subtitles, the expression 동갑, which means being of the same age, has been abbreviated to 갑 (kap).

is because ideologies, rather than being static and unchanging, are always tied to a particular context.

For the purpose of this project, Locher (2020) termed negotiations such as illustrated in extract (1) "moments of relational work" in fiction and it was found that these are of particular importance in Korean TV dramas. It is not surprising that these Korean fictional artefacts contain many scenes where interactants explicitly negotiate relationship by means of referencing (im)politeness norms. On the one hand, this is because in non-fictional contexts too the positioning of each other is an important factor for relationship creation and, on the other hand, fiction uses character development to drive plots (Locher 2020). In her study of four TV dramas (68 episodes), Locher (2020: 149) found that, on average, there were 3.2 scenes per 60 minutes episode that contained explicit moments of relational work in the form of characters either using lexemes from the semantic field of (im)politeness (in Eelen's terms this corresponds to 'classificatory politeness1') or engaging in meta-discussions about (im)politeness, such as in extract (1) (in Eelen's terms 'metapragmatic politeness1') (Eelen 2001: 35). TV dramas thus provide ample staged data to observe negotiations of politeness norms.

Acknowledging the fictionality of the data is important because the staged nature of the conversations means that (im)politeness ideologies are explicitly and consciously included for effect and convey and play with cultural ideologies. This means that (im)politeness ideologies can be rendered in more or less faithful ways (in comparison to face-to-face non-fictional interaction) but their inclusion always serves the needs of the drama. Studying these effects is of interest in and of itself (see, among many, Alvarez-Pereyre 2011; Jucker & Locher 2017; Locher & Jucker 2021). The target audience is the Korean viewership, who will be able to pick up on nuances of relational work not just in the explicit scenes of moments of relational work but throughout every utterance included in the artefact. The viewership who has no direct access to the Korean language, i.e. the target audience of the subtitles, will be able to pick up on relational work in particular in those scenes containing moments of relational work as defined above. As a consequence, translators face the challenge of giving access to the dialogue that negotiates relational work. Locher's (2020) research on fan subtitles in Viki shows that, while they cannot give access to the full gamut of nuances of the original relational work in Korean, subtitles contain pointers for the non-Korean audience to pick up on relationship negotiations. For example, subtitlers often retain the Korean address term system through Korean borrowings instead of domesticating by using first or last names. In this way, positionings of characters visà-vis each other by role are retained. Sometimes these address terms are used without translation, sometimes the subtitlers add a translation in brackets. The subtitles also add comments in brackets about changing politeness levels in the verb/adjective endings or vocatives and thus point to changes in and negotiations of relationships through language. Examples can be seen in the subtitles (2) and (3):


In a previous edited collection, Pizziconi and Locher (2015) have brought together scholars who work both on first and second language acquisition and throughout the many possible spaces of teaching and learning were highlighted. In the 2015 collection and in this chapter, we conceive of teaching not only in the formal classroom setting but also in everyday life contexts outside of the second language acquisition classroom. The same goes for learning that can take

place in manifold ways, and knowledge about others can be brokered through many parties. Here we are concerned with the online fan community of Korean TV dramas, which is exposed to scenes of moments of relational work and their translation. Brown (2011: 1) reports that learning about the Korean honorific system is one of the most challenging parts of learning Korean for second language learners, and we would argue that the drama scenes and their translations give a glimpse of this complexity. In this chapter, we are thus not concerned with classroom language acquisition, nor will we explore effectiveness of language learning. Instead, we explore how viewers of Korean TV drama talk about moments within the artefact that contain relational work negotiations or aspects about relational work that the viewership picks up on. In addition to using the term "moments of relational work" to point to scenes within the fictional artefacts, we would thus like to posit that these staged interactions can and often do present "moments of learning and teaching" about Korean (im)politeness ideologies. In order to position this topic further, we next turn to previous work on the international followership of Korean TV drama.

#### *2.2 K-Wave and Interest in Korean Artefacts, Culture and Language*

Interest in Korean culture and cultural artefacts such as music, webtoon (Korean online manhwa/comic), TV drama, or movies has increased over the last 20 years, a phenomenon that has been referred to as the Korean wave or *Hallyu* (Hong 2014; Kiaer & Kim 2021; J. Kim 2014b; Y. Kim 2013; Lee & Nornes 2015). As scholars who work in the field of English linguistics, our point of entry is the large international fan community that the Korean wave has generated. The international Korean TV drama fandom often engages with the cultural artefacts through English as a lingua franca and via computer-mediated means. There are many ways in which this can take place. Fans of K-pop groups are active in translating lyrics and following the group members' activities in media appearances and concerts, share clips with translations on general social media, such as youtube, tiktok and Instagram, and in fandom dedicated websites and fan associations, etc. fans can access Korean TV drama series online in manifold ways. Next to illegal access through streaming websites and peer-to-peer sharing such as torrents, there are websites that are dedicated to giving legal access to Korean audiovisual artefacts to the growing fandom, such as the now defunct DramaFever (2009-2018) or the still active Viki (since 2007), which focuses in particular on Asian series. Big market streaming platforms such as Netflix or AppleTV have recently also started to heavily invest in Korean TV drama series and movies and have expanded access to artefacts from this cultural context through professional translations (see Messer & Locher 2023). In other words, the reception of these artefacts has expanded to a global viewership.

Locher (2020) showed that fan subtitles on Viki are oriented towards a target audience whose members are assumed to be interested in the source culture (plot, sights, cuisine, language, etc.). In this sense, the translations are not prioritizing ease of comprehension (via domestication) by aiming at aesthetic target texts or adapting the language to comparative cultural experiences in the target language. Instead, Locher has shown that many address terms are taken over, with many borrowings other than the previously mentioned Korean address terms making it into the translations. This presumes a certain acquaintance of the target audience with the Korean language. Moreover, the subtitles also employ comments on culture in brackets about diverse aspects such as cultural practices, cuisine, idioms, or currency equivalence. In this chapter we want to explore further how relational work in the fictional scenes is taken up and commented on by the viewers (see Section 3.1 on the commenting possibilities).

K-drama fans do not only watch dramas, but many also actively engage in commenting on dramas and actors. The diverse and heterogeneous fandom meets in sundry online communities in order to discuss their favorite dramas, critically review them, share favorite scenes and plot musing or talk about favorite actors. For example, a long-standing platform for such fans is www.dramabeans.com, which has been recapping and critically discussing series since 2007 (see Schultze 2013, 2016). In the case of Viki, interactive possibilities are many, from subtitling, review writing, creating a multimodal profile to commenting in so-called 'timed comments' – written comments tied to particular moments of streamed episodes (see also Dwyer 2012, 2017, 2019; Kiaer & Kim 2021). The longer an episode has been available on the streaming site, the more comments are accumulated. In Section 3.1, we explain in more detail how this works, since this is the main data for our empirical analysis in this chapter.

#### *2.3 Research Questions*

From the brief literature review above, it has become clear that (1) K-drama is rich in scenes transporting cultural ideologies about relational work in Korean and (2) that the K-drama fandom is active in many ways and willing to engage with and learn about Korean culture. In what follows, we want to explore how the particular group of viewers who use the streaming platform Viki comment on relational work and language. (See also Kiaer et al., in this collection, who explore how drama fans pick up on embodied (im)politeness issues through questionnaire and interview data). With an interest in moments of teaching and learning in the Viki community in mind, we ask the following general questions:


#### **3 Data and Method**

#### *3.1 Viki's Fan Translations and Timed Comments: Previous Results*

In the previous sections, we have already introduced Viki as a streaming platform that gives access to licensed Asian TV dramas. The platform provides the technical affordances for teams of fans to add subtitles<sup>3</sup> . English is usually the first language into which Korean is translated and from English, subtitles in many more languages are added depending on the availability of fan translators. Locher (2020) established that the subtitlers orient towards the source language and function as cultural mediators who design texts for an audience that they assume to have an interest in Korean culture. This is evidenced in comments in brackets and in the inclusion of Korean address terms (and further borrowings) as in examples (1) to (2) above.

In addition, Viki allows fans to get active on the platform by rating episodes and dramas, and writing reviews. This study is in particular concerned with *timed comments* as another option

<sup>3</sup> The fan translators are credited by a team name and individual user names within the artefact. However, no further information about them exist. Since Viki is not available in South Korea, translators are likely to be Korean native speakers who live abroad or advanced learners.

to engage with the cultural artefact. When viewers activate the timed comments option, they see a list of comments displayed either below the video window or to the right of it. Another option users can choose is to see only select comments directly within the video frame at the top (like a surtitle). The comments are made by viewers who stop the video and post a remark. Viewers watching and commenting simultaneously cannot directly interact with each other, since new comments are only loaded when the stream is started, but subsequent viewers will see all previous comments as if they were written synchronously by other viewers. The effect of the comments is thus the illusion of co-watching with others. This illusion is created because when you watch, you get access to a stream of other viewers' voices, and their input influences your own uptake of the artefact. The co-presence is an illusion in so far that the timed comments can in fact be written at very different moments in time. Rather than being time-stamped and hierarchically ordered like in a blog or forum thread, the comments are directly linked to the minute and second within the video and show up whenever a new viewer arrives at this moment within the video. Over time, many voices are thus added and the viewing experience changes every time (Locher & Messerli 2020).

In Locher and Messerli (2020) and Messerli and Locher (2021), we conducted case studies of two episodes from two different dramas that contained 5,919 comments in total. Since then, we expanded this data by including the last two episodes of the same dramas to check the representativeness of the codebook, resulting in 8,930 analysed timed comments. This expansion confirmed the representativeness of the four episodes for TC functions in general. The aim was to establish what viewers actually 'do' in the timed comments. The content of each comment was thus qualitatively coded for its communicative function.

**Table 1.** Functional codining in four episodes of two K-dramas (*Meloholic* and *You Are All Surrounded*)


\*The numbers document that the comments contained at least one code from this group.

The results showed that viewers commented on artefact-oriented issues (comments such as on plot, intertextuality, characters, diegetic technique, etc.) in 71% of all comments (Table 1) as well as more community-oriented issues (comments such as sharing information on the time and place of watching or the nationality of the viewer, and asking/answering questions) in 34% of all comments. Apart from predominantly artefact- or community-oriented functions, we identified two categories for which no clear distinction between artefact-oriented and community-oriented functions could be made (i.e. a comment either performed both functions at the same time or the contributions could not be clearly attributed to one of these functions only). These were comments on culture (more on this category in Section 4.1) and especially comments revealing

emotional stance, which include emojis, emoticons or laugh particles (72% of all comments). A timed comment often performed more than one function at the same time (69% of all comments contained more than one category). Overall, the three most important functions are comments on 'sharing emotive stance' (n=6,299, 71% of 8,930 comments), 'plot' (n=4,086, 46% of 8,930 comments) and 'interaction with commenter' (n=1,760, 20% of 8,930 comments; not shown in Table 1).

To give the reader an understanding of what these comments look like, consider (4), in which a commenter shares their emotive stance (OMG and capitalization) of having figured out a plot element about the drama; (5) in which a commenter supports a previous plot comment made in the timed comments, so that interactivity between commenters becomes apparent; (6) in which the writer comments on character behavior within the plot and reveals emotive stance (capitalization) as well as where they are from; and (7) where a viewer shares the place of watching, thus implying that the drama is addictive.


In both the subtitles and the comments, the fan translators and viewers engage with Korean culture and also reflect on their own culture. We thus argued that the translators function as cultural mediators in their orientation towards the source culture and that the fan comments display active engagement with and interest in Korean culture. This chapter explores this finding further and focuses on discussions of (im)politeness in moments of teaching and learning within the timed comments.

#### *3.2 Data Description*

Our data compilation consists of two large corpora as well as of specific samples of these corpora we created for the current study. On the one hand, we work with the fan-translated subtitles of a convenience sample of seven dramas belonging to the genres romance, comedy and action (see Table 2), entitled "Moments of Relational Work in Fiction" (MoRWF) corpus. In comparison to the Locher-2020 data set, this corpus has been expanded from four to seven dramas (from 68 to 110 episodes, that add up to 613,000 words in 78,568 subtitles).

This corpus has been used to identify scenes that contain moments of relational work (i.e. scenes that use classificatory politeness1 comments and/or metapragmatic politeness1 comments by the characters and meta-comments by the subtitlers, as explained in section 2.1 and illustrated with extracts 1 to 3). Table 3 gives an overview of the scenes containing such relational work. Overall, there are 323 scenes containing moments of relational work, which means that there are

2.9 scenes per episode on average.<sup>4</sup> Scenes could contain more than one relational work moment so that that the total of these moments is higher with 428 occurrences.

Table 2. Number of subtitles and words in the MoRWF-Corpus (adapted and expanded from Locher 2020: 145)


Table 3. Number of scenes and types of relational work moments in the MoRWF-Corpus (expanded from Locher 2020: 149)


Legend: MH (Meloholic), OMHE (One More Happy Ending), TA (Twenty Again), W (W), YAAS (You're All Surrounded), RM (relational work moment)

To analyze viewer comments, we work with the K-drama Time Aligned Comment Corpus (K-TACC), which contains the timed comments of five of these dramas, amounting to 320,118 timed comments (2.9 million words), added to 80 episodes (Table 4) (Messerli & Locher 2021) and build on the results of the case studies reported on in Section 3.1 (Locher & Messerli 2020), which established the functions of the timed comments in four episodes from two dramas.

In order to be able to compare what viewers do during scenes containing relational work moments and compare their practices to those outside of those scenes, we created the K-TACC-RW subcorpus as a corpus of interest and the K-TACC-Non-RW subcorpus as a reference corpus. Starting point for the creation of these subcorpora was the manual annotation of the MoRWF corpus and the identification of relational work scenes. In R, timespans tied to specific Korean TV-drama episodes were extracted from the manual annotations and the K-TACC corpus was then automatically tagged so that comments written during such scenes were labelled as belonging to a particular RW scene. Using base R (4.2.0, R Core Team 2022) as well as

<sup>4</sup> This average decreased from 3.2 scenes in the smaller corpus of four dramas reported on in Section 2.1 (Locher 2020: 149) to 2.9 scenes per episode in the larger corpus that contains three more dramas (seven in total).

quanteda (3.2.1, Benoit et al. 2018), the K-TACC-RW subcorpus was then compiled based on only those comments written during any RW Scene, whereas the K-TACC-non-RW corpus contains all other comments. In order to facilitate corpus-driven methods such as keywords, comments identified as non-English (see Messerli & Locher 2021) were removed from both subcorpora. Table 5 provides an overview of the data that each of the K-TACC sub-corpora contains. Table 6 shows how many timed comments were added to relational work scenes for each of the dramas in K-TACC-RW.


Table 4: Overview of K-drama Time Aligned Comment Corpus (K-TACC) (Messerli & Locher 2021: 414)

#### Table 5: Overview of K-TACC-RW and K-TACC-non-RW subcorpora


\* We confirmed that comments posted during relational work scenes are significantly longer than other comments by means of a Mann-Whitney U test (W = 2234933380, p-value < 2.2e-16) in R (R Core Team 2022).



Legend: MH (Meloholic), OMHE (One More Happy Ending), TA (Twenty Again), W (W), YAAS (You're All Surrounded), RM (relational work moment)

The subtitles in Viki are open source as they are user generated fan translations. The timed comments are visible to anyone viewing the videos with and without subscription and can thus be considered to be publicly accessible. We removed all user name indications in order to maintain anonymity of commenters. Subtitles and timed comments are presented as they were written. Korean subtitles are only indicated when a relational work issue is in need of further clarification. In other instances, only the English subtitles are shown because this is what the viewers refer to in their discussions.

#### *3.3 Methodology*

We approached our first question on whether viewers comment on relational work in the scenes identified as containing relational work moments by creating the K-TACC-RW (relational work) subcorpus, which contains comments written by viewers during scenes we identified as containing relational work based on the subtitles in the MoRWF Corpus (see Section 3.2). For manual tagging, we created a relational database by combining data from MoRWF and K-TACC-RW in Filemaker. This database allowed us to make visible relational work scenes as sequences of both English subtitles and timed comments that appear during the scene. On this basis, we manually checked the timed comments for evidence that the viewers talk about the staged relational work in the scenes. This work was done by two coders. 5

For the subsequent corpus-assisted steps, we worked with the K-TACC, K-TACC-RW and K-TACC-non-RW (sub-)corpora (see Section 3.2) in R 4.2.0 (R Core Team 2022) and using quanteda 3.2.1 (Benoit et al. 2018). We first compared surface statistics (see Table 5 in Section 3.2) and did further text statistics ("textstat\_frequency()" in quanteda.textstats). As a next step, we performed a keyness analysis using log likelihood, with K-TACC-RW as a corpus of interest and K-TACC-non-RW as a reference corpus, in order to get an overview of the aboutness of those comments that were written during relational work scenes in particular.

Potentially, the keywords in K-TACC-RW could have already pointed us to particular teaching and learning moments and thus to answers regarding our second research question on the evidence of learning and teaching about (im)politeness in timed comments. We will briefly report on these finding in Section 4.2, but in summary, the keywords in K-TACC-RW alone did not seem to form a sufficient basis for a corpus-assisted discourse analysis of the comments' teaching and learning potential. Instead, we manually extracted those keywords we deemed relevant and complemented them with searches for questions and search terms based on Framenet (Ruppenhofer et al. 2016). The joint list was used to arrive at keywords in context (kwic) lists which we could then analyze in detail by means of a close-reading.

In particular, we categorized terms from several frames in Framenet into the three categories illustrated in Table 7. *Explanations* was created by collecting relevant terms from three frames in Framenet (*Expertise, Awareness* and *Explaining\_the\_facts*), whereas as *Negative* and *Positive evaluation of social interaction* were created by manually sorting the terms in *Social\_interaction\_evaluation* on Framenet based on the polarity of the evaluation. The rationale for the selection of frames consisted in looking for frames in Framenet based on lexical items we had identified as potentially indicative of teaching and learning moments, while not being so broad as to create much noise, i.e. false positives when used to search our corpora. We focused on moments where fans may explain aspects of culture to other fans on the one hand, and those where fans may evaluate the social interactions they see on the other. We then complemented

<sup>5</sup> Thanks go to Christan Feige, who worked with us as a research student intern for a semester and co-coded the timed comments posted during relational work scenes with the first author.

these lists with selected terms we deemed to be missing in our Framenet-based lists, or which we thought could be indexical for such teaching and learning moments within the particular communicative setting we investigate. This led to another list of potentially relevant terms: *accent\*, age\*, argu(e)\*, dialect\*, old\*, year\*, culture, formal, informal, sato(o)ri, relationship, treat, young* (terms marked with \* are derived from keywords).

Table 7. Categories of search terms based on Framenet and perusal of timed comments


Legend: Terms marked with \* are keywords we found in K-TACC-RW and added to the lists from Framenet.

For all five lists – the question words, the three Framenet-derived and the list of customized terms, complemented with the terms we extracted from the keywords – we created kwic-lists in quanteda. We used regular expressions to include e.g. comparative forms of adjectives and verb forms and included automatic clean-up of some false positives (e.g. constructions where *mean* is used as a verb rather than an adjective) before qualitative analysis. It is important to note that we performed these quantitative steps in the tradition of corpus-based discourse analysis, with the specific aim of creating a methodologically sound and systematic basis for the qualitative analysis of examples.

Finally, we go beyond linguistic aspects of relational work and complement our corpusassisted methods with a purely qualitative analysis of multimodal aspects of interpersonal pragmatics. This also allows us to look at a larger context for select examples and to move beyond the utterance level of individual comments to interactivity and interactual context.

#### **4 Analysis**

In what follows we will first present our qualitative results by exploring the link between the subtitled scenes and the timed comments with discussion of relational work in mind (Section 4.1). We then turn to quantitative, corpus-assisted methods that are more systematic but also

more limited in their scope (Sections 4.2). We end with a qualitative section that presents insights on interactive negotiations of relational work as expressed through multimodality (4.3).

#### *4.1 Qualitative Analysis: Evidence of Awareness of Relational Work in the Timed Comments*

In Locher and Messerli (2020: 31), which presented the exhaustive coding of 5,919 timed comments in four drama episodes from two dramas, we already established that relational work is being commented on, albeit not in large numbers from a quantitative perspective. From the category "culture" of which 'relational work' was coded as a sub-type, only 11 out of 289 instances referred clearly to relational work. The other comments evoked the source culture by employing Koran borrowings of address terms and other lexemes or by comparing Korean culture with the viewers' own culture. With 15 relational work moment scenes in total within the four episodes, we would have expected to see more uptake in the timed comments.

From the perspective of viewer experience, we felt that this result needed further scrutiny since there is indeed evidence that viewers pick up on relational work and comment on it. Example (1) in Section 2.1, in which the character Han Woo-tak suggests reciprocal use of panmal, but is rejected by Jung Jae-chan, did indeed receive comments on relational work. For ease of reference, the English dialogue in the subtitles is rendered here again, followed by a selection of timed comments.

(8) *While You Were Sleeping*, Ep. 3, 00:54:53, English subtitles in English from Viki (comments in square brackets are not in the original subtitles display)


For example, in (9) to (12) viewers comment on the fact that a Chinese zodiac sign is used to establish age and help each other in establishing the reference of what 'year of the dragon' refers to. In doing so, they engage in discussing age as one of the parameters that determines Korean politeness levels.


Viewers also pick up on the fact that Han Woo-tak is rejected in his bid for a closer relationship:


#### (15) Bromance ship has sailed

As a consequence, those viewers who watch the scene with comments activated obtain additional information concerning relational work that goes beyond the original artefact. The meaning conveyed in the timed comments thus adds additional voices to the relationship construction that is staged within the video.

In Locher and Messerli (2020: 21-22), we also discuss such a scene where the subtitles mention changing levels of politeness. They do this in a simplified manner by referring to changes in politeness levels as changes in 'formality' (see also (13) above). The two main characters face each other at the kitchen table. They playfully negotiate how to address each other again since, due to the plot which involves alternative worlds where time flows differently, one of the characters has aged more quickly than the other. The dialogue in the subtitles thus evokes (im)politeness ideologies around the factor age and expresses this by oscillating between formal and informal verb/adjective endings as well as mentioning address terms ('call me Oppa', line 1).

(16) *W*, episode 15, minutes 36 to 37, comments in brackets and italics are part of the original subtitles with the exception of [sic]


The timed comments posed during this scene (all by different viewers) pick up on and ask about the change in politeness levels:


In contributions (17) to (19), we see two comments about relational work with question marks that invite responses and also a statement of not understanding relational work aspects, which is an indirect invitation to explain. In (20), we see a response and an explanation followed by a comment on the challenge of using appropriate relational work in Korea. In (21), we find a comment that evokes a teaching frame by writing "Imma lay down some educational stuff", thus clearly marking that what follows is an opportunity for learning. What we see, therefore, are moments of teaching and learning triggered by the artefact and made possible through interaction in the timed comments. Viewers ask explicit and implicit questions about (im)politeness ideologies and receive responses from co-viewers. While these responses might not be entirely accurate or complete (for example, the reduction of politeness concerns to formality), they do add to a further understanding of the complexity and importance of (in this case linguistic) relational work in Korean and contribute to explaining why such scenes are included in the artefact in the first place.

The timed comments on (8) and (16) show that drama scenes can indeed function as triggers for discussions involving Korean relational work. In both cases the scenes and comments were about morpho-grammatical negotiations of politeness levels. Further evidence of uptake of relational work indexicals can also be found in questions and answers about address terms. As outlined in Section 2.1, one possibility for positioning in Korean is the use of diverse address terms derived from different semantic fields such as the family, workplace or seniority within different contexts (Koh 2006). These terms are challenging to translate and a handful of the better known ones are often retained in the fan translations. In Locher (2020), it was also established that the negotiation of address terms was a frequent trope in the scenes containing relational work moments. Looking at the comments of four episodes in our case study, our results in Locher and Messerli (2020: 31) show that the category "culture", which contained borrowing, was used in 289 times in the small case study, of which 100 were about the use of address terms. If we look at the K-TACC in its entirety, we can first of all mention that – in comparison to the plethora of Korean address terms available – only a small number of address terms are among the Korean borrowings that the fans frequently use (e.g., comments containing appa (n=261), omma (52), oppa (n=2087), hyeong (n=249), nuna (n=118), unni (n=211), ahjumma (n=108), sunbae (n=29), hoobae (n=11) – all in various Romanized spellings). Within the Korean borrowing practice of this fan group, the group of address terms is similarly important as the use of emotive interjections (e.g., comments containing omo (n=1946), aigoo (n=359), daebak (n=248), aish (n=336) and heol (n=59) in different spellings).

The address term borrowings are predominantly used to simply refer to characters as in (22) and (23 and/or to address them directly in connection with the plot as in (24) and (25).


However, there is also evidence that there is active teaching and (assumedly) learning in that viewers ask questions and obtain answers about address terms in the timed comments. Below is a selection of such exchanges (26–28). Intermittent non relevant timed comments have been removed to only display those that pertain to the interchanges about address terms.


User 4 @lc it's a way of younger boys addressing older men\n


User 10 Hoobae is junior

User 11 Hoobae mean same school but later then him is hoobae

Not every question asked received an answer and not all the responses are always entirely comprehensive or accurate in their explanation, but Examples (26) to (28) show that the fan community is willing to learn and teach and to interact with each other with respect to these linguistic markers of relational work. From our qualitative approach we can thus confirm that viewer engagement with Korean relation work does occur. In order to explore this insight more quantitatively, we turn to corpus linguistics methods in Section 4.2.

#### *4.2 Quantitative Analysis: Corpus-Assisted Exploration of Timed Comments*

In order to explore further whether there is uptake of staged relational work in timed comments, we created the K-TACC-RW subcorpus, which contains all comments written during relational work scenes, both as a version for manual coding in Filemaker and for quantitative and qualitative corpus analysis in R (see Sections 3.2 and 3.3). Our initial aim was to establish whether the scenes that the trained coders had identified as containing relational work negotiations indeed trigger discussion in the timed comments and whether this is a rare or frequent phenomenon. We started from 238 relevant scenes, of which 8 received no comments whatsoever. Even with only 230 scenes left, there are almost 20,000 comments in K-TACC-RW (see the last row in Table 6, Section 3.2), and detailed qualitative coding as we had done in Locher and Messerli (2020) was thus not feasible for time-reasons. Rather than manually coding each timed comment, we thus only skimmed the material and coded for presence and absence of timed comments categories, among them relational work comments. From this rough analysis, it transpires that about half of the scenes contained some form of relational work comment, while extensive meta-discussions like in the responses to the scenes in (8) and (16) appear to be rare. However, this should not diminish the fact that the viewers can learn about (im)politeness ideologies merely by being exposed to the scenes and by reading the comments that do exist.

We now turn to corpus-assisted analyses to help us find out whether relational work is taken up and whether potential moments of teaching and learning occur. We first compared the comments inside and outside of relational work scenes with a keyword analysis (Section 4.2.1), then we turn to questions and explanations as a potential location where relational work might be discussed (Section 4.2.2) and discuss character evaluations from the semantic field of relational work (Section 4.2.3), before turning to borrowings (Section 4.2.4).

#### 4.2.1 Comments Inside and Outside of Relational Work Scenes

We first wondered whether the lexicon employed during relational work scenes somehow differs systematically from those made outside such scenes. The first finding concerns the average length of comments during scenes that we categorized as being concerned with relational work. With a median of 7 and an average of 8.39 words in length, comments during RW-relevant scenes (henceforth RW-comments) are significantly longer than other comments (see Table 5 in Section 3.2). This indicates quite simply that viewers find more to write during RW scenes than during other scenes.

In order to compare what these comments during relational work scenes are about, we looked at the most significant 100 keywords in K-TACC-RW (based on log likelihood) and manually established those topoi that are of relevance to our interest in relational work and teaching moments (Table 8 below). Most of the terms are specific to the plot of particular scenes, e.g. "Conan" for a scene in episode 10 of *One More Happy Ending* in which the US entertainer Conan O'Brien appears as a cameo, whereas another group focuses on the cinematic realization of Korean TV-drama, e.g. "camera", "cameraman", "shaking", "shaky" during a scene in which viewers complain about camera shaking (see also Locher & Messerli 2020). For these scenes, keywords simply tell us what comment-worthy plot elements occur, without any obvious direct connection to relational work. A third group points to terms of *Korean wave* fandom (with "army" and "bts" both referring to the k-pop group BTS and their fans). More interesting to the question at hand is the category *Emotive stance* which shows that commenters during RW-scenes are particularly expressive and keen to share their emotional stance.<sup>6</sup> This group consists mainly of emojis. Finally, the most clearly relevant group, *Relational work relevant* shows that viewers evaluate character behavior as "creepy", "harsh", "inconsiderate", "rude" or "ungrateful". They further seem to pick up on age differences ("older", "age", "years") and also on linguistic behavior ("dialect", "accent", "arguing") that may be of significance also for the learning of culturally bound interaction within K-dramas and in Korean culture more generally.


Table 8. Select top 100 keywords in K-TACC-RW by category, ordered thematically

It is evident that the quantitative comparison of subcorpora by means of keywords does not in itself provide in-depth insights into the relational work aspects viewers address in comments. However, the fact that as an aboutness-measure these keywords point beyond aspects of plot and

<sup>6</sup> We should stress that emotive stance was the most used function expressed in the case study of all comments, to the point that 71 per cent of all timed comments have some type of emotive stance (Locher & Messerli 2020: 414). What we find here then is that even compared to a reference corpus in which emotive stance is already very frequent, K-TACC-RW contains several emotive keywords.

include expressive as well as relational elements gives us some further confirmation that interpersonal aspects of fictional Korean interaction are also taken up by commenting fans. Whereas the emotive stance words indicate that viewers may exhibit a stronger emotional reaction to these scenes, the pertinent relational work terms show an explicit interest particularly in the negative evaluation of social behavior and in the recognition of age as a relevant factor in the negotiation of politeness norms.

#### 4.2.2 Questions and Explanations in K-TACC-RW

Having established the most relevant keywords and grouped them into relevant topoi, we expanded our exploration of commenters' relational work related practices by first looking for questions and then by using Framenet as a semantic dictionary (see Section 3.3). In order to find moments where fans may explain cultural and perhaps relational work aspects to each other, we will first consult the list of concordances or keywords in context (kwic) we arrived at by searching the corpus for the terms in the categories *Questions* and *Explanations* (Table 7 in Section 3.3).

In order to operationalize the search for questions in K-TACC-RW, we chose to limit ourselves to surface features consisting in the question mark and English question words. The resulting findings in Table 9 indicate first of all that question marks are not something that occurs more frequently during scenes containing relational work and that questions are unsurprisingly a very non-specific way of looking for specific interactions. It was not possible for this study to explore questions in detail, but already a cursory glance revealed that they are used for a plethora of different purposes that perhaps deserve a separate study. The same was true for *how*, which turned out not to lead us to any relevant comments.


Table 9: Questions in K-TACC-RW and K-TACC-non-RW, ordered alphabetically

**\*** Terms marked with an asterisk appeared at least 5 times per 100,000 words and are more frequently in K-TACC-RW

The next step was then to search for the terms related to explanation we extracted from Framenet<sup>7</sup> , thus aiming at questions and answers in which fans request or share others' and their own expertise. We found a total of 1,014 occurrences in K-TACC-RW (Table 10). In this and

<sup>7</sup> Explanation keywords were: account, explain, explanation, expert, expertise, familiar, knowledgeable, aware, believe, know, reckon, suspect, think, understand.

subsequent Tables 10 to 12, we collapsed terms that do not appear in either subcorpus into one row and marked by asterisk term that occur at least 5 times per 100,000 words and are more frequent in K-TACC-RW.


Table 10: Explanation concordances in K-TACC-RW and K-TACC-non-RW, ordered alphabetically

**\*** Terms marked with an asterisk appeared at least 5 times per 100,000 words and are more frequently in K-TACC-RW

While the terms *explain*, *believe*, *think* and *understand* may be worth exploring in the K-TACC corpus, our focus on K-TACC-RW here means that we will only look at *know* and *suspect* in this section. Looking at the concordances for *suspect*, it turns out that most instances (12 out of 15) refer to the noun 'suspect', and we can thus discard it as a useful indicator of explanation sequences. Looking in more detail at the lemma *know,* the syntactic word *knows* appears as the main difference between the subcorpora. Some of the examples of *knows* do indeed point to requests for information. However, they do not point to relational work in particular, and more generally, viewers appear to use *knows* to engage in mind-reading, i.e. in expressing their interpretation of a character's knowledge.

Here and in the other analyses we document in this section, we focused on results that were more frequent in K-TACC-RW than in other scenes. Based on the particular explicitness of *explain* as a lexical item for the finding of explanations in the corpus, we decided to also include the 18 comments containing *explain.* However, out of the 18 cases we thus included, only one pointed us to an actual request for an explanation (29). This comment does not make specific reference to what it is the viewer is confused about, nor do any surrounding comments give any pointers, and the request for explanation does not receive any response in the comments. Our best guess is that the comment voices confusion about the plot rather than the positioning of characters.

(29) SOMEONE EXPLAIN IM CONFUSED (W, Ep. 1)

Insofar as explanation-words serve as indicators of explanation practices, we can thus say that the relational work scenes do not seem to trigger more explanatory comments than other scenes in the Korean TV-dramas in our corpus.

#### 4.2.3 Evaluations of Social Relationships in K-TACC-RW

Another frame from Framenet that seems a promising indicator of discussions of relational work are terms about the evaluation of social interactions. In particular, we saw already in our list of keywords (Section 4.2) that some negative evaluations appear to be key during relational work scenes (e.g., creepy, harsh, inconsiderate, rude, ungrateful).

Looking first at the positive terms in Table 11, we will again only investigate the most promising terms further, which are *kind, mature* and *smart.* For *kind* we first removed cases that referred to *kind of* or *kinda*. The remaining references to *kind* appear to be constituted mostly by character evaluations (30). The same is true for *mature* which seemed promising, because it also refers to age, but usage of *mature* in fact also consists of positive evaluations of character appearance and/or attribute (31).


Table 11: Positive evaluations of social interactions in K-TACC-RW and K-TACC-non-RW, ordered alphabetically

**\*** Terms marked with an asterisk appeared at least 5 times per 100,000 words and are more frequently in K-TACC-RW

(30) oh that's so **kind** - she doesn't even ask for anything but her motherly instinct makes her want to protect him (YAS, Ep. 13)

(31) She's my favorite. She's so **mature**. (OMHE, Ep. 4)

Similarly, *smart* is used as a positive evaluation of intelligence as a character trait, sometimes informed by character actions (32). When negated, such terms also occur as negative evaluations (33).

(32) that boy **smart** (W, Ep. 12)

(33) A little doubtful about her being **smart**...lol (YAS, Ep. 11)

Judgement on character by means of lexical items from the semantic field of (im)politeness can be interpreted as evidence of the link between relational work and identity construction and the inherent process of judging the others' and one's own behavior (Locher 2008). Overall, we find that relational work scenes do not lead fans to contribute many positive comments about characters' interactions, while negative comments are more frequent, as exemplified in the next step.

Given that some terms within that category are keywords of K-TACC-RW, negative evaluations of social interactions appear to be a particularly good category for the exploration of discussions of relational work. Our searches, as documented in Table 12, led us to six terms in particular that are worth pursuing further in our list of concordances: *cruel, harsh, horrible, mean<sup>8</sup> , rude* and *ungrateful.*


Table 12: Negative evaluations of social interactions in K-TACC-RW and K-TACC-non-RW, ordered alphabetically

**\*** Terms marked with an asterisk appeared at least 5 times per 100,000 words and are more frequently in K-TACC-RW

<sup>8</sup> In the case of *mean* we removed false positives, references to the verb *to mean* before examining the concordances.

In this category, too, we find many instances of evaluation of character traits which again points to the importance of relational work vocabulary for character identity construction. In addition, we see some awareness of relational work relevant aspects of the ongoing interaction on screen, in particular when the evaluative adjectives are connected to linguistic aspects.

(34) He seems a little **cruel**. Saying unneccesary things... (OMHE, Ep. 2)

In Example (34) from the second episode of *One More Happy Ending*, for instance, a commenter picks up on the way Kim Seung Jae talks to his ex, Han Mi-Mo. Similarly, a commenter finds harsh how Ah Soon Soo rejects Goo Hae Joon in a scene in episode 9 of the same Korean TVdrama (35). In that case, the assessment is challenged by another commenter immediately (36), thus starting a negotiation of what is and is not appropriate interpersonal behaviour from the perspective of the viewership.


We find similar comments also for *horrible* and *mean* (37 and 38). These terms are used for the evaluation of characters in general, and they are employed frequently to specifically address interpersonal behaviour and social interactions and thus represent what we were looking for in K-TACC-RW.


Perhaps the best indicator of negative evaluation of social interaction is the term *rude*, which is unambiguously used in reference to relational work behaviour. Of the 69 instances in K-TACC-RW, 3 times *rude* is used to negatively evaluate commenters' behaviour – all other cases refer to character interactions, which are deemed inappropriate by commenting fans. A typical example is (39) and (40), where a commenter negatively evaluates the character Han Mi-Mo's behaviour as inappropriate. Sharing a meal with her boyfriend Goo Hae Joon in a restaurant, Han Mi-Mo keeps talking about a different man instead of focusing on her company. Goo Hae Joon appears to visibly dislike this and then voices his displeasure.


The comments in (39) and (40) are not the only negative comments about Han Mi-Mo's behaviour in this scene. We find other viewers sharing their negative assessment, a selection of which is shown in (41). By identifying a comment that has a clear meta-pragmatic marker such as the comments containing "rude", we can find further comments made nearby and thus arrive at further discussions of relational work.

(41) Selection of comments on Han-Mi-Mo's behaviour (OMHA, Ep. 7)


Given the fact that assessments similar to (39) and (40) are four times more frequent in K-TACC-RW than in other scenes, we find that commenters are particularly keen observers when it comes to behaviour they perceive of as very impolite during relational work scenes.

Finally, *ungrateful*, which we included because it is a keyword in K-TACC-RW, turns out to point to particularly strong negative evaluations by fans of behaviour that goes against their expectations (i.e. they expected gratitude and evaluate negatively when none is displayed by the character). This is best illustrated by the nouns that are pre-modified by *ungrateful*, which include *bitch, biatch, bastard,* and *son of a bitch*.

In general, we find that negative evaluations of character interactions are a key part of fan discourse in K-TACC-RW and thus in comments added to scenes in which relational work is saliently staged. While expressivity seems the main direction these comments take, in the form of one-to-many sharing of social evaluations, we do find some evidence of communal negotiation of evaluations when initial appraisals are challenged. In context, both the scenes themselves and the comments as complementary discourse are rife with moments for learning about (im)politeness norms in Korean TV-drama, but also about the inferred norms shared and negotiated by the fans. In contrast, *explicit* teaching and learning moments in the form of question and answers that would take place in the comments themselves appear to be rarer. We cannot offer a satisfactory explanation of the scarcity of such explicit moments at this point in time. However, one tentative rationale that could be explored further is whether it may indicate that rather than considering (im)politeness as something that needs to be learned, viewers refer and reveal their own norms – presumably a blend of norms they have witnessed in other Korean TV-drama episodes and the norms they transfer from their own culture.

4.2.4 Learning About Language Use, Relationship and Age

As a final corpus-assisted approach to comments during relational work scenes, we looked for keywords that did not fit the topoi we extracted from Framenet and complemented them with our own introspective list of terms. The goal was again to find engagement with the negotiation of interpersonal relationships and more specifically differences in this area between Korean and other cultures. Table 13 shows that, contrary to the Framenet-derived lists, the more specific

terms were indeed mostly relevant for K-TACC-RW, with 9 out of 15 terms being frequent and relevant enough within the parameters we set (>5 per 100k words and more frequent in K-TACC-RW than in the reference corpus).

Table 13: Culture and relationship terms in K-TACC-RW and K-TACC-non-RW, ordered alphabetically


**\*** Terms marked with an asterisk appeared at least 5 times per 100,000 words and are more frequently in K-TACC-RW

We will discuss these concordances based on three subgroups: *Speech* (*accent, argu*), *Relationship* (*relationship, treat*) and *Age* (*age, old, year, young*). The terms in the *Speech* group point us to comments about accents fans identify as Busan, or Seoul, mostly together with a positive evaluation, as in (42), where one character's accent is seen as a positive change from the more common Seoul accent.

(42) um, i love her **accent** actually...the Seoul accent gets old after while! (YAS, Ep. 1)

Commenters thus pick up on the potential of linguistic indexicals for character positioning (Locher & Jucker 2021), which we argue belongs to relational work. This is also evidenced by the use of dialect and satori (the Korean expression for dialect) in both sub-corpora.

The term *argu\*,* on the other hand, seems to mostly be used in episode 5 of *W* to compare styles of conflict, with commenters juxtaposing character arguments with those they experience in their real life (43 and 44).


The *Speech* categories thus exhibits awareness of characters' choice of linguistic code and perhaps of the comparability of conflict situations inside and outside Korean fiction.

Similar comparisons are also found in the *Relationship* group, where commenters use the term *relationship* to take sides in social relationship negotiations (e.g. what is going wrong in a fictional relationship and who is to blame (45). The term *treat* in the same group interestingly is mostly used to predict or request future character behaviour and thus to perform imagined interaction between fans and fictional characters (46).


Especially relevant for relational work is the *Age* group, which points to one of the deciding factors in the choice of linguistic features in Korean (Yuh 2020). The term *age* itself is used in descriptions and evaluations of the physical appearance of characters, assessment of behaviour relative to age, but more importantly also to specifically comment on the appropriateness of characters' linguistic choices (47), or the Korean TV-drama genre norm of discussing age (48).


While *young* (including *younger* and *youngest*) is predominantly used to positively assess character appearance and thus bears little relevance for our study, *old* (including *older* and *oldest*) points us to interesting fan observations about the relevance of age. Examples (49) and (50) on the one hand discuss the significance of age for the choice of linguistic features and on the other hand tie it to particular Korean address terms which are borrowed in the comments (also see Section 4.1). Examples (51) and (52) indicate that age is also discussed in the comments as a more general norm-giving factor that is deemed relevant for judgements of appropriateness by commenters themselves as well as the judgements commenters expect fictional characters to make.


Our corpus-assisted exploration of viewer comments has given us a systematic way of comparing what fans talk about during relational work and other scenes, but has also forced us to examine individual comments rather than longer sequences of comment interactions, and also to look at the comments detached from the original video input that triggered the comments. In order to add a more context-inclusive lens we will dedicate the final section to the multimodal aspects of relational work.

#### *4.3 Multimodality: Hitting, Bowing and Lowering Your Gaze*

The fact that relational work is multimodal is an established fact (see, e.g., Brown & Winter 2019; Kiaer et al., this volume) but it is rarely researched in depth due to methodological laborintensive challenges of aligning multi-modal corpora. We too have only made comments on multi-modality in passing within our research when we discussed scenes in context. The subtitle MoRWF Corpus is unlikely to show much textual evidence of mulitmodality since the video and sound input is not transcribed in the subtitles. Whenever there was a link to relational work negotiations, we did however include the scene in the MoRWF corpus. For example, in (53) from *Meloholic*, a scene is depicted in which the main character Han Ye Ri, who is portrayed as having two personalities, one gentle and kind and one rude and obnoxious, is introduced through a flashback as dominating the classroom. The students are all sitting at their desks before class starts and a classmate accidentally trips over Han Ye Ri's foot while walking to his seat, which causes her to lose her shoe.


(53) *Meloholic,* Ep. 3, 00:38:18, subtitles in Korean and English from Viki (comments in square brackets are not in the original subtitles display)

In line 1, the student who tripped complains almost inaudibly but Han Ye Ri does not apologize. Instead, she commands him to not look at her and sit (line 2) and other students to return her shoe (line 3). She then challengingly looks around the classroom and puts her classmates in their place by asking a number of girls to lower their gaze (in line 4) and to quit staring (line 5). This is accompanied by swearing noises (aishh, 씨), which are translated in the English version as 'the heck' (l. 2, 5) and is accessible through the audio track. In this way, Han Ye Ri positions herself as having higher status as a direct eye gaze is considered as challenging the addressee when done by a person of lower status. In the context of the classroom, where in theory classmates would be equal, she thus claims a higher status among her peers. This scene did not receive any explicit comments on relational work but the viewers commented on the character creation ("Her baddie side is out" / "I love her baddie side. Omg. Her and i would be the best of friends"); it is thus recognized by these viewers that her verbal and non-verbal behavior is part of a character creation that transgresses norms.

In contrast, a much longer scene from *W* that contains embodied as well as linguistic relational work of various kinds is commented on abundantly by the viewers (54). In this scene from *W*, the main character Oh Yeon-joo, who is a doctor, has just finished surgery (to which she came late) with her professor. They are getting ready to wrap up and Oh Yeon-joo wants to quickly leave the theatre because she needs to check on the other main character, who is waiting

for her. A quick departure is prevented by the professor who had postponed reprimanding her for being late until after the surgery was finished. However, the scene quickly moves from discussing work related issues to private issues and the Professor's passion for the webtoon *W,* which is written by Oh Yeon-joo's father.


(54) *W,* Ep. 5, 00:27:27.260, subtitles in English from Viki

In this scene we see power difference and different status enacted on several levels. On the one hand, the professor claims the right to stop Oh Yeon-joo from leaving the theatre and reprimanding her for her unprofessional behavior. He also uses meta-pragmatic markers when he claims that she had "no manners" (55, line 8) and was "acting up" (line 9), thus rebelling against his authority which he presents as the natural order of things. These linguistic markers are enforced by spatial ordering and physical action as well. In line 1, Oh Yeon-joo's body is oriented towards the door and she expresses urgency to leave through this orientation, which is underlined in line 6, where she resists the professors order to approach (uttered in line 5), and later in line 13, where she repeatedly glances up to the side (presumably to see the time on a clock mounted in the theatre). The professor insists in her taking on a subordinate body position – right in front of him with head and eyes lowered – and then starts his tirade about her acting up, repeatedly smacking on the side of her head with his face mask. Oh Yeon-joo maintains the lowered head and eye gaze until the end of the extract but in the continuation of the scene she raises her eyes again once the professor is clearly discussing the plot and genre development in the webtoon *W* and has thus veered from the medical context. In the webtoon, Oh Yeon-joo claims authority herself and this is expressed in her heated retorts concerning the development of the plot in *W* and body stance (not shown here). The scene is thus a nice example to show how physical action, body comportment, eye gaze and language are drawn on in combination to stage characters and their relationships.

Comments by viewers made about this scene are abundant (N=578) and some pick up on the mention of 'manners' in line 8 but especially on the fact that a superior is smacking a subordinate, reprimands her for work-unrelated matters and in general is behaving in what the fan community calls unprofessional. Here is a selection of comments to illustrate this stance.

(55) A selection of comments on *W*, Ep. 5, 00:27:27.260


The comments thus show awareness of negotiations of status and underlying hierarchies. The observed behavior is also assigned to being "Korean", thus implicitly contrasting what the commentators see with (presumably) their own culture.

To give justice to the breadth of opinions in the comments on this scene, it should also be mentioned that some commentators like the doctor for his passion for the webtoon *W* and thus applaud him for exactly the opposite of what other commentators highlight (i.e. unprofessional, non-work related conversations). They make analogies between the professor and Oh Yeon-joo as fans of a webtoon who discuss their favorite characters, just like the Viki community discusses their bias in the TV drama.

(56) A selection of comments on *W*, Ep. 5, 00:27:27.260,


In other words, while some viewers choose to focus on the multi-modal relational work enacted in the scene and on the link to Korean culture, others focus on different aspects of the scene, such as its comic potential and links to fandoms.

When considering K-TACC overall, there are more avenues one could explore. For space reasons, we cannot do so but would suggest that collocations around bowing, nodding and eye gaze that have been associated with expressing Korean politeness would be worth exploring further.

#### **5 Conclusions and Outlook**

Our argument in this chapter was to explore evidence on teaching and learning about Korean (im)politeness ideologies in the Viki fandom who watches Korean TV dramas. Our findings on the subtitles corroborate our earlier study in that Korean TV drama scenes on relational work that (a) there are many moments of relational work scenes included in Korean drama and that (b) these scenes are rich in relational work indexicals. In addition, we found evidence in the timed comments that (c) these scenes can act as introductions to the importance of relational work negotiations in Korean society.

Our findings on the timed comments show that there are indeed many implicit comments on Korean relational work in our corpus but that explicit meta-comments are not particularly frequent when compared to the overall number of comments that viewers write and the many other functions that viewers use the comments for. Having said that, the mere fact that there is an average of 2.9 scenes containing moments of relational work per episode means that viewers can hardly fail to notice the importance of relational work negotiations in Korean society. This is corroborated by the question and answer sequences about relational work that we were able to identify and discuss.

The idea of moments of teaching and learning about Korean relational work that we took as our leading theme throughout the chapter thus plays on two levels. First, the Korean drama, made accessible through subtitles, provides ample examples for viewers interested in Korean culture to notice, pick up on and question relational practices. Such scenes present implicit teaching moments. Despite the fact that these plot-relevant scenes were not created with the second language classroom in mind, their inclusion in the classroom might easily spice up the second language classroom as lively examples of fictional interactions that are set in Korean contexts. With respect to fictionality, it is at the teachers' discretion to choose examples that are close to face-to-face encounters rather than using atypical, artful or playful examples that might distract from the teaching aim. Viki itself is quite aware of the teaching potential that the platform has and also provided subtitles in Korean, written in Hangeul, where each word is interlinked with dictionary entries (the so-called learner mode). Second, we found moments of teaching and learning in the timed comments where the contributions in their sum can add to a better understanding of Korean culture and Korean (im)politeness. With respect to both subtitles and viewer discussions in timed comments, we should stress again that these instances are

starting points for awareness raising. Neither the space provided in subtitles nor the discussions in timed comments usually go beyond scraping the surface of the complex relational work in place. As moments of awareness raising, however, both practices work hand in hand.

It is important to stress that we have no evidence of 'learning' in the sense of effectiveness. We do not know whether viewers *actually* understood and learnt from being exposed to the scenes and reading the timed comments. We do, however, see the possibility that viewers open to learning about Korean (im)politeness have opportunities of such learning on the streaming platform.

The scope of our paper is modest in so far as we looked at one streaming platform only and did not follow up on neither individual users nor on other practices that viewers might engage in outside of Viki. Through our combination of qualitative and quantitative methodological steps, we were however able to paint a picture of the fan community as interested and active and as creating a fan identity through their engagement with the cultural artefact.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **"Koreans Are Always Nodding or Bowing": K-Wave Fandom's Perception and Learning of Non-Verbal Politeness**

#### **Jieun Kiaer, Loli Kim, and Alfred W.T. Lo<sup>1</sup>**

**Abstract** This chapter explores how K-wave viewers' exposure to dramas and reality TV shows informs their perception and understanding of non-verbal politeness. In particular, we focused on how participants in K-wave fandom acquire the notion of non-verbal honorifics and politeness. We adopted a multimodal qualitative design by including learners in a multilingual Korean classroom in the United Kingdom to express their perceptions of non-verbal politeness through a think-aloud protocol, as well as creating a multimodal text to express the mental thinking process of their notion of politeness through non-verbal acts. Findings revealed how different levels of learners perceived non-verbal politeness differently. Beginners were unaware of Korean pragmatics and the significance of semiotic resources. While intermediate learners demonstrated a better grasp of non-verbal behaviour, it still lacked nuance. Advanced participants displayed a thorough understanding of non-verbal politeness. This paper also demonstrates how watching Kdramas and K-films can provide insight into learning of pragmatics without the need to exert much effort on the learners' part and supplement traditional textbook-based learning. It also sheds light on how integrating media into language learning could prove particularly beneficial for learning East Asian languages, providing pedagogical and research implications for both language educators and researchers.

**Keywords** Non-verbal Expressions Gestures Language Learning Politeness

L. Kim

Asian and Middle Eastern Studies, Oxford, England, United Kingdom e-mail: loli.kim@ames.ox.ac.uk

A. W. T. Lo Asian and Middle Eastern Studies, Oxford, England, United Kingdom e-mail: alfred.lo@ames.ox.ac.uk

<sup>1</sup> J. Kiaer ()

Asian and Middle Eastern Studies, Oxford, England, United Kingdom e-mail: jieun.kiaer@ames.ox.ac.uk

#### **1 Introduction**

With the rise of the Korean Wave (*Hallyu*) since the start of the new millennium, learning the Korean language has sparked huge interest around the globe, both online and offline. For online language learning, this was exemplified by a 94% increase in people wanting to learn Korean on one popular language learning app following the release of Netflix's smash hit *Squid Games* in 2021 (Memrise, 2021). The growth of Korean language study is visible offline in schools and universities: one study by the Modern Language Association shows that Korean is the only language with a growing up-take in US universities with an increase of 5.3% between 2013 and 2016, despite an overall decline in all foreign language enrolments. The BBC also reported that 14,000 students are studying Korean in American universities, as compared to 163 two decades ago (Pickles, 2018).

Despite the proliferation of Korean language learning, learners—particularly those from countries not geographically close to South Korea—often struggle with acquiring the language. One major obstacle is the socio-pragmatic complexity of Korean, especially in terms of 'politeness'. The level of politeness employed significantly influences the identification of relationships between speakers and hearers during communication, as Korean is a sociopragmatically rich language. Although essential for understanding the Korean language, culture, and society, acquiring these interpersonal skills using textbooks alone can be challenging. These textbooks are often criticized as inauthentic, leaving students feeling ill-equipped to discuss Korean culture and language in everyday life (Brown, 2010; Kiaer, 2017b; Kiaer et al., 2022). For instance, written materials like textbooks typically provide only 'plain text', exposing students to just one level of honorifics and politeness. In contrast, other speech levels are neglected, and classroom teaching centred around textbooks can make students feel more challenged when encountering a broad range of social situations in daily communication with Koreans. Effective communication relies on both verbal and non-verbal expressions. In particular, non-verbal expressions such as bowing, nodding, and posture are common in the Korean context, serving as means to convey attitudinal meanings. They are almost inseparable from the honorific system of Korean pragmatics. However, the presence of these complex gestural, non-verbal honorifics can potentially impede learners from becoming effective communicators. Where pragmatic competence cannot be acquired through studying written materials alone, students' primary motivation for studying Korean—such as Korean pop music, dramas, and films—can provide valuable input. Far from being frivolous, K-dramas and K-films are rich and useful learning tools that exhibit both the verbal and non-verbal aspects of communicating in Korean. In our observations of Korean language learners, we discovered that some expressed the notion that 'Koreans are always nodding or bowing'—a phrase so striking that it also appears in the title of this chapter, which has motivated us to explore how and why Koreans might perceive these nonverbal behaviours as polite.

Research on Korean language learning has extensively explored how learners understand and employ politeness through verbal communication (e.g., Byon, 2006; Suh, 1999). However, the non-verbal aspects of the Korean honorific system, which are crucial to Korean pragmatics, have often been overlooked (Kiaer, Shin, Driggs, 2022). Notably, few studies have examined how fandom culture influences the non-verbal learning of politeness. The Korean Wave (K-wave), or Hallyu, attracts fans to Korean dramas and films, providing strong evidence that media-mediated learning can help students grasp various levels of politeness, thereby overcoming the limitations of repetitive classroom language instruction. This paper investigates how exposure to K-wave dramas and reality TV shows shapes viewers' perceptions and understandings of politeness (see also Locher & Messerli, this volume). Our research aims to examine how media supports the Kwave fandom in acquiring non-verbal honorifics and politeness in the Korean language. In this chapter, we begin with a literature review on Korean verbal and non-verbal politeness. We then describe the methodology used to study the K-wave fandom's perception and learning of nonverbal communication. The results are presented and discussed critically, focusing on the effectiveness of media-mediated learning in acquiring non-verbal politeness. Finally, this study highlights advancements in Korean language pedagogy through the use of media and technology to facilitate embodied language learning and proposes future research on interventions to study how learners can acquire non-verbal politeness through media-mediated language learning.

#### **2 Literature Review**

#### *2.1 Politeness and honorifics*

Politeness is one of the main ways that humans devise and maintain interpersonal relations. At the core of polite expressions is a conveyance of consideration to the hearer, or to the person, subject, or object that is being spoken about, whether genuine or manufactured simply for social lubrication. Locher (2004: 91) theorises politeness as a 'speaker's intended, marked and appropriate behaviour which displays face concern' and 'the motivation for it lies in the possibly, but not necessarily, the egocentric desire of the speaker to show positive concern for the addressee and/or to respect the addressees' and the speaker's own need for independence'. However, despite politeness having this common goal, how it is executed is far more complex than simply choosing considerate over inconsiderate expressions.

As Kadar and Haugh (2013: 1) explain, '[…] in many cases politeness does not come into existence simply through what is said in the moment, as many social actions and pragmatic meanings that are understood in locally situated contexts in fact follow pre-existing (often formalised) patterns', adding that some forms of politeness are 'historically situated' too. Politeness is also multimodal (Kiaer, 2022; Kim and Kiaer, 2022) and multifaceted (Kiaer and Kim, 2021). As Brown et al. (2014) argue, 'politeness research has long recognized that politeness resides not just in what people say, but also in how people say something'. Thus, this sensitivity to local contexts and their conventions is extended beyond simply words to multimodality, and 'how' multimodality, as opposed to only verbal language, is used.

Perception of politeness is therefore a similarly complex matter, since formalised patterns are read by the same means by which they are employed, and even more so if one considers that unlike interactions, in which known languages are used in order to obtain the understanding of the hearer, perception is constant, and total, in that it is how everything one experiences is mediated on the individual level. Inevitably expressions are encountered that are not in the hearer's semiotic repertoire, and as such these expressions may be rendered untranslatable—either entirely invisible or mistranslated (Kiaer, 2019).

In Korean, the formalised patterns that Kadar and Haugh (2013) refer to take form in the Korean system of interpersonal relations, which is firmly rooted in Confucian ideology (Hong, 2009). This system guides how interlocutors of various ranks and statuses should interact with each other, using a repertoire of honorific and non-honorific verbal and non-verbal expressions that are meaningful when formulated with certain social factors, and of which there are a vast

variety. Korean communication is thus different from English in almost every respect (Kiaer, 2017, 2019). Thus, difficulties are often commented on arising for Korean second language learners, both in learning and, for the many learners whose interest in Korean is influenced by Korean popular culture, in understanding Korean content too.

To elaborate further, in Korean honorifics are used to be polite, by showing respect and humility. Korean is not the only language to use honorifics. Japanese has perhaps one of the most similar systems to Korean. However, the extreme level of complexity of the Korean case is quite unlike other honorific systems. While 'honorifics' refer to forms of language that honour the hearer (e.g., respectful address terms), Korean also has speech levels, first- and second-person pronouns that involve the increase or decrease of one's status (e.g., the humble *jeo* (저) or more self-centred *na* (나), use of *dangsin* (당신) 'you' instead of an address term), and one honours objects and subjects as well as the hearer. As Kiaer (2023) explains, 'Even if interlocutors are speaking about an impersonal subject, e.g., an inanimate object, it is unavoidably laced with meanings to the effect of "I don't care who you are" or "I value you" or "you are an important person" in addition to "I think…", "I feel…" or "my attitude is…".

Speech styles and address terms are often referred to separately, however, if we consider how they are 'aligned' together with other modes in the orchestration of Korean communication, and the potential for communicating sarcasm otherwise (Brown and Winter, 2019: 32, Kiaer and Kim, 2021: 143, Kiaer, 2020: 93), then speech styles and address terms can be grouped together as components of the same styles, according to their politeness level. It is often agreed that there are six speech styles (Yeon & Brown 2011: 17): 1) *formal*, 2) *semiformal*, 3) *familiar*, 4) *plain*, 5) *polite informal*, and 6) *informal*. *Formal*, *polite informal*, and *informal* styles are commonly used in present-day Korea, with *polite informal* style also referred to as *'jondaenmal'*, recognisable by the154olitenesss marker *-yo* added to the end of sentences. *Informal* style is *known as 'banmal', 'half-talk', or 'intimate style', and* recognisable by the absence of the politeness marker, which conveys rudeness or intimacy. There is then a plethora of address terms, which can be paired with these styles, including honorific suffixes such as *-nim* (-님) or -*kkeseo* (-께서); the former corresponds with *jondaenmal* (존댓말) and also *formal* styles, and the latter with the *formal* style only.

Example 1 illustrates a variety of ways that a teacher can be respectfully addressed when they are the subject of the sentence. This example demonstrates how speech styles and address terms are orchestrated, and common reasons for doing so. Further categorisation in addition to the politeness styles named above is made in I, to demonstrate the granularity to these levels, which could not be explored in-depth in this chapter, but nonetheless are important considerations.


'**Respected** teacher Kim, could you **please** do this?' (informal polite – could be used to junior/equal but if used inappropriately, could be conflicting)

*c) Kim seonsaeng-nim, i-geo hae-jusibsio* (김 선생님, 이거 해주십시오)

#### '**Respected** teacher Kim, would you **please** do this?' (polite)

*d) Kim seonsaeng-nim-kkeseo, i-geo hae-jusiki parabnida* (김 선생님께서, 이거 해주시기 바랍니다)

#### 'Respected and **Honourable** teacher Kim, we **humbly implore** you to do this' (formal)

The selection of these components is dependent upon three primary factors: (1) the relative status of the speaker and hearer, (2) whether they have an intimate relationship, and (3) whether the situation is a formal one (e.g., a special occasion, or an interaction being witnessed by nonintimates or spectators who hold seniority or higher status). Koreans will consider these factors, amongst others, such as medium, to work out how to communicate in a way that meets the requisite for politeness.

This is vastly different from Western linguistic and cultural contexts, not only in how social factors are calculated, but also in the strategies employed to meet the social requirements. For example, if we consider the creation of 'distance' in communication, while an American might create less distance to be friendly, a Korean might create more distance to come across as respectful and polite. Ultimately both have the same goal of a smooth interaction, through consideration, however, the strategies being employed to achieve this follow opposing conventions. In Korean, the strategy is to show respect and therefore honorifics are employed.

Brown (2011: 2) argues that the 'difficulty and importance' attached to Korean honorifics and politeness in second language learning is the very reason for further research, stating that 'previous research into the acquisition, development or usage of honorifics of L2 Korean is limited'. Brown, whose study investigates the use of Korean honorifics by second-language learners from Western backgrounds, explains that lack of research is due to honorifics being incorrectly treated as a grammatical feature of the language, when honorific usage is actually 'always and undeniably linked to social and cultural factors'. This means that learning honorifics requires more than simply learning the language 'system', but requires what Brown describes as 'the negotiation of socio-pragmatic knowledge and questions of what it means to use the language appropriately and to be a "polite" speaker'. Interestingly the data collected by Brown also indicated that this is an element of Korean that is especially difficult to master, since his findings showed that participants, regardless of their proficiency, often applied honorifics outside of local norms.

#### *2.2 Non-Verbal Politeness*

Despite the multimodal nature of communication, focus has traditionally been placed on the verbal aspects of (im)politeness rather than the non-verbal. This trend is part of a general attitude towards non-verbal gesture as being less significant than verbal language, even if communicative. For instance, Kendon (1988: 132) argues that words must follow a standard set of syntactic rules, whereas gesture is less regulated in how it is used. McNeill (1992: 19) agrees, arguing that there is a fundamental difference between non-verbal gesture and verbal language. Wharton (2009: 153) argues that non-verbal gestures are communicative, but only if they are salient and relevant.

In fact, even beyond (im)politeness research, many mention a lack of research on nonverbal gesture in general, and especially when it comes to East Asian gesture. Within the small pool of research that exists on Korean gesture Kiaer and Kim (2021) and Kiaer (2022), and more broadly on Asian gesture, Kiaer and Kim (2023) are the first to tackle the subject in depth. All these works argue for the essentiality of non-verbal gesture in systems of communication and emphasise the significant role of non-verbal gesture in Asian languages. In the case of Korea, Kiaer and Kim (2021) and Kiaer (2022) argue that due to their inseparability and mutual dependency, non-verbal gestures are as important as verbal language and their collaboration with verbal language as meaningful as mono-modal expressions. For example, alignment and misalignment of non-verbal with verbal can change the meaning entirely (Brown, 2013). Kiaer and Kim (2023) extend the conversation to Asia, stressing the need for cross-cultural and socio-pragmatic-inclusive perspectives on non-verbal gestures across the continent. Brown and Winter (2019: 26) likewise argue that research on non-verbal politeness is lacking and neglected, as do McKinnon and Prieto (2014) and Brown et al. (2014).

The notion of non-verbal politeness being as integral as verbal politeness is however now increasing in politeness literature, and especially in Korean politeness. Watts (2003 :8) argues that politeness involves more than merely speech, but indeed space and time, arguing, 'It would also seem that whether or not a participant's behaviour is evaluated as polite or impolite is not merely a matter of the linguistic expressions that s/he uses, but rather depends on the interpretation of that behaviour in the overall social interaction'. Kiaer (2022) argues that Korean 'bodily speech' is aligned with Korean verbal speech to clearly communicate one's attitudes, emotions, and thoughts, following the same rules for navigating interpersonal relations as verbal language. Kiaer states:

> In Korean, one's 'bodily speech' must continuously inform one's fellow interlocutors of the evolving interpersonal dynamic on Confucian-terms. This means that one's bodily speech must inform the other person of how they think of them and themselves in relation via the hierarchies in the social dynamic at play. So, if the hierarchy is junior-senior, then the junior's attitude will be conveyed by either submitting or contending with that hierarchy; and if the hierarchy is conflicting, the attitude of either will be conveyed by how they balance the value of their hierarchical claim with the other person's.

Kiaer and Kim (2021) also stress the effect of alignment and misalignment on the meaning expressed by non-verbal gestures. Kiaer (2020) proposes a multimodal modulation hypothesis that also encompasses this concept of (mis)alignment, in references made to modes being orchestrated in 'harmony'. The hypothesis argues that politeness levels are constantly in flux, according to a huge variety of social factors that are also changing, and as such the negotiation of these factors is reflected in the verbal and non-verbal communication of interlocutors. Kiaer (2020) explains:

> The core linguistic ability found in human communication is to be able to modulate or attune/orchestrate different levels/modes of information in a harmonious way, sensitive to the socio-pragmatic needs of each situation. If conflicting or inconsistent meanings are communicated, the communication will become sociopragmatically inappropriate, insincere, or unreliable.

Kiaer and Kim (2021: 71-72) even categorise the gestural components of bodily speech according to their politeness levels, demonstrating the inherence of politeness in the Korean nonverbal communication system and of non-verbal expressions in Korean politeness. They determine politeness levels based on a variety of social factors, including age, socio-economic status (SES), class (for period films), position (e.g., professional, familial), gender, intimacy, and additional factors such as generational differences are also considered in some examples. Kiaer and Kim argue that conventional meanings can be inferred in a range of common contexts using these politeness classifications as guides for the interpretation of non-verbal and verbal expressions. Kim and Kiaer (2022) conduct analysis demonstrating this further, collecting substantial data to support Kiaer and Kim's (2021) framework. Both studies serve to demonstrate the systemic nature of the link between politeness and non-verbal gestures in the Korean communication system.

In a Korean context, gestures are inseparable from the honorific system. They are an essential part of any Korean interaction and can support or even undermine verbal expressions. Some of the main non-verbal 'primitives' that have been identified include bowing at a variety of degrees (e.g., 15-, 30-, and 45-degrees), nodding, broad/compact posture, direct/indirect eye contact, the position of the hands, giving/receiving with one or two hands, patting, and other forms of *skinship* (Kiaer and Kim, 2021: 70; 75-147); beckoning with an open hand, and crossing the arms or legs (Kim and Kiaer, 2022). To learners of Korean and other Asian languages, the presence of these gestural honorifics may pose challenges in becoming pragmatically effective communicators, and it is pragmatics that are fundamental to mastering Korean, because meeting the requisite for expected politeness ensures smooth and effective communication. Not doing so can have serious consequences.

The cross-cultural difference specifically between non-verbal gestures in Korean and Western cultural contexts is significant, with non-verbal gestures often playing a major role in the 'culture shock' experienced by Westerners when they encounter East Asian cultures (Kiaer, 2017a, Kiaer and Kim, 2021). Since these non-verbal gestures inherently communicate pragmatic meanings, this shows how integral the perception of the non-verbal is for understanding and learning how to communicate in Korean. It also indicates the association between Korean nonverbal competency with higher levels of Korean language competency, since to be competent, one needs to develop Korean interpersonal skills. It is striking, then, that explicit instruction on any kind of non-verbal communication other than bowing is typically absent from traditional, textbased Korean teaching material.

#### **3 Methodology**

#### *3.1 Research Question*

Given that the non-verbal aspects of pragmatics are essential in the Korean language, this paper explores how Korean language learners inform their perception and understanding of politeness. In particular, we focused on investigating how K-wave viewers' exposure to dramas and reality TV could facilitate the acquisition of the notion of non-verbal honorifics and politeness. The following research questions were formulated:

1. In which ways do Korean language learners perceive Korean politeness, verbal communication, non-verbal communication, or both?


#### *3.2 Study Design*

A multimodal qualitative design was employed to investigate how the K-wave fandom might enhance the acquisition of non-verbal honorifics and politeness. Given the complexity of participants' perceptions of politeness, sophisticated methods were necessary to capture the nuanced understanding of complex social situations that extend beyond what plain text alone can convey. In this study, participants engaged in a workshop where they reflected on their experiences with Korean language learning through verbal, visual, and written presentations. This integrated approach, which combines visual and word-based research methods, provides a means to explore the multiple dimensions of human experience foundational to much social research (Guillemin, 2004, p. 273). Relevant to the study's focus on non-verbal politeness in language learning, the inclusion of students' drawings aligns with the visual methods in qualitative research that emphasize the importance of the non-verbal dimensions of human experience, which often elude verbal articulation (Silver, 2013, p. 480). Harper (2002) argues from both a phylogenetic and ontogenetic perspective that images provoke stronger reactions at and below the level of conscious awareness than words, as the visual-processing parts of the brain are evolutionarily older than the verbal-processing parts. Exploring the non-linguistic dimensions in politeness research enables access to and representation of diverse experiences from Korean language learners who are also members of the K-Wave fandom. In this context, visuals serve not only as research outcomes but also as a process that enables participants to access and interpret their prior Korean language learning experiences. The participants' texts, speech, and images are interwoven to generate data, thereby providing insights into their perceptions of the non-verbal aspects of politeness in Korean language learning.

#### *3.3 Research Context*

The study was conducted during weekly voluntary Korean language lessons in 2022 at a UK university, organized by the Korean Society. These sessions offered free Korean language instruction and cultural discussions without requiring prior registration, allowing anyone interested in learning Korean within the district to participate at their convenience. Consequently, participants were not pre-selected for the study. The instructional team comprised a native Korean speaker, a native Cantonese speaker, and a native English speaker.

The class was divided into three sections based on the students' self-assessed proficiency levels: elementary, intermediate, and advanced. This categorization also helped differentiate responses within the study. At the elementary level, students were familiar only with Korean vowels and consonants and basic phrases such as '안녕하세요' (Hello) and '감사합니다' (Thank you), possessing limited vocabulary. Intermediate students were expected to have reached at least Level 2 on the Test of Proficiency in Korean (TOPIK), enabling them to use basic commands,

express personal thoughts, comprehend everyday conversations, and distinguish between formal and informal contexts. Advanced students generally achieved at least Level 4 on the TOPIK, equipping them to engage in transactions in public settings, maintain social relationships, and understand news articles, general social issues, and abstract concepts with fluency.

#### *3.4 Participants*

As both teachers and students volunteered for the lessons, there was no purposeful sampling in recruiting Korean language learners. For the study, students received an invitation email describing the session as a workshop focused on exploring their perceptions of verbal and non-verbal communication in Korean films, specifically targeting those with an interest in Hallyu. While informed that there would be no immediate benefits from participating, attendees were provided with refreshments, including Korean snacks and drinks.

The study comprised fifteen participants, aged between 19 to 32 years (mean age = 23.3, *SD* = 3.42), consisting of four males and eleven females. The group included ten undergraduate students and four postgraduates from three different universities, along with one full-time worker. The linguistic background of the participants varied, with eight native English speakers, and one each of Bulgarian, Cantonese Chinese, Danish, German, Mandarin Chinese, Nepali, and Spanish. All participants were either bilingual (*n* = 1) or multilingual (*n* = 14), knowing more than two languages. Regarding Korean language proficiency, nine had been learning Korean for less than a year, four for nearly a year, and two for over a year (1 year and 5 months, and 6 years, respectively). Participants reported using a combination of self-study (using Korean textbooks), attending regular classes, engaging with media (Korean pop songs, dramas, and films), and using language learning apps as sources for learning Korean. The primary motivation for learning Korean was the K-wave (*n* = 12), such as listening to music or watching dramas without subtitles, while other reasons included general linguistic interest, the desire to converse fluently with native Korean speakers, and family-related motivations (*n* = 3).

#### *3.5 Research Instrument*

#### *3.5.1 Demographic Survey*

The survey was designed to identify their linguistic and educational backgrounds for analysis. It included questions to collect demographic information from participants, such as their gender, age, educational level, Korean learning experiences, the occurrence of travelling to Korea, their communication with native Koreans, their feelings, and perceptions when speaking to L1 Korean speakers.

#### *3.5.2 Wordcloud*

*Wordcloud* is a collection of words depicted in different sizes generated by Mentimeter, a digital interactive presentation software. Participants would be asked to enter a maximum of three words to describe their perception of Korean verbal and non-verbal communication, and the words would be immediately entered into the system to create the Wordcloud. The larger the word, the more often it was written by participants.

#### *3.5.3 K-film Roundtable Discussion (Think-Aloud Protocol)*

The *Roundtable discussion* consists of five sections where K-films are screened and discussed. One clip was from *The New World* (신세계) and the other four clips were from *Train to Busan*  (부산행). In each session, participants would watch the clip (around 1-2 minutes) attentively, then engage in a discussion where they would share their thoughts on the verbal and non-verbal aspects of politeness in the clips.

#### *3.5.4 Intercultural Awareness Drawing*

The *Drawing* took place after the K-film roundtable discussion, where students got exposed to a series of Korean film clips. Participants were given a worksheet of 'Intercultural Awareness Drawing' that they used to express the cultural, pragmatic, and linguistic differences of non-verbal politeness between the language and culture of Korean and those of theirs visually. Participants were also asked to provide a brief explanation for each of the drawings they provided. In addition, they could alternatively choose to provide a dialogue for the interlocutors in their drawing to depict the differences.

#### *3.5.5 Intercultural Awareness Reflective Journal*

The *Reflective Journal* also took place digitally, where students would need to type digital input to critically reflect on their intercultural awareness of politeness in Korean's non-verbal communication. Participants were asked to type in at least 50 words to reflect on how Korean films (and dramas) had shaped their understanding of 'politeness' in the Korean language, and how that was similar to and different from their own language and culture.

#### *3.6 Procedures*

Before the study commenced, ethical approval was obtained from the university. The second author, also an instructor of the Korean lessons, briefed the other instructors about the research purpose and procedures. The participants were involved in four activities: survey completion, interactive brainstorming, a K-film roundtable discussion, and completing an intercultural awareness reflective journal. Each session lasted 50 minutes, and there were three sessions total, corresponding to the elementary, intermediate, and advanced levels.

Firstly, participants were instructed to scan a QR code and fill out a Google Form with their demographic information, which took about five minutes. Next, during the brainstorming session, they entered a PIN into the Mentimeter platform. The participants were asked to contribute a single word describing their views on verbal and non-verbal communication in Korean. Their responses were aggregated into a Wordcloud, with more frequently mentioned words appearing larger. They were also asked whether they perceived Koreans as polite and, if so, in what ways. In the subsequent film analysis, participants watched clips from *The New World* (신세계) and *Train to Busan* (부산행). They discussed their impressions of the relationships between the characters,

the level of politeness, non-verbal communication, and the broader implications at individual, organizational, and societal levels. The researchers provided question prompts on paper for each participant to facilitate the discussion. The researcher observed the discussions with minimal involvement to avoid influencing the outcomes, only intervening if participants struggled to generate discussion. Finally, in the reflective journal session, participants were asked to illustrate how non-verbal politeness in their own cultures compared to Korean practices. They then digitally recorded their reflections on Padlet, allowing them to privately write and share their thoughts with classmates.

#### *3.7 Data Analysis*

The primary data sources from the workshop included texts from participants' individual responses, audio recordings of group discussions, and participant reflections. All audio was transcribed for thematic analysis. We utilized Braun and Clarke's (2006) method of thematic analysis, which involves examining both semantic (explicit meanings) and latent (underlying ideas or assumptions) content. Two researchers meticulously read and coded the transcripts multiple times, refining and revising the codes throughout the process to enhance the analysis.

Students' drawings were also scanned and converted into digital files, each accompanied by a description contextualizing the artwork. This approach not only provides a rich, nuanced view of participants' perceptions and learning processes but also highlights the expressive and personally relevant aspects of their contributions. Analysing these visual representations allows for the uncovering of subtler messages and more obscure realities than can typically be achieved through solely text-based research methods.

#### **4 Results**

#### *4.1 Ways of Perceiving Korean's Verbal and Non-Verbal Communication*

Participants' perceptions of verbal communication in Korean are characterized by various descriptors reflecting the honorific nature of the language. Terms such as 'respectful', 'polite', 'hierarchical', 'traditional', 'self-conscious', 'distant', and 'structured' were commonly used. In contrast, their views on non-verbal communication elicited both negative and positive connotations. Negative terms included 'troublesome', 'cumbersome', 'restricting', 'controlled', and 'conservative', while positive descriptors were 'expressive' and 'pleasant'. Participants supplemented these single-word descriptions with examples illustrating their understanding of politeness through non-verbal pragmatics.

One participant provided a compelling comparison between her native Bulgarian culture and Korean culture. In Bulgaria, loud speech and shouting are typically perceived as rude or indicative of anger. However, she observed that in Korea, louder voices and more pronounced gestures are often used expressively in a broader range of situations. This observation led her to suggest that there may be numerous culturally ingrained methods of expressing non-verbal politeness in Korean, indicating that these louder expressions are not necessarily impolite or negative (see Fig. 1).

**Fig. 1** Cultural differences of non-verbal politeness in terms of verbal message delivery

Participants identified several non-verbal features commonly used in Korean communication, some of which are universal across different languages and cultures. However, they specifically noted how certain non-verbal behaviours are uniquely tied to Korean culture and history, examples including 'bowing with their head often' and 'eye contact changes depending on hierarchy'. Several participants suggested that the non-verbal patterns could convey an underlying and implicit social order in the Korean society. These behaviours were seen as reflecting the implicit social order within Korean society.

A common observation was the contrast between the Korean practice of bowing as a greeting and other cultural greetings, such as a kiss during a first meeting, as practiced in one participant's culture (see Figures 2 and 3). Additionally, one participant noted that younger individuals often use non-verbal gestures like bowing to request favours or make entreaties, adhering to the traditional hierarchy in Korean society, rather than engaging in verbal disputes or direct requests.

I think there is a culture of being mindful of the people around you and following unspoken rules e.g., etiquette in public transport. Also, I have an image that the staff are very polite to customers.

(Participant 5, intermediate)

Deferential towards superiors or elders, bowing, not arguing with them, becoming a 'yes man' agreeing with ideas and not opposing.

(Participant 9, intermediate)

**Fig. 2** To bow or not to bow: Greeting difference

Several participants highlighted the cultural significance of handshakes in Korean interactions. One participant observed that Koreans often support their dominant hand with the other when shaking hands, a gesture intended to convey politeness. Another noted the common use of a two-handed handshake among Koreans, which is seen as a sign of respect, akin to handling a precious object. This contrasts with the practices in their own cultures, where handshakes, even during first meetings, tend to be less formal and aim to establish equality between the parties. Additionally, a participant pointed out an intriguing cultural nuance: in their culture, the more relaxed one's hands are during a handshake, the more experience or confidence it suggests. For example, a participant noted, 'Casualness can be used to show dominance in a meeting — 'I'm more relaxed than you, so clearly I have more experience' (see Fig. 3).

**Fig. 3** Handshaking - one hand or two hands?

#### *4.2 The Role of Competence in Perceiving Non-Verbal Politeness*

#### *4.2.1 Advanced learners: Correlating and Criticising*

The advanced students in our study were able to interpret non-verbal communication not only by means of description and identification on the surface level but also by providing more information

and being able to understand the cultural connotation behind such actions. In *The New World*, one of the participants could clearly describe all the non-verbal features of the bowing scene and the significance of using non-linguistic representation in presenting politeness.

There's like all different levels of the hierarchy...so when they were walking in the triangle, clearly the most front bows. This is the important guy in this selection. The fact that as they walk in, everyone bows, puts all three of them above...everybody else, and then the way the important guy in front bows to the man in the corridor and roughly dismisses them and puts him above everybody else. It's just from without using any words to show seniority. (Participant 14, advanced)

As compared to other levels, advanced participants demonstrated a greater understanding of how Koreans process physical and mental information when they encounter situations in which they may or may not need to show politeness. For example, in *Train to Busan*, a participant gave an example of a junior employee feeling higher up when he came in, showing a sense of excitement. Nevertheless, he reminded himself to be respectful by lowering his eyes.

The guy with a brown suit is harmless, respectful almost like a soldier at rest...sort of position at all times. Even he came in and was a bit higher up because he was excited about something and then Gongyoo interrupts him, and he lowers his arm and also lowers his eyes. Like, yeah like, remembering to be respectful.

(Participant 15, advanced)

Additionally, the participants were able to identify non-verbal politeness linked to societal norms, such as submissiveness, in which juniors or younger people would trivialise themselves in order to show respect to the seniors and not be blamed or judged. Using the metaphor of a dog, one participant illustrated how avoiding eye contact would demonstrate a hierarchical relationship and show respect to the seniors.

Like societal submissiveness...You are more important than me, I will make myself as small...and make sure that I'm not looking at you, almost the way that dogs don't look each other in the eyes. Yes, like okay, you're clearly the one who's in charge. I just take a step back.

(Participant 14, advanced)

Like the guy standing is trying to keep himself like making an effort to keep himself composed. Whereas the other guy (senior) is tired but he's not trying to hide his tiredness… like annoying. He is just like letting go…without expecting any negative repercussions or frustration.

(Participant 15, advanced)

The advanced students were distinguished from the other levels of students by being able to describe how the characters in the movie use different intonations while speaking. Some participants could perceive that the senior was talking to the 'more senior' person on the phone, presumably his boss, and he raised his voice to show confidence and reassurance in order to please the boss.

I guess he is talking to his boss. He's actually raised the tone of his voice as well. Rather than speaking like oh, I am really frustrated at this. He's like 'Ohh don't worry, I'll get this sorted for you and I'll make sure everything is done.'

(Participant 12, advanced)

Another participant noticed that the senior was slightly nodding his head on the phone as if his boss was watching and staring at him. It could be seen that advanced students had an indepth understanding of non-verbal communication in Korean by linking their own understanding of the Korean language and culture to the scenario happening in the Korean film clips.

#### 4.2.2 *Intermediate learners: Predicting and Inferring*

Intermediate students showed the ability to connect their understanding with Korean drama, and most of their understanding comes from Korean drama, which shaped their understanding of nonverbal politeness. In *The New World*, some participants were able to identify the main character as the senior with other characters being the juniors who needed to bow. They also explained the necessity of such non-verbal behaviour with reference to contextual factors. For example, given that the characters were in an official space, Koreans should bow to their superiors, a form of social convention or norm which is universally applied to all the official settings in the Korean context.

The guy…. I don't know his name, but from Squid Game... when he bowed and could tell that when he bowed, you could tell that the guy at the far end of the corridor was superior but the people that were walking, perhaps like a step behind him, seem to be his junior. (Participant 7, intermediate)

It seems they are in an official space, maybe workspace, maybe the environment that like compels them to act this way.

(Participant 8, intermediate)

For *Train to Busan*, some participants explained the non-verbal communication between the speakers in detail and noted that some patterns appeared polite, while others did not. For example, a participant could notice the employee came in, but the boss was still stressed and ignored the employee. The eye contact between the two speakers was also demonstrated in their example, as the junior's eyes immediately dropped when he was interrupted, and he also looked around to avoid someone's eyes in front of him. In this study, intermediate students were able to convey the hierarchy of speakers—one being higher status and one being lower status. An interesting phenomenon noted by the intermediate students was the recognition of the male character's real name. As the male character (Gongyoo) was famous across K-wave fandom, participants would normally be able to name one of the speakers in the video clips with his real name—Gongyoo.

Did the young man come in and he did something that annoyed him? Because it seemed like at first, he was looking in Gongyoo's direction. But then the moment Gongyoo looked back, and he did this with his hands. And then he looked down as well. He always needs to lower his face.

(Participant 9, intermediate)

Some intermediate participants were aware of the non-verbal behaviours common in normal interactions between Koreans in daily life. They could express the use of non-verbal acts such as bowing as a way to show respect and gratitude to older people or senior in the Korean culture. They also believed that it could be acceptable that the younger generation can use nonverbal impolite behaviours to each other, despite having significant differences in age or title. They believed that it would not be acceptable among the older generation as it could cause a serious, negative consequence from the senior negatively evaluating their disrespectful behaviour. Another participant also expressed the importance of the scriptwriter's age—the older the writer is, the more 'polite' the K-drama/K-film is.

I think also depends on the writer though, because if the writer is someone who's quite old, they'll probably stick to the very hierarchical level in the setting of the drama.

(Participant 12, intermediate)

In addition, some intermediate participants predicted that the pattern of non-verbal features in Korean interaction expressing politeness might have changed even though some K-films were very popular years ago. It was mentioned by a participant that the film (*Train to Busan*) was six years old, and the interactional pattern might have changed considerably since then.

#### *4.2.3 Elementary students: Describing and Identifying*

The results of the Roundtable discussion revealed that elementary students would find it very hard and challenging to interpret non-verbal behaviours and their underlying meanings. It was possible for them to discern some patterns that were universal across cultures, such as bowing to convey politeness, but they were unable to interpret their perceptions or respond to why particular gestures or patterns were used in Korean dramas and films. During the discussion session of *The New World*, one participant only noticed that the stare of the boss indicated that he was important and would need respect. However, he could not provide any further personal explanation about why the other people would look at the '*Squid Game* guy'.

I think the 'Squid Game' guy looks important, but the other guy looks more important. Everyone was looking at him as well and they were waiting for them to do it.

(Participant 1, elementary)

For another K-film, *Train to Busan* (부산행), participants found it more familiar because most of them had watched it before. Having already mastered some basic honorific structures in Korean (- *yo*), they were able to detect it in the video clips, which led them to note that a girl on the train attempted to speak formally to the older man. However, they could not provide further explanation of the reason behind the non-verbal gestures in the K-film. This is possibly due to their lack of understanding of pragmatic usage and cultural hierarchy in the Korean language system.

The girl was asking the guy formally, but the older guy just because he's older, he just speaks informally to her.

(Participant 1, elementary)

It seems she can't argue with him because he is older... and she wants the older guy to try and help.

(Participant 2, elementary)

#### *4.3 The role of the Korean Wave in understanding politeness in the Korean language*

Participants reported that Korean dramas provided them with sources to learn about different styles of respect. Being a fan of Korean dramas and films provided them with an opportunity to see the non-verbal honorific system used appropriately in 'real life'. Some participants also demonstrated a thorough understanding of how seniority or the relationship between people in Korean society could be predicted by non-verbal behaviour. Using K-dramas and K-films as examples, they mentioned that they had seen Koreans making an effort to bow when greeting other people, without the need for verbal expressions to express their respect. Participants could learn from Korean dramas that verbal and non-verbal aspects can have an intertwining and influential role in politeness that might have something to do with age or organisation.

I feel like watching Korean dramas has made me realise that although hierarchy can be strict in terms of body language and speech, Korean people can be casual and relaxed when they are close or familiar with each other. Age and status in a group or organisation are big factors.

(Participant 8, intermediate)

Non-verbal pragmatic behaviour also provided instrumental values for participants because the verbal pragmatic features in the Korean language can be very complicated, especially for beginners. The learning of the honorific system from the Korean dramas and movies can also be understood easily, as it provides a backdrop to demonstrate the use of verbal and non-verbal pragmatic usage in an authentic context to help break down the complex hierarchical relationships underlying the Korean language system. As an example, one participant commented that the nonverbal perspective displayed in Korean dramas and films illustrates the interrelationship between speakers, social status, and the environment. One participant's drawing depicted that he learnt when Koreans drink, the younger person at the table tends to turn away to show respect (see Figure 4).

The fact one usually needs to wait till the end of a sentence to understand what the speaker wants to say, shows a deeply ingrained structure of politeness. Things like bowing, degree of bowing, and staying humble through body language are great examples of the levels of respect. The variation in politeness endings in speech not only reflects not only the

environment of the speakers, but also their social status and relationship. It's very interesting about the Korean language that one can understand so many things about people just by the way they speak and behave during a dialogue.

(Participant 3, elementary)

**Fig. 4** Learning the drinking culture from Korean drama

Korean dramas and films also serve as a virtual immersion for learners to acquire both the language and culture of the Korean language. This approach allows learners to gain an understanding of authentic daily interactions among Koreans, even when they reside in countries where they are not surrounded by Korean speakers. In this way, learners feel that even if they cannot immerse themselves in the culture by travelling to or living in Korea, they can still experience the use of non-verbal expressions when interacting with superiors or elders. One participant drew parallels between a scene from a Korean drama and his perceptions of the formality associated with a Korean CEO or a *Goryeo*-era king. In addition, Korean dramas and films also inspired learners to adapt the honorific standard to their own culture, imagining encounters with the Queen and considering the non-verbal expressions they would use to demonstrate politeness.

K-dramas and movies allow me to see different levels of respect that I may not use. For example, I don't have very close Korean friends to use *banmal* with, so watching a k-drama, however, allows me to see when and how it is appropriately used. The same for formal situations, I will probably never meet a Korean CEO or a *Goryeo*-style King, but k-dramas allow me to see how such interactions would unfold, including the appropriate language and the deference required in my body language. English speakers don't usually use deferential or polite body language unless it is a very formal situation, such as meeting the Queen, as it can be seen as rude or snobby. Casualness is more common, even with people older than you.

(Participant 7, intermediate)

Furthermore, one participant stated that while her first language also has polite ways to address people, Korean dramas and films demonstrated more vividly the non-verbal aspect of Korean politeness with a much greater sense of hierarchy and respect than their own languages. The non-verbal aspects of polite expression may be universal across languages and cultures. For example, bowing expresses politeness in many languages and cultures, but Korean dramas and

films provided Korean language learners with a broader understanding of how these non-verbal behaviours in Korean are similar or different from their own.

Koreans use body language such as nodding and bowing to show their respect to elders and higher ups. Koreans also use polite words to elders/higher-ups. This doesn't exist/isn't as common within the English language including body language.

(Participant 13, advanced)

Germans are generally known to be very polite as well… we also have a polite form to address someone, but still from what I've seen from K-dramas, body language plays a much more important part in Korea. It's not archaic to use these symbols of respect and hierarchy there.

(Participant 9, intermediate)

That being said, participants explained that acquiring Korean politeness from non-verbal aspects enhanced their cultural sensitivity toward their origins and the Korean language because a 'polite act' in one language can be an 'impolite behaviour' in another. One participant, for instance, expressed that although avoiding eye contact may seem like a way to express politeness nonverbally, they noted it could in fact create impoliteness in their culture because avoiding eye contact could appear as being impolite or rude when listening to someone. It was also demonstrated in a participant's drawing that, in the English culture, it is natural and often expected to maintain eye contact when talking with other people, especially in a professional setting. While in Korean non-verbal interaction, the participants reported that in order to show respect for elders and seniors in social situations, bowing was expected and there is a tendency for interlocutors not to look into one another's eyes in professional settings (see Figure 5).

Politeness in English manifests to a lesser degree in physical habits than in Korean. So we don't bow, or avoid eye contact with superiors. In fact, avoiding eye contact can be considered rude or inattentive.

(Participant 3, elementary)

**Fig. 5** Being polite - with or without eye contact?

Clearly, the use of body gestures and facial expressions in the non-verbal expressions of the Korean language is ubiquitous and that could be of instrumental value for the 'members' of the K-wave fandom in understanding the Korean language and culture. In particular, from our results, the explicitness of politeness expressed non-verbally from K-dramas and K-films has also prompted participants to consider how appropriate it could be if a similar honorific system was to be applied in their own language and culture.

#### **5 Discussion**

#### *5.1 Interactional Competence through K-Wave: Supplementing Textbook-Based*

#### *Learning*

As globalization advances and technology evolves, the aspiration to become bilingual or multilingual intensifies. This is equally true for Korean language learners, many of whom are driven by their passion for K-wave culture. Despite their enthusiasm and desire to become 'true fans' by mastering the language, these learners often rely on textbooks. Scholars such as Brown (2010), Kiaer (2017b), and Kiaer et al. (2022) have criticized these textbooks as inauthentic, leaving students feeling ill-equipped to navigate real-life conversations about Korean language and culture. Typically, textbooks teach the 'polite form' of Korean, which can create awkward social dynamics when learners use overly formal language with younger Koreans, potentially making them feel uncomfortable. Moreover, genuine communication encompasses both verbal and nonverbal elements that contribute to perceived politeness, challenging students to acquire these complex interpersonal skills through textbooks alone. This situation raises concerns about students' interactional competence, which is essential for effectively adapting linguistic and interactional skills to meet immediate communicative needs.

Interactional competence, as defined by Kecskes (2019), is 'the ability to deploy interactional resources (turn-taking, repair, boundaries, speech acts, etc.) using available linguistic resources to fulfill communicative intentions in actual situational contexts' (p. 69). Young (2011) further explains that interactions are influenced by social, institutional, political, and historical factors. The honorific system in Korean, often emphasized in textbooks, poses additional challenges for learners. Korean dramas and films, however, offer a valuable alternative. They provide learners with semi-authentic settings where they can observe how characters navigate identity resources, linguistic resources, and interactional resources—crucial components of interactional competence, which is distinct from communicative competence often targeted by language curricula. Korean media thus serves as a practical supplement to textbook learning, plunging learners 'into the bathwater' of language use, as will be further discussed in the following section. This approach helps bridge the gap between classroom learning and real-world application, enhancing learners' overall ability to communicate effectively in Korean.

#### *5.2 Virtually Bathing in Language*

Politeness is very difficult to teach, but it is essential for learning the Korean language properly. In the case of Korean, it clearly goes beyond words to multimodal components too, as it does in so many Asian and other languages (Kiaer, 2023). As such, it is very difficult to teach this very crucial aspect of the Korean language because it does not fit neatly into standard textbooks and

classroom language teaching practices. However, as K-fandoms have grown and been able to access more and more K-films and K-dramas over time, awareness of Korean politeness has grown. The huge-scale success of *Squid Game* (2021) and *Parasite* (2019) have allowed learners of the Korean language to glimpse how Korean politeness functions. Internationally successful artefacts such as these can also act as 'gateway' content, leading fans to be interested in watching other Korean content, so they do not feel that they are forcing themselves to watch something just because it can help with their language acquisition.

K-films and K-dramas are of course works of fiction. Their significance, however, should not be downplayed. They allow viewers to see how actors, directors, and scriptwriters believe a character *should* behave depending on whether they are trying to be either polite, rude, or intimate (Alvarez-Pereyre, 2011). In seeing these depictions, viewers begin to learn conventional Korean politeness in an embodied way. This means that they learn subconsciously, picking up the significance of different multimodal signs without being explicitly told by a teacher or textbook. Simultaneously, they can learn which multimodal signs should be used in which context. Despite the practical and narrative limitations that generally prevent the full portrayal of activity frames, viewers of Korean media are nevertheless exposed to segments of interactions amongst people of various relative statuses K-viewers nevertheless see at least portions of different interactions between people of a variety of different relative statuses. They see a range of non-verbal behaviours carried out by characters of a range of different genders, age groups, occupations, and personality types. Thus, K-fans can learn much from K-content with very minimal effort. The biggest K-content fans have watched hundreds, if not thousands, of hours of content in the Korean language. As such, they may gain a nuanced and complex understanding of all the non-verbal and multimodal aspects of Korean communication. How this translates into actual production of language, though, remains an open question.

While good for dealing with matters of lexis and syntax, classroom settings deal with language learning in a manner isolated from reality. K-dramas allow for learning in-situ with clear contextual subtext. They show dynamic interactions that allow learners to acquire Korean language skills in a different way. Thus, media-mediated language learning overcomes the barriers of classroom and textbook learning. It uniquely compliments classroom and/or independent learning, providing a taste of what it might be like to actually interact in reality. We would argue that mediamediated language learning can be a beneficial way to acquire intercultural awareness of pragmatic differences.

Media-mediated language learning may also be able to increase foreign language enjoyment (FLE) and thus lessen foreign language anxiety (FLA). Watching K-dramas and Kfilms provides a way for Korean language learners to be exposed to Korean pragmatics without requiring them to respond. There is almost no other situation in which language learners can just observe people of different relational dynamics communicating in such a wide variety of contexts. The lack of pressure on the viewer to make any response means that they can acquire pragmatic awareness without having to worry about offending anyone. What is more, K-dramas and films are entertaining and exist in a wide variety of genres, meaning that viewers can choose content to watch for genuine entertainment. As a result, they acquire knowledge of Korean verbal and nonverbal language without feeling it like a chore. With this gained pragmatic competency and familiarity with *in-situ* Korean interactions, language learners lessen the burden upon themselves when speaking Korean. We believe further research is required, but this approach may probably diminish their potential for FLA.

#### **6 Pedagogical Implications**

#### *6.1 Curriculum Innovation: Media-Mediated Pragmatic Teaching and Learning*

Curriculums matter. As we have seen in this paper, language learners of different levels have different needs, and we need to think about how to accommodate those needs effectively. Korean dramas and films being a strong motivating factor for learners to acquire the Korean language, can also serve as a key role in the curriculum innovation of Korean language teaching and learning.

For the elementary level, our elementary participants showed that our curriculums need to incorporate pragmatics because they were unable to interpret the meanings of verbal and nonverbal signs in the K-dramas that they were shown. As such, a focus on pragmatics in pedagogical settings could help better support their language acquisition. Korean dramas could be a powerful tool to be employed in elementary classrooms, but teachers should be aware of the instruction that they are providing to students. Teachers should not assume that students will be able to understand the underlying pragmatic meanings because students coming from distinct backgrounds may not have possessed adequate pragmatic competence and awareness. Accompanying the video clips from Korean dramas and films, explicit instruction of the meanings conveyed is encouraged in Korean teachers' pragmatic teaching. Our intermediate-level participants showed a good grasp of the non-verbal elements of Korean, but they could benefit from further developing their pragmatic, semantic, and syntactic awareness.

This study focuses on the non-verbal aspect of politeness in Korean pragmatics, but both verbal and non-verbal communication are often interrelated in all languages: we could not only focus on either verbal or non-verbal aspects. Possessing an adequate grip of honorific systems in terms of syntactic, semantic and pragmatic capability, students should be provided opportunities to practise an authentic engagement with the Korean language by incorporating the non-verbal elements in pragmatics. For example, students can engage in role plays where they have to perform the use of the honorific appropriately not only in the verbal aspect but also the non-verbal aspect to display a thorough understanding of how 'politeness' actually works in authentic communication. An interesting pedagogical approach will be incorporating a Korean film or drama into the role-play teaching by giving each student a character. Students will receive the character card with the person's demographic information including age, job title. The teacher will invite students to walk around the classroom and stop after a minute, and they have to find a partner who is close to him/her. They then have to show the card to each other and decide what honorific style they should adopt in their communication. Students are encouraged to use not only the polite or plain form of honorific in their communication, such as adding '-yo', but also employ the appropriate non-verbal acts in their communication. Meanwhile, teachers should provide feedback to students on both verbal and non-verbal aspects to students' practice and they should also maximise the possibility of using non-verbal features, making students' learning meaningful and authentic.

Our advanced-level participants gained a good pragmatic understanding. In response, teachers could facilitate more spoken practice with other advanced learners and L1 Korean speakers so that their students would be able to employ pragmatics spontaneously and with ease. It should be noted that our aim is not to promote 'native-speakerism'—as language belongs to anyone. Students should be motivated to use suitable non-verbal features in their conversation, but teachers should not judge their usage only on the basis of whether or not 'native' speakers will use it in their daily communication. Teachers should provide feedback based on context appropriateness so students' verbal and non-verbal acts will create effective communication and

will not hinder the listener's understanding. Meanwhile, Korean drama and film will serve as a supplement for students to further explore and interpret the rationale of using non-verbal behaviours in relation to Korea's history, culture, and tradition.

#### *6.2 Education Equity: Affordable and Enjoyable Language Learning Experiences*

As teaching moved toward a focus on communicative competence during the 1970s and 1980s, it was recognised that immersion may be the best way to acquire a language. As Snow et al. assert, 'language is learned most effectively for communication in meaningful, purposeful social and academic contexts' (1989, p. 202). Immersion contexts aligned with natural approaches to language acquisition, and language learning can be more effective when it was integrated with content. Lo (2022, 2024a, 2024b) also found that motivation was found to be the most significant factor when content and language are learned simultaneously.

It is not always so easy to travel, as we have learnt much from the COVID-19 pandemic. Language immersion requires money and time, too. For ecological reasons, flying across the world for language acquisition may not be some people's first choice. These issues lie at the heart of why media-mediated language learning can be useful. It moves beyond input-poor environments, such as classrooms, and introduces embodied learning at a very little cost, time, or effort.

With the rapid development of Virtual Reality (VR) and Artificial Intelligence (AI), there is also the possibility of integrating media-mediated language learning into them. For example, students may create their own avatars and may communicate in the metaverse (Kiaer & Lo, 2025). Teachers may also set the context as in the Korean dramas or films and ask students to come up with a solution to tackle the conundrum that appeared in the dramas or films, using the appropriate verbal and non-verbal features. An authentic task engagement and an enjoyable learning environment would allow for further immersion and possibly even greater reduction of FLA, which has been found to have a negative correlation with learning performance—the more anxious students are, the worse the learning performance (Zhang, 2019).

#### **7 Conclusion**

This paper has looked at how different proficiency levels of K-wave language learners displayed an increased awareness of non-verbal politeness cues and were more adept at explaining politeness. Through a workshop with Korean learners of a range of levels, we assessed their awareness of Korean politeness, and its non-verbal aspects, using scenes from two Korean films. Our findings demonstrated that their awareness of Korean pragmatics was not high at the beginner level, especially at the beginner level. At the intermediate level, their understanding of more noticeable non-verbal behaviours was seen, but it still lacked nuance. Our advanced participants had a thorough understanding, but this raises the question as to whether the knowledge of non-verbal politeness is an advanced feature of the language. Considering how crucial pragmatics and politeness are to successful communication in Korean, students should be learning about these non-verbal features from the very start. Teachers may protest that these features are too confusing and complex to learn as a beginner, but this paper has highlighted that watching K-dramas and Kfilms can provide insight into pragmatic learning without learners having to make any special effort. Media-mediated language learning allows language learners of any level to gain familiarity with Korean politeness, requiring only minimal effort. Overall, greater integration of media into language learning syllabuses could prove greatly beneficial, especially for Asian languages.

#### **8 Future Research**

This study has preliminarily proven that watching media is beneficial for the acquisition of intercultural and pragmatic competence. To further verify these findings, it would be necessary to undertake an intervention study with a control group. For example, researchers may divide participants with the same proficiency into an experimental group and a control group. Before the study is conducted, a pre-test set including a language proficiency test, an intercultural awareness questionnaire, and a scenario-based pragmatic competence assessment (e.g. What would you do in this situation?) can be taken by participants. The experimental group will then receive Korean dramas or films as input to their pragmatic learning. A post-test can be administered to see if there may be differences within and between groups to further reveal how effective Korean drama and films could facilitate students' intercultural and pragmatic competence. In addition, there would also be a need to look at other languages and their specific pragmatics. Outside of this, a study into other effects of media content on language acquisition could also be useful. Media may not only teach pragmatic skills and cultural competence, but other skills and capabilities too. As Kiaer and Lo (forthcoming) note, students who are engaged in the K-wave fandom would find themselves more joyous in language learning, thus sustaining their language learning motivation. A longitudinal study tracking and tracing students' intercultural and pragmatic development supplemented by media content is strongly recommended.

#### **References**


Hong, J. O. (2009*). A discourse approach to Korean politeness: Towards a culture-specific Confucian framework* [Unpublished doctoral dissertation]. Nottingham Trent University.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Negotiating Age, Epistemic Stance, and Category Memberships in Korean Talk Shows**

#### **Mary Shin Kim and Jaehyun Jo<sup>1</sup>**

**Abstract** This chapter examines an integral part of Korean politeness—age. Prior research has focused on investigating age as a key determining factor that influences speakers' choice of speech styles, address terms, and word choices. On the other hand, this study explores what speakers do with age and how age is relevant to them at the moment of their interaction. The focus of the study is on media talk as it reveals how age is represented, negotiated, and utilized by social members in public discourse. The study identifies a categorial practice related to age by utilizing Stokoe's (2012) five guiding principles for membership categorization analysis (MCA). A collection of data segments from various talk show interviews shows that the speakers routinely evoke age while displaying an epistemic stance, such as in claiming or disclaiming their epistemic rights, to the matter being discussed. Moreover, speakers classify themselves and others into age categories (Stokoe, 2012; e.g., the old, the young, *acessi*) based on their epistemic status or rights. These categories are determined not by the speakers' chronological age, which is external to the interaction, but are rather spontaneously formulated in the local context of the interaction, which are subject to challenge, resistance, and negotiation by speakers. The study reveals how speakers use age as an interactional tool to negotiate their epistemic stance and category membership.

**Keywords** Age Category Membership Epistemic stance Epistemic status Identity Talk shows

M. S. Kim ()

e-mail: maryskim@hawaii.edu

J. Jo

Department of Asian Languages and Cultures, Rutgers University, New Brunswick, New Jersey, USA. e-mail: jj778@alc.rutgers.edu

Department of East Asian Languages and Literatures, University of Hawaiʻi at Mānoa, Honolulu, Hawaiʻi, USA.

#### **1 Introduction**

Age is an integral part of everyone's identity.<sup>2</sup> The age status between interlocutors is particularly important in Korean, an honorific language with a rich repertoire of detailed and elaborate honorific and politeness linguistic resources (Sohn, 1999). Depending on the speakers' relative age difference, social status, and relationship, they design their utterances with different speech styles, address terms, and word choices. Accordingly, if the relative age status is unclear or unknown, Korean speakers seek to clarify the much-needed demographic information (Cho & Jo, 2022), as exemplified in the question and answer sequence from a talk show interview (Excerpt 1).

The interaction involves a group of co-hosts and a team of four Korean national fencing members as guests. As soon as the talk starts, one of the hosts inquires who is the oldest guest (line 1). In response, the guests and another co-host verbally and nonverbally explain that they are seated by age order: from the oldest member seated closest to the hosts to the youngest member seated further away (lines 2–6 and Figure 1). This brief introduction clearly demonstrates how age is a crucial factor in determining the social and physical positioning of the participants (cf. Kjaer & Krummheuer, 2018). <sup>3</sup> Age overrides the guests' other possible identities,<sup>4</sup> such as being a world champion by professional rank, as shown in lines 8–11.

Excerpt 1. Who's the oldest?

```
01 H1: → 여기:는 누가 제일 (.) 형 큰 형,
          yeki:-nun nwuka ceyil(.)hyeng khun hyeng,
          HERE-TOP WHO:NOM MOST BROTHER BIG BROTHER
          'Here, who (is) the oldest, (.) the big brother?'
02 G1: → [%저요.]
          %ce-yo.
          ME-POL
          'Me.'
          %raises and keeps his hand up-->
```
<sup>2</sup> Raymond (2016a, 2016b) differentiated the 'identity status' (i.e. "more or less settled aspects of individuals' identities") and the 'identity stance' (i.e. "the moment-by-moment invocation of one or more of these identities") in talk-in-action. In this paper, we analyze identity stances that speakers select for themselves and attribute to others in connection with age.

<sup>3</sup> Kjær and Krummheuer (2018) discuss how the bodily positioning of healthcare professionals enables or limits the agency of the other participants.

<sup>4</sup> When the rank overrides age (for instance in the workplace), precursor phases are used which permit interlocutors to verbally license the 'oddity' (e.g., 'Since we are in the same rank, I will speak comfortably (despite our age difference)').


'In this (order)' g4 +swirls both hands to display their age order+ g1 ->%


```
09 잘 생겼는데, 펜싱도 [잘 하구:,]
          calsayngkye-ss-nuntey, pheynsing-to [cal ha-kwu:,]
          HANDSOME-PST-AND FENCING-ALSO WELL DO-AND 
          '(He's) the youngest. Wow, the youngest(.) is so 
          handsome, and also very good at fencing, and'
10 H2: [일 위- 일위시- 세계 랭킹 일위시죠?
           [il wi- il wi-si- seykyey layngkhing il wi-si-cyo?
           ONE RANK ONE RANK-SH WORLD RANKING ONE RANK-SH-COMM:POL
           '(You) are No. 1-, No. 1- No. 1 world ranking right?'
11 G4: 예.
          yey.
          'Yes.'
```
Upon specifying the relative age of speakers, age may be considered only as contextual information. However, as the subsequent interaction from the same interview with the fencing team members shows (Excerpt 2), age is constantly present in the interaction ("omnirelevance," Fitzgerald et al., 2009; following Sacks, 1995), particularly when it comes to the participants' epistemic domain. After introducing the fencing team members as guests, the host (H1) is curious to know how the members came to start fencing. According to the host, fencing is an uncommon sport that is not as accessible as other sports, such as kendo, which used to be common and accessible to the public (lines 1–9).

Excerpt 2. I am quite old.


[yaykiha-ki-]waynyamyen kulen(.)club-i iss-kena animyen, SAY-NML BECAUSE SUCH CLUB-NOM EXIST-OR IF.NOT '(cannot easily) say- Because it's not like they have such(fencing) club (at school),' 06 H1: 우리 어렸을 때 검도부는 있었어. wuli elye-ss-ul ttay kemtopwu-nun iss-ess-e. OUR YOUNG-PST-RL TIME KENDO.GROUP-TOP EXIST-PST-IE 'When we were young, there was a Kendo group (at school).' 07 H2: 아 그렇지. a kuleh-ci. OH BE.SO-COMM 'Oh, that's right.' 08 H1: 검도는:, 좀 흔해 가지구:, kemto-nu:n, com hunhay kacikwu:, KENDO-TOP A.LITTLE COMMON BECAUSE 'Because kendo was pretty common,' 09 쉽게 할 수 있었는데, >학원도 많았구.< swipkey hal swu iss-ess-nuntey,>hakwen-to manh-ass-kwu.< EASILY ABLE.TO.DO-PST-BUT SCHOOL-ALSO MANY-PST-AND '(we) could easily try and also there were many (kendo) schools.' 10 G1: → 모래시계 (.) 그 당시에 이정재 씨 때문에. molaysikyey (.) ku tangsiey icengcay ssi ttaymwuney SANDGLASS THAT AT.THAT.TIME NAME MR. BECAUSE.OF '*Sandglass* (.) at that time, because of Lee Jung Jae (on the drama)' 11 H3: 아:


13 (0.2) 14 G1: → 아- %저 나이 되게 많아요.% a- %ce nai toykey manh-a-yo.% OH I AGE VERY.MUCH MANY-IE-POL 'Oh- I am quite old.' %points at himself%

During the host's contrast between kendo and fencing, observe how Guest 1 (G1), the oldest member of the fencing team, adds his knowledge of the popularity of kendo at the time which the host refers to. G1 mentions Lee Jung Jae (line 10), an actor who practiced kendo in a popular Korean drama, "Sandglass" (aired in 1995), which led to the popularity of kendo. Rather than acknowledging or accepting G1's additional information, the host asks for confirmation on the guest's knowledge of the drama (line 12, '(You) know (the drama) *Sandglass*?'). Despite knowing the matter to bring up, G1 does not confirm his knowledge with a yes response. This request for confirmation is a negatively valenced question (Kim, 2021) which displays the host's doubt or disbelief that G1 knows the drama. The question is designed with the object particle ('sandglass'+*l*) (Kim & Kim, 2014) and accompanied by the host's head tilted movement (Figures 2-1 and 2-2), which all serve to display the host's doubt or disbelief toward G1's knowledge. Take note of how the guest responds to this question that challenges G1's knowledge level: he asserts that he is quite old (line 14). G1's response indicates that he believes the host does not think G1 is old enough to know the drama that aired in the mid 90's. Speakers routinely display what they know and how they know it by encoding evidentiality (e.g. Chafe & Nichols, 1986; Heritage, 2012a, 2012b; Kärkkäinen, 2003), the source of or evidence for their information (e.g., sensory experiences, hearsay, inference, etc.). Instead of disclosing the source of information, in Excerpt 2, the speaker discloses his age as a basis for warranting or justifying his epistemic status and rights.

This study is interested in investigating how age is evoked as a locally situated device or practice during the unfolding interaction, as seen in Excerpt 2. The paper examines what speakers do with age and how age is relevant to the speakers at the moment of their interaction. The study is particularly interested in studying media talk as it reveals how age is represented, negotiated, and utilized by social members in public discourse. The study will deploy conversation analysis (Sacks et al., 1974; Schegloff, 2007a), which will help us reveal what actions and activities participants perform with age through their turn designs and turn-taking and sequence organization. Additionally, the study utilizes membership categorization analysis (Sacks, 1995; 1972a, 1972b, 1979; Stokoe, 2012) to unveil how participants categorize each other and how these categorization practices are essential for understanding how identities are co-constructed in situ.

The findings of this study will first show how speakers routinely evoke or allude to age while displaying an epistemic stance, such as in claiming or disclaiming their epistemic rights to the matter being discussed. Age serves as an account or license for having access to certain knowledge or lack thereof. The study will further show how speakers classify themselves and others into age categories (e.g., the young, the old, *acessi*, grandpa) based on their epistemic status or rights regarding certain matters. The categories are determined not by the speakers' chronological age or stages in their lives, but by their knowledge or lack thereof of certain matters. These categories are spontaneously formulated in the local context of the interaction and

are subject to challenge, resistance, and negotiation by speakers. Overall, the study will reveal how speakers use age as an interactional tool to negotiate their epistemic stance and category membership.

#### **2 Data and Methodology**

This study utilizes conversation analysis (CA) and membership categorization analysis (MCA) to investigate participants' orientation to age as an interactional device for managing epistemics and category membership in public talk. In particular, the study follows Stokoe's (2012) five guiding principles to examine both categorization and sequential aspects of interactions, which derived from Sacks' and subsequent MCA work (Evans & Fitzgerald, 2017; Fitzgerald & Housley, 2015; Hester & Eglin, 1997; Housley & Fitzgerald, 2002; Schegloff, 2007b). Through such systematic analysis, we will demonstrate how and when participants locally orient to age or age group and position themselves and others into different membership categories.

Stokoe's (2012) five guiding principles for MCA (Modified):

1. *Collect* data across different sorts of settings and interactions depending on the focus of the study (i.e., diverse talk show interviews)

2. *Build* collections of explicit mentions of categories (e.g., *acessi* 'old man', *halapeci*  'grandfather', etc.); membership categorization devices (e.g., *nai* 'age') and category descriptions (e.g., the descriptions 'I'm quite old.' 'In my days, ….', 'You need to hang out with a young person.')

3. *Locate* the sequential position of each categorial instance with the ongoing interaction

4. *Analyze* the design and action orientation of the turn in which the category, device or resonant description appears.

5. *Look* for evidence that, and of how, recipients orient to the category, device or resonant description; for the interactional consequences of a category's use; for co-occurring component features of categorial formulations, and for the way speakers within and between turns build and resist categorizations.

Using these guidelines, we observed cases in which age was made relevant or crucial during media talk. Then, we gathered instances from various talk show interviews and identified a categorial practice.

The data collection comes from 11 different episodes of four popular Korean talk show programs: *You Quiz on the Block*, *Dolsing Fourmen* 'Return to Single Four Men', *Shinkwa Hamkkey* 'Along with the Gods', and *Jessiwauy showterview* 'Showterview with Jessi'. The participants are either public figures or experts from various fields. Their ages ranged from their twenties to their seventies. The data collection contains 23 instances where age is present in the local, sequential context of the interaction moment. The next section examines cases that are representative. Gail Jefferson's (2004) transcription conventions for CA are used to transcribe

talks. When relevant to data analysis, Mondada's (2018) multimodal transcription is used to represent embodied actions.

#### **3 Age and Epistemics**

Speakers' knowledge is the essential element in interaction. Speakers constantly position themselves and others with respect to what they know and what they don't know, and their rights to the matter being discussed ("territories of knowledge," Heritage, 2012a, 2012b; Heritage & Raymond, 2005). In this section, we will examine how age is intricately tied to speakers' epistemic rights in Korean media interaction. The following excerpt from a different talk show interview than Excerpts 1 and 2 also shows this point.

Excerpt 3 is taken from an interview between the two hosts and an actor who played a role in a renowned drama. In response to the main host's request to explain the storyline, the guest begins her explanation by stating that the drama is based on a famous novel written by a Korean-American writer.

Excerpt 3. How do you know such things?




While constructing her response, the guest struggles to find the appropriate version of the word 'Korean-American' in Korean as there are variations (lines 5–6). The guest eventually comes up with *chaemi tongpo* (line 8). However, as one of the co-hosts comes up with the word *chaemi kyopo* (line 9), the guest quickly accepts the host's choice of word (line 10). Interestingly, note how the guest immediately invokes age to explicate or excuse herself for the reason for her earlier choice of word (line 11, 'Back in my day, we used…').<sup>5</sup> The guest's swift adoption of the host's word choice, followed by her age-resonant explanation, suggests her awareness of the significance of age in word selection. It also implies that she perceives the host's choice as more in line with current trends.

<sup>5</sup> Similar to *wuli ttay-nun* 'In our/my days', there is another frequently used expression, *na ttay-nun maliya* 'back in my day'. This expression represents the way older generations like to talk about how it was back in the old days. Older generations are frequently ridiculed for using this expression and it has evolved into a pun *latte-nun maliya*.

Take note of how age continues to be an important factor in the ongoing interaction. When the co-host validates the guest's earlier word choice and discloses his knowledge of her word choice (lines 12–13), the guest displays surprise to his knowledge and asks for an explanation for his epistemic status (line 15, 'How do (you) know such things?'). The speaker's question here resonates with age. The guest, now in her 70s, has just shared her word choice from her youth period. She does not expect the 30-year-old host will be familiar with the word. In response, the co-host explains that his knowledge stems from exposure to such words from an early age through the media (line 17). Similar to Excerpt 2, the speaker's doubt about the recipient's knowledge arises from their age difference, and the recipient challenges this assumption by citing another age-related account. The epistemic domain and age are interwoven in the interaction between speakers. After uncovering the latest term with the help of the co-host, the guest returns to address the main host's initial question about the drama's storyline, employing that term (lines 22-23).

Excerpts 2 and 3 above have shown how speakers invoke age when claiming and justifying one's knowledge. The following excerpt illustrates how speakers attribute the speaker's lack of knowledge to age.

Excerpt 4. You need to hang out with young people.


The interaction takes place during a talk show interview between four co-hosts and a guest. While one of the co-hosts, H1, distributes drinks to everyone, he mentions that he made *avala* specifically for the guest (G1). H2, another co-host, displays that he is unaware of the meaning of the word *avala* (line 2), which is a newly coined short word for ice vanilla latte. In response, H1 and the guest tease him for not knowing the word (lines 3–4). Note how the guest invokes age in the immediately following turn as she suggests hanging out with younger people (line 5). This recommendation presupposes that H2 does not know the term *avala* due to his age and his epistemic status can be rectified through interaction with younger age groups. Age is oriented to as an essential factor in people's epistemic status.

Not knowing something can be treated as problematic by the participants, as seen in Excerpt 4. On the contrary, knowing certain matters can also be a problem as it is conventionally associated with certain age groups, such as old people. In Excerpt 5, the speakers actively deny their knowledge. As the four co-hosts wait for the guest of the day to arrive, they begin conversing about the pastries on their table for the guest. One of the co-hosts notices that the prepared pastries are old-style baked goods (line 7). This observation evolves into a comparison between pastries and black-and-white television, a relic of the past (lines 12–15). The discussion about the old-fashioned pastry then piques their curiosity about dating culture at bakeries in the past. Before asking H1 about the bakery dating culture (lines 19–20), take note of how H3 explicitly denies his and another co-host's knowledge due to age. There is no way that H2, the youngest co-host, would know about this, and even someone at H3's age would not understand the matter (lines 16–17). It is interesting that this prompts all other co-hosts to actively deny their knowledge of the past dating culture at bakeries (lines 21–22). Participants not only disclaim and disassociate themselves from the matter, but even display relief for not knowing such culture (line 23).

Excerpt 5. I don't know either, luckily.

05 우리는 알고 있지만 시청자분들은



'Woomin would absolutely have no idea (about this), and' 17 내 나이 때도 이해가 안 되는 게:, nay nai ttay-to ihay-ka an toy-nun key:, MY AGE TIME-ALSO UNDERSTANDING-NOM NOT BECOME-RL THING:NOM 'it is something which at even my age,(I) do not understand' 18 H4: (뭐가요?) (mwe-ka-yo?) WHAT-NOM-POL '(What?)' 19 H3: \*예전에는:, 그렇게 빵집에서 만나요? \*yeyceney-nu:n, kulehkey ppangcip-eyse manna-yo? IN.THE.PAST-TOP LIKE.THAT BAKERY-AT MEET:IE-POL 'In the past, why (did people) date at bakeries?' \*looks straight at H1 -> 20 [그게 뭐야:? \* kukey mwe-ya? THAT.THING-NOM WHAT-BE:IE 'What's the deal?' -> \* 21 H4: [아:: 저도 >아니에요 아니에요.< [a:: ce-to. >aniey-yo aniey-yo<. AH I-EITHER NOT.BE-POL NOT.BE-POL 'Ah: neither neither (do) I (know).' 22 H2: 저는 티비를 통해서 알았어. ce-nun thipi-lul thonghayse al-ass-e. I-TOP TELEVISION-ACC THROUGH KNOW-PST-IE 'I learned it through the television' 23 H3: 나도 아니야. =다행히. na-to ani-ya. = tahaynghi. I-EITHER NOT.BE-IE LUCKILY 'I don't know either.=Luckily.'

What is the reason for speakers ardently disclaiming this age-resonant knowledge and what are they collectively and collaboratively doing during the interaction? In the next section, we will discuss these questions further through the lens of MCA.

#### **4 Age and Membership Categorization**

We have observed speakers' invocation or orientation towards age or age group in regards to their own and others' epistemic status and rights. In this section, we will examine what speakers do regarding age and its relevant epistemic status and rights (knowledge and authority over the

age-resonant matter). Raymond and Heritage (2006) demonstrate that claims of knowledge are closely linked to membership categorization practices. Their study demonstrates how the identity of a speaker as a 'grandparent' is highly relevant and important in producing assessments about her grandchildren. The following excerpts will show that the way speakers assert or deny knowledge is intertwined with the process of categorizing them as either young or old.

Let's take a look at Excerpt 6 which shows the interaction that occurred after Excerpt 5. The discussion of old-fashioned bakery goods led participants to split into two groups based on their knowledge asymmetry (Raymond & Heritage, 2006), those who know the bakery dating culture and those who do not. This categorization is not just about their knowledge state but is also deeply linked to their age or age group. Host 3's mention about age in line 17 invokes the association of the knowledge of such dating culture with someone who is old or from an older generation. Interestingly, the participants tacitly invoke such membership categories without using any specific categorial labelling, such as 'old.' However, participants understand its association with being old and eagerly disassociate themselves from such a category. While they deny having knowledge of the matter, they identify the oldest host, Host 1, as someone who belongs to the category. Interestingly, without using any categorial term, they cast this category to H1 via speaker selection (Butler & Fitzgerald, 2010) through eye gazes and verbal cues (lines 19–20), explicitly asking him to explain the culture for the younger.

Excerpt 6. You're the storytelling grandpa.



However, speakers can challenge one another. Host 1 actively denies having such knowledge, resisting categorization (line 24). Others continue to challenge Host 1 and request him to provide an answer by identifying him as the sole member who can possibly tell them about the 'old' bakery dating culture (lines 25–26). Despite Host 1's continuing resistance (line 27), other speakers ultimately name the host as a storytelling grandpa, an age category, which justifies for his role and responsibility to answer the question (line 28). Just like how other hosts categorize Host 1 and assign him to carry out the ascribed activity, Host 1 recasts the category and expected activity to the guest, someone who has gone through the old times and can answer the age-resonant question (line 29). Although the guest is not yet present on the interview set, Host 1's gesture of pointing to the empty seat where the guest will be seated demonstrates his action of ascribing this category to the guest even before she appears on the show. Here, it is important to note that the disagreement between H1 and other co-hosts on H1's categorization is not a serious one. It is produced in the context of teasing and laughing contributing to the creation of humor (Okazawa, 2021). Despite H1's repeated resistance to categorization, he eventually engages in the category-bound activity ascribed to him, which is to tell what he knows about the past of bakery dating culture (lines 30–32). This activity resumes again when the guest finally appears on the interview set (data not shown here).

As previously shown, age plays an important role in how Korean speakers construct interactional identities, establish shared understandings of knowledge expectations, categorize members, and position themselves in relation to one another. What is interesting is that they tacitly categorize members into different age groups without referring to their physical age. The display of knowledge or lack thereof of certain information or knowledge situated in a particular time is a way for speakers to represent themselves. Similar to Excerpt 5, Excerpt 7 also shows how speakers portray themselves as young or younger groups. However, in this case, speakers ardently display their knowledge to represent themselves as a young person.

Excerpt 7. I know (the word). I am not an *acessi*.



This talk is derived from an interview with five singers in their 30s and four co-hosts in their 40s and 50s. When guests mention the latest newly coined words they use (lines 1–3), the co-hosts display different responses. H1 reveals his lack of knowledge by asking for the meaning of the newly coined words, which G2 finds surprising (lines 4–5). On the other hand, H4 selfselects to take the turn and discloses his knowledge (line 6). Interestingly, he immediately categorizes himself as not being an *acessi*, a reference term used for middle-aged men. He justifies that he is not a typical middle-aged man by claiming to have knowledge of the newly coined word. Although H1 and H4 are similar in physical age, H4 disassociates himself from H1, thereby indirectly categorizing H1 as an *acessi*. G1 is impressed by H4's explanation of the word meaning (lines 7–9), which indicates that speakers are highly valued or expected to keep up with new or up-to-date knowledge and information of younger or youth cultures. The association of the category 'young' with new trends was also observed in Excerpt 4, wherein one speaker recommended another speaker to hang out with young people to keep up with newly coined words, such as *avala*.

Above excerpts demonstrate that speakers do not simply categorize members based on their chronological age, but they create categories based on their knowledge of certain timeresonant matters locally in and through interaction. What is noteworthy is that the connection between the category (e.g., old, young) and the activity (e.g., bakery date, *avala*) is contingent on the interaction but not predetermined by a prior condition. Accordingly, due to these contingencies, these categories are subject to negotiation and challenge.

As shown in Excerpts 6 and 7, speakers seek to identify themselves as belonging to a certain category, such as a youthful group. Speakers may not always want to be a part of a

younger group. In Excerpt 8 (a continuation of Excerpt 2), when the hosts discuss the popularity of kendo back in their days, Guest 1 (G1) discloses that he shares the same knowledge associating himself with the host group. As 39-year-old, G1 deems himself to be quite old, at least old enough to share the same knowledge with the co-hosts (lines 14–16).

Excerpt 8. You're mimicking to be a big brother.



As previously mentioned, categorization can be subject to resistance. As the youngest person in the host group is 47, the hosts reject G1's categorization of himself as old (lines 17– 21). The hosts tease G1 as someone who still needs to be breastfed (line 23) and who is merely mimicking being a big brother (lines 25–26). Note that the negotiation and disagreement of the membership categorization between the speakers is produced jokingly in the context of teasing and laughing creating spontaneous humor in this excerpt again as previously shown in Excerpt 6.

It is striking to observe how consistently speakers in Korean media discourse categorize members into specific age or age-related categories based on their knowledge or lack thereof. Furthermore, speakers publicly display either embarrassment or pride regarding their age-related knowledge and affiliations.

#### **5 Conclusion**

In this chapter, we thoroughly investigate a crucial aspect of Korean politeness, which is age. While previous research has concentrated on age as a fundamental factor that shapes speakers' decisions regarding speech styles, address terms, and vocabulary selection, our study takes a distinct approach. We aim to uncover how speakers actively engage with the concept of age and its relevance in the very moments of their interactions. The media talk data examined in this paper are particularly good sources to observe the interplay between age, epistemics, and

identities in interactions. The data shows that Korean speakers frequently reference or indirectly refer to age while expressing their epistemic stance, whether by asserting or disavowing their right to knowledge about the subject under consideration. Age effectively acts as a justification or qualification for possessing or lacking access to specific information. We have discussed that there is also the interactional need to (re)categorize members and position themselves to one another based on their knowledge or lack thereof of certain age sensitive information at various moments of talk.

Prior studies have shown that speakers' claims of no knowledge or forgetfulness serve as means for various purposes: to avoid potential blame and responsibility (Drew, 1992; Hutchby, 2002), and to encourage another knowing speaker's participation in the conversation (Goodwin, 1987). Interestingly, our data show how speakers claim no knowledge to represent themselves as the young or younger group or generation (Excerpts 5–6, bakery dating culture). On the contrary, speakers can claim to be young or younger by actively claiming knowledge as well (Excerpt 7, I am not an *acessi*). Epistemics play a crucial role in shaping our interactional identities and our understanding of one another.

This paper contributes to our understanding of what speakers do with age and how age is relevant to the speakers' epistemic rights at the moment of their interaction in Korean. These findings inform us that the epistemic domain and age are tightly interwoven in the interaction among Korean speakers, which becomes procedurally consequential to the relevant interactional and membership categorization practices of the participants.

The results of this study highlight an interesting contrast in how politeness is practiced by Korean speakers when it comes to age. In Korean culture, there is a strong tradition of showing respect and politeness to individuals older than the speaker. However, when we look at media talk shows, a different pattern emerges. In this context, speakers openly express discomfort or tension when they are perceived as older during interactions, which is considered undesirable, similar to many other cultures (Jolanki, 2009). Future research could delve deeper into how speakers manage, reconcile, or navigate these dual aspects of politeness and age. One way we have observed in the current data was through teasing and laughing contributing to the creation of spontaneous humor (Excerpts 4-7). Further explored age discourse in the Korean content will confirm a collaborative process of establishing the asymmetric age dynamic, which reveals a tacit yet powerful cultural understanding by the speakers.

#### **Abbreviations**



#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Exploring Frames and Negativity Strategies in the News during an Election Campaign**

#### **Ji Young Kim<sup>1</sup>**

**Abstract** This study aims to investigate news frames and the use of negativity in the mainstream news coverage during a presidential election campaign. From a strategic communication perspective, the messages' cognitive and emotional (affective) attributes have been explored per time and political orientation of the media to discuss the implications on the campaign strategies.

**Keywords** Frames Negativity Impoliteness Emotion News Election

<sup>1</sup> J. Y. Kim ()

Communication Program, School of Communication & Information, College of Social Sciences, University of Hawaiʻi at Mānoa, Honolulu, USA e-mail: jkim22@hawaii.edu

#### **1 Introduction**

In political public relations, the role of practitioners' strategic communication has been emphasized in affecting news media stories, and the effects have been explored regarding the frames and emotions that refer to the cognitive and affective attributes (i.e., Curtin, 2009; Froehlich & Rüdiger, 2006; Kiousis et al., 2015; Turk, 1986). In the context of the 2022 presidential election in Korea, this study examined news frames and the negative remarks in the news coverage of the two mainstream newspapers.

Negativity has become a popular campaign strategy in today's election contexts due to its benefit of gaining media and public attention. Previous studies showed that negative messages obtained increased media attention and further engaged the public (Druckman, Kifer, & Parkin, 2010; Druckman & McDermott, 2008; Theilmann & Wilhite, 1998). This study aims to explore whether the media in Korea would cover negative and aggressive messages and what types of negative remarks are popular in the Korean context.

During the election campaign, Korea has experienced several political scandals, and people called this one of the most unfavorable elections in the country's history (Gallo, 2022). Candidates tried to associate their opponents with negative aspects, and campaign rhetoric focused on the disagreement and differences between the two major candidates on several social issues, including economy, gender, and foreign affairs. Through analyzing news stories, this study explored how the news media portrayed the campaign message strategies.

#### **2 Literature Review**

Political communication scholars have studied media portrayals of campaign messages to explore the effectiveness of their strategic communication efforts. For example, during an election campaign, practitioners provide several types of information subsidies—such as news releases or social media messages—to the journalists hoping their campaign messages are picked and portrayed in the media (i.e., Conway, Kenski, & Wang, 2015; Fountaine, 2017; Nadeau, Pétry, & Bélanger, 2010; Tedesco, 2001).

In the context of the 2000 presidential primaries, scholars analyzed the relationships between the candidates' news releases and network news stories regarding the issues and frames, and found strong correlations between them, particularly for Republican candidates (Tedesco, 2001). A content analysis study also showed that government communication practitioners successfully drew media attention to a certain issue (health care) during the 2000 Canadian federal election through the party's campaign materials (Nadeau, Pétry, & Bélanger, 2010).

Not only traditional news releases but also social media messages, such as Facebook or Twitter, also became important strategic communication tools for practitioners. A study shows that young women politicians used Twitter as a strategic communication platform during New Zealand's 2014 general election to increase their visibility and manage their campaign frames (Fountaine, 2017). An intermedia agenda-setting relationship was also found between the candidates' Twitter feeds and the nation's top newspaper stories during the 2012 presidential primary, meaning that candidates successfully set the traditional news' media agenda on certain issues (Conway, Kenski, & Wang, 2015).

This strategic communication effort has been a main domain of the political public relations field, and scholars often use the agenda-building theoretical framework to explore the relationships between the practitioners, media, and the public (i.e., Curtin, 2009; Froehlich &

Rüdiger, 2006). Public relations practitioners use information subsidies to influence the media agenda for their clients—such as political parties or candidates—and consequently to affect the public agenda (Curtin, 1999; Turk, 1986). Because of the agenda-building process, practitioners communicate and highlight certain aspects of issues, hoping to earn public understanding and support (Froehlich & Rüdiger, 2006).

#### *2.1 Agenda building and news frames*

Transferring the salience of certain issues (or the interpretations of the issues) between the media and public was the traditional agenda-setting theory's main conceptualization (McCombs & Shaw, 1972). Expanding from the classic model, scholars have developed and applied an agendabuilding theoretical framework in the numerous political communication contexts, highlighting the public relations practitioners' strategic role of communication (i.e., Kiousis et al., 2015; Turk, 1986).

Several content analysis studies have explored the relationships among the practitioners, media, and the public regarding the agenda salience (i.e., Sweetser & Brown, 2008; Roberts & McCombs, 1994). Scholars found strong correlations of issue agendas between political advertising and newspapers in the 1990 Texas gubernatorial campaign context (Roberts & McCombs, 1994). Moreover, scholars have analyzed the salience relationships between multiple sources (practitioners, media, or public) on the three different levels—object, attribute, and network associations (i.e., Guo, 2012; Kiousis et al., 2015; Schultz et al., 2012).

On the first level, scholars explored the transfer of object salience between different agendas regarding the issues, candidates, parties, or organizations (i.e., McCombs & Shaw, 1972; Turk, 1986); while the transfer of attribute salience has been explored on the second level (i.e., Entman, 1993; Sheafer, 2007). For example, issue attribute refers to the certain aspect of issues—regarding the frames (Entmans, 1993) or evaluative tones (Sheafer, 2007). These attributes affect how the media and public interpret the issues. Moreover, scholars proposed a concept of associative frames referring to the third level (i.e., Guo, 2012; Schultz et al., 2012). Due to the complexity of reality, the issues or attributes easily intertwined with each other so that the association should be explored simultaneously (Guo, 2012). During an election, candidates try to associate themselves with positive attributes while associating their opponents with negative attributes to earn voters' trust and support (Kiousis et al., 2015).

Scholars have highlighted the studies of frames, saying that it is important to study how individuals understand an issue and perceive it in a certain way (Froehlich & Rüdiger, 2006) and certain condition (Myslik et al., 2021). The framing process selects certain aspects of issues more salient than others in the messages (Entman, 1993) with a hope that the particular aspect of the issues becomes more salient on other media or public agendas (Froehlich & Rüdiger, 2006). Frames are one of the popular examples of cognitive attributes of issues under the agenda setting/building framework (i.e., Entman, 2007).

There are different types of frame categorizations identified from previous content analysis studies (Kim & Wanta, 2018). Popular news frame examples are conflict, human interest, responsibility, and economic consequences. That is, whether a news item contains conflict between individuals or groups (conflict frame), an individual's story or emotional evaluations (human interest), an attribution of problem (responsibility), or economic effects (economic consequences) (i.e., de Vreese, Peter, & Semetko, 2001; Neuman, Just, & Crigler, 1992; Valkenburg, Semetko, & de Vreese, 1999).

To explore news frames for a particular social issue, scholars also suggest other types of frames such as thematic or horse race (Kim & Wanta, 2018). Different from an episodic news frame (e.g., illustration of an individual case), the thematic frame illustrates an issue in a larger social context that considers its history or effect on the community (Iyengar, 1996). The thematic frame would help readers/viewers to understand the issue more in-depth. In contrast, the horse race frame focuses more on who is winning in a campaign context (Kim & Wanta, 2018).

Froehlich and Rüdiger (2006) also explored two levels of framing—thematic and position—linking frames to political players' strategic role in the framing process. Practitioners use thematic frames to highlight certain aspects of subtopics of an issue to guide public how to interpret the issue in a certain way. Moreover, position frames illustrate political players' solutions or actions connected to the subtopic of an issue. Practitioners aim to earn public support by telling them how an issue/problem would be solved (p. 19).

#### *2.2 Frames and Message Negativity*

Considering the role of players (e.g., political parties, candidates, or media professionals) in the framing process—from the strategic communication perspective—scholars also examined frame patterns the media or parties' ideology creates. In the context of the Smolensk plane crash, scholars analyzed Polish and Russian government messages along with the event's media coverage and found that the government's strategic communication success may differ by a medium's political ideological profile—pro-government, mainstream, or opposition (Myslik et al., 2021).

In the context of European politics, a content analysis study showed an effect of populist communication on the party's visibility in the media coverage (Schmidt, 2020). Populist messages contain the factors equivalent to the traditional news values emphasizing in-group community and social elites' critique (Schulz, 1976 in Schmidt, 2020). In nature, populist messages are more negative and hostile, and consequently, those messages have a higher chance of the media covering them (Schulz, 1976 in Schmidt, 2020).

Negative campaign is one of the popular campaign message strategies to obtain increased media attention and to engage voters with the candidate's rhetoric (Druckman, Kifer, & Parkin, 2010). In the previous studies, scholars explained that challengers generally use rhetoric that is more negative with the hope that their message will receive increased attention from the public (i.e., Druckman & McDermott, 2008; Theilmann & Wilhite, 1998).

In a hypothetical experiment, scholars tested a theoretical model, adopted from Skaperdas and Grofman's conceptualization (1995), asking practitioners whether they would use negative advertising when their candidate is leading or losing in the campaign (Theilmann & Wilhite, 1998). Study results show that practitioners would use negative advertising more when the race is getting close or when their challengers gain ground. When their own candidate is winning far ahead, practitioners said they would not use negative advertising at all and may just use positive advertising strategies.

Moreover, scholars also proposed a different pattern in using positive or negative campaign strategies between the two parties (Theilmann & Wilhite, 1998). The Republican consultants relied more on the negative advertising strategies than did the Democratic consultants. In other words, Republican voters are more open to the attacking strategies while the negative strategies influenced Democratic voters less. In their discourse analysis, Khajavi and Rasti (2020) also found that Romney, the Republican Party candidate, used more negative representation in his strategy than Obama, the Democratic Party candidate, in 2012 election

campaign speeches. Positive self-presentation or negative presentation of others (i.e., Dijk, 2005) has been a useful tool to analyze the rhetorical characteristics of political messages.

Negative or offensive remarks often appear in today's political discourse, particularly connected to the populist communication strategies. Culpeper (2011) uses the term "impoliteness" to describe all types of offensive remarks against social norms or expectations (e.g., what ought to be) (p. 23). This message strategy can be easily found from the speeches of the former U.S. President Donald Trump or former Prime Minister of Italy Silvio Berlusconi (Wodak, Culpeper, & Semino, 2021). Analyzing their official press conference messages and comments on the YouTube videos, scholars concluded that both politicians may intentionally violate social norms as a part of populist communication strategies, and their explicit and implicit impoliteness may contribute to increasing their authenticity (Wodak, 2021; Wodak, Culpeper, & Semino, 2021).

Moreover, Wodak (2021) also explained that this impolite behavior would be accepted in today's political discourse and proposed a new normal, using the term, "shameless normalization" (p. 6). When former president Trump made some remarks violating traditional norms (such as insults) without apology, a substantial amount of people's comments on YouTube videos were even positive (Wodak, Culpeper, & Semino, 2021).

In a democratic society, political disagreement or dispute is inevitable (Mutz & Reeves, 2005). The more important question concerns the manner of how to show the public that disagreement. Mutz and Reeves's (2005) experimental study showed some paradoxical findings: People reacted negatively toward the uncivil political discourse, but at the same time, they also enjoyed watching it (p. 13).

Previous studies have proposed different types of incivility categorizations. Sobieraj and Berry (2011) coded outrage media content with 13 modes including insulting, name calling, verbal fighting, mockery, or conflagration (p. 28). Incorporating some revisions from his previous work (Culpeper, 1996), Culpeper (2016) summarized the following strategies: bald onrecord impoliteness, positive impoliteness, negative impoliteness, off-record impoliteness, withhold politeness, sarcasm, or mock politeness (p. 425).

#### *2.3 Effects of Emotional Messages*

In the strategic communication perspective, framing effects (on public opinion or the decisionmaking process) research has focused mostly on the cognitive understanding of the issues. However, in today's political campaign contexts, emotions or evaluative tone can also play a significant role in the framing process (Druckman & McDermott, 2008). Scholars explored the effect of emotion on framing results (e.g., voters' risk assessment) and found that emotions significantly affected individuals' decision-making process (Druckman & McDermott, 2008). When the cognitive information in a message is the same, the emotional representation (e.g., positive or negative) of the information can lead to different reactions from the individuals (Druckman, 2004).

Moreover, scholars also examined the effects of several emotional orientations on judgment beyond a valence (positive–negative) such as enthusiasm, distress, or anger (Lerner et al., 2003, p. 144). For example, hostility or anger produced an optimistic assessment about future outcomes and people took more risks (with the hope to revenge in the future), while distress or fear produced pessimistic assessment leading toward risk-aversion decisions (Lerner & Keltner, 2000; Lerner et al., 2003).

In their campaign strategy study, Maier and Nai (2020) found that candidates who used a more negative campaign tone and more emotional appeals received greater media attention. Particularly, a negative campaign message with the use of fear appeals attracted the most media coverage (Maier & Nai, 2020). Fear (nervousness and horror), sadness (sympathy, neglect, shame, disappointment, sadness, and suffering), and anger (torment, envy, disgust, rage, exasperation, and irritation) are the common negative emotions drawn from previous studies (i.e., Culpeper et al., 2014, p. 74; Shaver et al., 1987).

Scholars (Klinger et al., 2023) explained that the use of negative emotions is part of political parties' campaign strategies because negative campaigning can bring more user engagement and reach a wider network. Klinger et al.'s (2023) quantitative content analysis study of the 2014 and 2019 European Parliament Election campaigns showed that on social media, more negative emotional posts yielded higher engagement—likes, shares, or comments than other posts.

#### *2.4 2022 Presidential Election in Korea*

On March 9, 2022, Korea had its 20th Presidential election. Among the final twelve candidates, the two major candidates were Yoon Suk-yeol (conservative, People Power Party) and Lee Jaemyung, the governing party's candidate (progressive, Democratic Party) (Choe, 2022). Voters were almost evenly split during the campaign, and Yoon won against Lee by less than 1 percentage point; his single five-year term began last May. Previously, Yoon served as the country's top prosecutor with no political experience.

During the campaign, the most salient social issues in the country were economic growth, housing prices, and income inequality (Lee & Kim, 2022). In addition, voters expected that the election results would affect Korea's future foreign relations with North Korea, China, or the U.S. (Choe, 2022; Lee & Kim, 2022).

With nonstop scandals and negativity during the campaign period, voters called this election the "election of the unfavorable" (Lee & Kim, 2022). Some concerned that this hostile and negative campaign would tire voters and would deter them from voting. However, this election recorded one of the highest voting rate—almost 77% voting rate among the registered voters.

Voters in their 20s were the major target swing group in this campaign, considering they are less ideological than the older generations are (Lee & Kim, 2022). Moreover, the campaign's youth strategy was becoming increasingly sophisticated with divided subgroups between men and women: feminist/anti-feminist rhetoric was one of the main campaign message strategies (Kang, 2022).

Hence, in the context of 2022 Presidential Election in Korea, news frames and negativity strategies in the two mainstream media coverage were explored with the following research questions:

RQ1: What news frames were presented in the mainstream media coverage in Korea about the 2022 Presidential election?

RQ2: What negativity strategies were presented in the mainstream media coverage in Korea about the 2022 Presidential election?

RQ3: How the frames and negativity strategies in the mainstream media coverage in Korea about the 2022 Presidential election differ by media's political ideology?

RQ4: How the negativity strategies connect to the frames presented in the mainstream media coverage in Korea during the 2022 Presidential election period?

#### **3 Method**

This study conducted a qualitative content analysis to identify thematic frames and campaign emotion discourses in the news coverage about the 2022 Presidential election.

#### *3.1 Samples*

Two mainstream newspapers in Korea were selected to gather news articles about the presidential election and to review the media portrayals of the presidential candidates' strategic communication efforts. First, considering newspapers' perceived political ideology, the following two newspapers were selected to study: 조선일보 *(Chosun Ilbo)* and 한겨레 *(Hankyoreh)*. Among the top daily newspapers in Korea, 조선일보 *(Chosun Ilbo)* is considered right-leaning and 한겨레 *(Hankyoreh)* is considered left-leaning (Lee et al., 2022). Previously, a number of scholars have shed light on the impact of the political orientation of the media on the strategic communication outcomes (i.e., Ban, 2018; Lee et al., 2022; Myslik et al., 2021). In the context of Korea's superintendent election in 2014, Lee (2014) conducted a comparative study of news contents between the two newspapers and found different patterns based on the political disposition of the newspaper organizations.

In this study, news articles were collected from each news outlet's main website (chosun.com and hani.co.kr) using a search function. Three search keywords were '대선 (presidential election)' and the names of the two leading presidential candidates—'윤석열 (Yoon, Suk-yeol)' and '이재명 (Lee, Jae-myung).' Between February 1 and March 8, 2022, all available news articles were selected excluding visual-only contents or irrelevant stories. The official election campaign began on February 15 and the election happened on March 9. A total of 485 (out of 683) and 401 (out of 518) articles were analyzed from 조선일보 *(Chosun Ilbo)* and 한겨레*(Hankyoreh)*, respectively.

#### *3.2 Analysis*

First, the collected articles were reviewed to identify the frames used in news stories. Some of the most common news topic frames found in previous media studies were conflict (e.g., disagreement between two or more individuals or organizations), human interest (e.g., individual and personal aspects), responsibility (e.g., attribute of responsibility for causing or solving a problem), economic consequences (e.g., cost, expense, or financial loss/gain), campaign strategies/process (e.g., management of campaigns—how to attract target voters and collaborate with other political leaders), issue policy (e.g., candidate's issue position and decision-making process), or personality (e.g., candidate's personality, manners, style, education, and experience) (i.e., de Vreese et al., 2001; D'Angelo et al., 2005; Valkenburg et al., 1999).

To understand news frames in the Korean election context, this study analyzed phrases or terms used in each article and then classified similar patterns into the groups, and this qualitative thematic approach allows researchers to find context-based frames directly from the data (Gioia et al., 2013; Linneberg & Steffen, 2019; Van Gorp, 2007, 2010). The identified news frames

from the news stories would help us to understand the main themes or major issues of Korean society related to the 2022 presidential election in Korea.

Moreover, two-leading presidential candidates' speech quoted in the news articles were analyzed to review the use of negative campaign strategies during the election period. In the previous communication discourse studies, scholars defined impoliteness as a communicative strategy to attack others' face causing social conflicts intentionally (strategically) (Culpeper, Bousfield, & Wichmann, 2003, p. 1546). In political debates, for example, 'conflictive talk' (see Culpeper, Bousfield, & Wichmann, 2003, p. 1545) is a part of communicative discourse strategies of the candidates. Scholars also proposed a few substrategies: positive impoliteness (e.g., ignore, exclude, disassociate, disinterested, obscure language, or calling names) and negative impoliteness (e.g., frighten, ridicule, invade the other's space, or interrupt) (Brown & Levinson, 1987; Culpeper, 2016, p. 425). These categorizations are neither mutually exclusive nor complete yet (Culpeper, 2016). From the news articles, a full length of debate is not available, so instead of the conversation between the candidates, quotations by the two-leading candidates were separately analyzed if presented in the news stories.

In this campaign context, the official election campaign period began on February 15, 2022, and we compared results between Time 1 (prior to February 15) and Time 2 (February 15 until election day) to ascertain how news frames and other characteristics were different before and after the main campaign.

This study also explored how specific negative patterns in the news about the Korean presidential election are different from previous studies and how emotional remarks are associated to news frames. Finally, the frames and negative strategies found in the news media coverage were compared between the two news media outlets to discuss the impact of the political orientation of news organizations on strategic communication outcomes.

#### **4 Results**

A qualitative content analysis was conducted to explore the frames and negativity strategy discourses in the 2022 presidential election news coverage. Two mainstream newspaper stories were collected for a month-long period prior to the election. The results of a qualitative content analysis summarized into the nine election news frames and emphasized three main negativity remarks from the candidates (i.e., Culpeper, 1996, 2016). Moreover, the results were compared by time and the mediums' political orientations.

The first research question was proposed to examine the news frames. *What news frames were presented in the mainstream media coverage in Korea about the 2022 presidential election?*

The most common news frames found in media coverage were conflict, strategy, issue position, public engagement, race, announcement, investigation, personal story, and critique (i.e., de Vreese et al., 2001; D'Angelo et al., 2005; Valkenburg et al., 1999). In addition, most news stories were either neutral or negative.

Similar to the previous studies, the most popular news frame was conflict. During election campaigns, candidates argue on many social issues (e.g., special prosecutor, media freedom, family risk, or COVID-19 risk) based on their political ideologies: "*Lee Jae-myung, the Democratic Party of Korea's candidate, said on the 18th that Yoon Suk-yeol, the People Power Party's candidate's Gwangju shopping mall complex proposal is 'far-right populism dividing* 

*sides'*" *1* (*Chosun Ilbo,* February 19, 2022: para.1), or "*Presidential candidates who will lead the next government said that they will prepare a policy to manage microplastics; however, they differ on the use of mandatory measure implementation*" *2* (*Hankyoreh,* February 20, 2022: para.1).

The strategy frame focuses on the management of campaigns. Results also show that it is important to deal with an opponent's negative advertising or scandalous messages: "*With the dramatic unification of the People Power Party's presidential candidate Yoon Suk-yeol and People's Party candidate Ahn Cheol-soo six days before the election day, it has become the biggest question whether it will change the election situation*" *3* (*Chosun Ilbo,* March 4, 2022: para.1), or "*Lee Jae-myung, the Democratic Party of Korea's presidential candidate, toured the Gangnam area in Seoul on the 6th to spread support.… [and] emphasized a 'country of opportunity for the youth'*" *4* (*Hankyoreh,* February 16, 2022: para.1).

The issue position frame describes the political issues and candidates' political decisionmaking processes. Some of the main political issues in this campaign period were housing prices, labor, gender, and economy. Candidates explained their positions on these issues by telling the media and public how they would solve related problems: "*Both Lee Jae-myung, the Democratic Party of Korea's candidate, and Yoon Suk-yeol, the People Power Party's candidate, criticized the Moon Jae-in government's quarantine policy that focuses on business restrictions.… If elected, I will stop unnecessary excessive quarantine policy and will allow business to remain open until midnight for those who have received booster shots*" *5* (*Chosun Ilbo,* February 21, 2022: para.1), or "*The two leading candidates' [Lee Jae-myung of the Democratic Party of Korea's and Yoon Suk-yeol of the People Power Party] economic policy has three main differences*" *6* (*Hankyoreh,* February 07, 2022: para.1).

The public engagement frame describes campaign activities during which candidates meet supporters and deliver speeches. Candidates run nationwide election campaign tours with their campaign teams and interact with voters: "*20th presidential election candidates' campaign style has become an issue*" *7* (*Chosun Ilbo,* February 17, 2022: para.1), or "*On the 15th, when the official campaign for the 20th presidential election began, both ruling and opposite candidates rushed into their 22-day campaign period*" *8* (*Hankyoreh,* February 15, 2022: para.1).

The race frame is also a popular political news frame during an election. Simply, news stories present who is winning or losing based on public opinion polls. The 2022 election was one of the most competitive elections in the country's history: "*Yoon Suk-yeol, the People Power Party's candidate, was 35% and Lee Jae-myung, the Democratic Party of Korea's candidate, was 31%*" *9* (*Chosun Ilbo,* February 08, 2022: para. 1), or "*Lee Jae-myung was 32.6%, and Yoon Suk-yeol was 38.8%*" *<sup>10</sup>* (*Hankyoreh,* February 07, 2022: para. 1).

The announcement frame shares candidates' campaign schedules, such as press conferences, debates, and other public activities. Furthermore, it includes election schedules, such as voter registration, early voting, and overseas voting: "*Candidate Lee went on a relay campaign that started in Suncheon, Jeollanamdo, and toured the Honam region for two days and two nights*" *<sup>11</sup>* (*Chosun Ilbo,* February 19, 2022: para.1), or "*The official campaign for the 20th presidential election begins on the 15th*" *<sup>12</sup>* (*Hankyoreh,* February 14, 2022: para.1).

The investigation frame focuses on the official actions of investigating and attributing responsibility when scandalous or illegal accusations are made against candidates. Investigation results may affect voters' risk assessment of each candidate: "*In the second TV presidential* 

<sup>1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12</sup> Translations are italicized.

*debate on the 11th, controversy around the spouses of Democratic Party candidate Lee Jaemyung and People Power Party candidate Yoon Seok-yeol became the main issue*" <sup>13</sup> (*Chosun Ilbo,* February 12, 2022: para.1), or "*The Daejang-dong [development project corruption] scandal backfires not only on the Democratic Party presidential candidate Lee Jae-myung but also on Yoon Seok-yeol*" *<sup>14</sup>* (*Hankyoreh,* March 7, 2022: para.1).

The personal story frame highlights candidates' individual and personal aspects. Candidates' social media channels become important sources for this frame: "*The official presidential election campaign began; however, spouses of candidates did not appear publicly at campaign events*" *<sup>15</sup>* (*Chosun Ilbo,* February 19, 2022: para.1), or "*A controversy erupted when People Power Party's presidential candidate Yoon Seok-yeol posted a photo on his social networking services (SNS) of a tangerine with an angry face along with a post supporting Ukraine*" *<sup>16</sup>* (*Hankyoreh,* March 2, 2022: para.1).

Finally, the critique frame is used when a political elite or public figure evaluates candidates' qualifications or morals: "*About 40 people, including former and current Democratic Party officials, members, and national delegates, declared their support for People Power Party presidential candidate Yoon Seok-yeol.… They said it is because the Democratic Party nominated a person with a four-time criminal record as their candidate*" *<sup>17</sup>* (*Chosun Ilbo,* March 5, 2022: para. 1), or "*Kim Dong-yeon of New Wave said on the 7th that he met Yoon Seok-yeol, People Power Party's presidential candidate, last month; it seems that he is not thinking deeply about constitutional amendment and political change*" *<sup>18</sup>* (*Hankyoreh,* March 7, 2022: para.1).

Results showed that the types of frames differ over time. In this campaign context, the official election campaign period began February 15, 2022, and the analysis results were compared between Time 1 (prior to February 15) and Time 2 (February 15 until election day). Prior to the official campaign period, the most frequent election frames were announcement, investigation, and personal stories. During the official campaign period, the most common election frames were strategies, public engagement, issue position, conflict, race, and critique.

The second research question was proposed to examine the negativity covered in the media. *What negativity strategies were presented in the mainstream media coverage during Korea's 2022 presidential election?* 

Analysis showed that during the election, negative impoliteness in Korean newspapers was most frequently reported (Culpeper, 1996, 2016). Particularly, newspapers reported a campaign message when it explicitly associated a negative aspect to the opponent. Among the positive impoliteness strategies damaging one's positive countenance desire (to be appreciated), name-calling was commonly reported. Other strategies, such as seeking disagreement, disassociating one from the other, using inappropriate identity, or frightening voters, also manifested in media coverage (Culpeper, 2016, p. 425).

For example, during the 2022 Korean presidential election, some of the campaign keywords used in the communication messages were populism, communism, feminism, and political revenge. Candidates also disregarded their opponents by using the words unfavorable, inexperienced, and unintellectual. Sometimes, one called the other "crazy" or referred to them as a "gangster." When the analysis results were compared between Time 1 (prior to February 15) and Time 2 (February 15 until the election day), negative politeness (explicit negative remarks toward the other candidate, such as populists or feminists) was found more often in Time 2.

<sup>13, 14, 15</sup>**,** 16**,** 17, 18 Translations are italicized.

Then, the third research question examined the effect of the medium's political ideology orientation. *How do the frames and negativity strategies in the mainstream media coverage regarding the Korean 2022 presidential election differ by the media's political ideology?* 

Results show that the frame types are different depending on the media's ideological orientation. The two newspapers selected in this study represented either right-leaning or leftleaning media (Lee et al., 2022). When comparing analysis results between the two newspapers, the most frequent election frames were issue position, announcement, investigation, and critique for 한겨레 *(Hankyoreh,* left-leaning) and strategy, conflict, public engagement, race, and personal story for 조선일보 *(Chosun Ilbo,* right-leaning). The results showed that negative impoliteness was found more in the left-leaning paper, perhaps because the popular news frames of that medium focus more on issue positions, candidates' moral evaluation, or investigations (of scandal or illegal incident) that may harm the others negatively, rather than campaign strategies or simple conflicts between the candidates (i.e., Brown & Levinson, 1987; Rheem & Ahmed, 2022).

Finally, the fourth research question explored how negativity strategies can be connected to the election frame in the media. *How do the negativity strategies connect to the frames presented in the mainstream media coverage during the 2022 Korean presidential election period?* 

The results showed that the negativity remarks in the news occurred more when the campaign strategy, issue position, or public engagement frames were used. Political issues successfully drew the media's attention when the issues were presented along with the candidates' communication efforts (e.g., candidates' approaches to the political issues and their solutions). Populism and feminism were popular campaign rhetoric during the election, and with these issues, a candidate also easily associated a negative remark to the other candidate by calling him/her a populist (or anti-populist) or feminist (or anti-feminist).

#### **5 Discussion**

This study examined the mainstream news coverage for the 2022 Korean presidential election to identify the most commonly used frames and negativity remarks during the campaign. From the strategic communication perspective, news media content can be considered an outcome of practitioners' campaign message strategies (i.e., Conway, Kenski, & Wang, 2015; Fountaine, 2017; Nadeau, Pétry, & Bélanger, 2010).

Traditionally, frame is about how one emphasizes a particular aspect of an issue with the hope that it affects the media and public's cognitive understanding of the issue (Entman, 1993; Froehlich & Rüdiger, 2006). Some of the popular news frames used in the previous studies include conflict, horse race, or human interest (i.e., de Vreese, Peter, & Semetko, 2001; Kim & Wanta, 2018). Semetko and Valkenburg (2000) also analyzed Dutch national news media content using the five common news frames: attribution of responsibility, conflict, human interest, economic consequences, and morality (p. 93).

In the current study, "election" news frames were also summarized into nine frame patterns. Still, some of the traditional news frames were prevalent, such as conflict and race (i.e., de Vreese, Peter, & Semetko, 2001; Kim & Wanta, 2018). Theses frames mainly describe the characteristics of the issues important in the election, and campaign strategy is often considered a simple win–lose game. On the other hand, this study's results show that, during the Korean presidential election, a campaign's active role was emphasized in the news coverage. Through

the strategy and issue position frames, candidates' campaign-related activities and their communication strategies (as well as the strategies' effect on the public) have been highlighted in the media. This may show that political public relations information subsidies play a significant role in affecting news agendas during elections.

Not only cognitive attributes but also emotional (affective) attribute issues have been emphasized in the previous agenda-building strategic communication studies (Druckman & McDermott, 2008; Kiousis et al., 2015). Scholars have explained that emotions can play a significant role in affecting an individual's decision-making process (Druckman & McDermott, 2008), and practitioners try to associate their candidates with positive attributes, while associating their opponents with negative attributes to gain public trust (Kiousis et al., 2015). Emotion is important in today's election context due to the development of social media tools and the involvement of younger generations (Fountaine, 2017).

Moreover, in today's political discourse, negative communication is considered an important campaign strategy. For example, scholars used the term "impoliteness" to describe certain types of offensive remarks (Culpeper, 2011, p. 23; Wodak, Culpeper, & Semino, 2021). Even though these negative remarks often go against the social norms, the public sometimes enjoys this intentional and strategic negativity in the campaign messages and perceives the message as authentic (Wodak, 2021; Wodak, Culpeper, & Semino, 2021).

Naturally, negative and aggressive messages have a higher chance of being covered in the media (Schulz, 1976 in Schmidt, 2020), as people are interested psychologically in the negative stories. During the 2022 Korean presidential election, negative remarks were found in the news media coverage; however, the frequency was somewhat moderate. Politicians' official information subsidies may present wide ranges of positive/negative impoliteness, such as insulting language, verbal fighting, exaggeration, or mockery (Sobieraj & Berry, 2011, p. 39– 41). However, only selected appeared in the media coverage.

During the presidential election campaign, ideological negative language, such as populism, communism, feminism, or political revenge, was used often in the Korean news coverage. After experiencing a series of political corruptions and social inequality (e.g., gender or income), people, particularly the younger generation (i.e., those in their 20s), feel deprived and divided. For example, feminism was a hot topic for the country, with an intensified gender war between young men and women (Seo & Hollingsworth, 2022). Considering the country's unique political situation, candidates also had different views on the country's foreign relations with North Korea, China, and the U.S. (Chandran, 2022). With the political uncertainty and risk, communism versus populism became important campaign rhetoric in Korea and a source for negative remarks during the campaign. Negative remarks associated with a populist/anti-populist or feminist/anti-feminist successfully drew media attention in this study.

Moreover, the results show that the frame categorizations differ over time: Time 1 (February 1 to 14) and Time 2 (February 15 to March 9). The most frequent frames were announcement, investigation, and personal stories in Time 1; and the most common frames were strategies, public engagement, issue position, conflict, race, and critique in Time 2. During the official campaign period (Time 2), more diverse and strategic frames were used in the media coverage.

When the analysis results were compared between the two newspapers, the most frequent frames were issue position, announcement, investigation, and critique for a left-leaning newspaper; and strategy, conflict, public engagement, race, and personal story were common for a right-leaning newspaper. Previously, scholars showed that a government's strategic

communication effect might differ by the political orientation of the medium (Myslik et al., 2021). This result also supports the effect of the media organization's ideological profile.

The analysis compared over time and by the political orientation of the media outlet showed that negative politeness occurred more often closer to the election. In addition, this study showed that negative impoliteness occurred more often in the news when framed within certain campaign strategies or an issue position. Practitioners may consider these message factors when developing campaign messages. For example, use of negative remarks can be considered a strategic campaign plan to draw attention from people and the media. When the election is competitive and split, candidates would use more negative emotions on other candidates (i.e., Dijk, 2005), which reflects in media content.

Future research should investigate the role of impoliteness message strategies with multiple media sources for not only mainstream newspapers but also TV news or social media platforms. The media type may affect the use of negative remarks in their news coverage by considering the media's purpose and target audience. Scholars should explore whether Internetbased social media would increase negativity in election campaigns. In addition, future study may expand on campaign messages or news analyses in different election situations. A local or statewide election may differ from a presidential election because of its competitiveness or the amount of public attention. Moreover, a country's culture and social norms may affect the use of negativity strategies, so future study should examine campaign message strategies and their impacts in other countries' election contexts.

#### **References**


D'Angelo, P., Calderone, M., & Territola, A. (2005). Strategy and issue framing: An exploratory analysis of topics and frames in campaign 2004 print news. *Atlantic Journal of Communication, 13*(4), 199-219. https://doiorg.eres.library.manoa.hawaii.edu/10.1207/s15456889ajc1304\_1


media relations and agenda building during the Smolensk airplane crash. *International Communication Gazette, 83*(2), 169-192. https://doi.org/10.1177/1748048519853766


https://doi.org/10.1080/10584609.1994.9963030


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Facework in Patient Requests for Treatment Recommendations in Korean Medical Interactions: The Use of Main Clause Omission**

#### **Yujong Park<sup>1</sup>**

**Abstract** This study examined the involvement of facework in main clause omissions employed by patients when performing requests. The data were drawn from a corpus of videotaped primary care visits collected from Korean medical practitioners. Drawing on previous work on facework in politeness research, requests in conversation analytic research, and main clause ellipsis in Korean, the study found that main clause omissions provided a mechanism for patients to minimize any threats to the participants' face when requesting a prescription for a treatment plan. By employing this practice, patients were able to prove the relevance for a proposed treatment without directly requesting it. By only providing the relevance for a specific type of treatment by omitting the main clause, the patient allowed the doctor to indirectly reject or grant the request. The analysis allowed the notion of facework to be respecified as one emerging from both participants' joint construction of the sequential unfolding of the interaction. This study supports the view that face is a discursive, interactional, and social accomplishment by examining a linguistic practice employed primarily by patients in Korean medical interactions. The role of epistemic imbalance and asymmetry in this setting is also discussed.

**Keywords** Facework Requests Main clause omission Treatment plans Medical interaction

<sup>1</sup> Y. Park ()

English Literature and Language Department, Sungkyunkwan University, Seoul, South Korea e-mail: yujpark@skku.edu

#### **1 Introduction**

In medical interactions, patients rarely propose for a specific treatment plan (Kravitz et al., 2003; Stivers, 2002) as it is considered a highly face-threatening act (FTA) because it imposes on the doctor's authority (Robins & Wolf, 1988). Because doctor-patient interaction is characterized by a knowledge-based asymmetry or epistemic imbalance (Pilnick & Dingwall, 2011; Drew, 2018), the performance of politeness in doctor-patient interactions had been studied in relation to how this imbalance is managed through a variety of linguistic or interactional practices (Cordella, 2007; Granger et al, 2005; Robins & Wolf, 1988). For example, doctors may threaten the patients' negative face by using warnings to make patients understand the consequences of noncompliance with treatment recommendations (Cordella, 2007). Although there are a number of studies which investigated how doctors can threaten a patient's face when prescribing treatment or conveying bad news (Grainger et al, 2005; Robins & Wolf, 1988; Silverman et al., 2005), fewer studies have examined the patient's performance of face-threatening actions in medical encounters, even though patients also have many opportunities to threaten the doctors' face, such as rejecting a treatment plan, advocating for medication (i.e., Stivers, 2002) or requesting more information. Filling this gap, this study investigates how patients' perform FTAs in an interactional, sequential accomplishment of requesting for treatment. We focus on how patients work to provide the relevance for treatment without having for themselves to request it, primarily using main clause omission as a face saving practice in the Korean primary care settings.

Past research has shown that when engaging in FTAs, Koreans tend to employ a greater number of hedges and accounts in longer turns (Sohn, 1986) and frequently omit the main clauses (Byon, 2006; Rhee, 2008; Sohn, 2003). In Example 1 below, the son omits a main clause that questions his mother's request to run an errand (i.e., Are you asking me to run an errand?) and only states the reason for his inability to fulfill this request (line 2, *konpwuhalamyense,* "you told me to study"). The interrogative main clause is omitted (put in parenthesis) leaving only a subordinate clause that ends with the Korean clausal connective -*lamyense.* 

Example 1. Rhee (2014, modified, invented p. 604)

01 Mom: 심부름 좀 해. simpwulum com hay. errand-DM-do **"Run an errand for me."** 02 Son: 공부하라면서 (심부름가래?) kongpwuha-lamyense (simpwulum-ka-lay?) study-CONN errand-go-QP **"(Are you asking me to run an errand) while telling me to study?"**

Since main clauses in Korean usually carry the speaker's assertion, main clause omissions can be used to protect the face of the hearer by withholding the assertion that puts an imposition on the hearer (e.g., Are *you asking me* to run an errand). However, despite the frequency of this construction in Korean request exchanges, in particular, there has been little research into main clause omissions and their relationship to facework.

The use of requests in this study is a broad one, akin to Searle's (1969) "directive" speech act category, which forms threats to the other's positive or negative face (Goffman, 1967). A variety of forms may be used to perform requests (Austin, 1962, Searle, 1969). Studies in line with Brown and Levinson's (1987) politeness theory have argued that in general, indirect linguistic speech act forms are less face threatening and polite as they increase the degree of optionality and decrease the force of the illocutions in speech acts. The graded notion of indirectness assumed in previous studies has taken "direct" and "indirect" to refer to the extremes of an (in)directness scale, with imperatives being the most "direct" request forms, and hints being the most "indirect" (Blum-Kulka 1987; Blum-Kulka & Olshtain, 1984; Brown & Levinson 1987; Leech 1983). The use of main clause omissions could be categorized as an indirect linguistic form similar to hints described in the literature. In conversation analysis (CA), the notion of preference is frequently employed to explain how requests are performed (Atkinson & Drew, 1979; Pomerantz, 1984). The basic claim is that dispreferred actions such as requests or invitations are accompanied by marked features such as delays, accounts, prefaces or mitigations. Preference choices are in many cases motivated and determined by face considerations (Brown & Levinson, 1989) although the face concept is not frequently employed in CA research to explain preference organizations. Requests from patients constitute dispreferred actions in doctor-patient interaction where preservation of face may be an important consideration in managing the epistemic imbalance between participants.

This paper proposes that main clause omissions may be employed as a locally and sequentially managed resource to accomplish facework when patients deliver FTAs during Korean medical encounters. The face concept was developed by Goffman (1967) as a derivation from Durkheim (1915) and refers to the "positive social value a person effectively claims for himself by the line others assume he has taken during a particular contact" (p. 5). Facework is a set of coordinated practices that communicators use language to build, enhance, maintain and challenge interpersonal communication relationships (Domenici & Littlejohn, 2006; Locher, 2004, 2013). Facework can be achieved through the various devices users employ that allow them to negotiate their social identities and maintain their sense of dignity and place in a given social space during the emergent discourse (Locher, 2011).

Conversation analysis (CA) studies that have criticized Goffman's face notion as being too general and as presuming the existence of an already constituted, recognizable action (Schegloff, 1988) have used the notion of (dis)affiliation (Heritage & Raymond, 2005; Stivers, 2008), with affiliation in this context being defined as "the affective level of cooperation." According to Stiver et al. (2011, p. 21), affiliative responses are "maximally pro-social when they match the prior speaker's evaluative stance, display empathy and/or cooperate with the preference of the prior action." Therefore, the (dis)affiliation notion subsumes issues related to both facework and politeness in this line of work. Schegloff (1988) pointed out that the constitution of an action that relies on generic features of the organization of talk is anterior to notions such as face and politeness. Generic features, such as turn taking, sequencing, and repair, are involved in the turnby-turn joint construction and negotiation of possible face-threatening actions in an interactional moment. In this way, the organization of talk such as preference organization provides opportunities to avoid face threats. While acknowledging the effectiveness of the "(dis)affiliation" notion in explaining politeness phenomenon, this study primarily discusses facework in relation to the generic features of the organization of talk.

While it is possible to employ a face perspective to understand CA notions such as repair and preference organization, the focus of CA work has been to emphasize the sequential and collaborative nature of talk; however, recent work has begun to explore the relationships between concepts developed in CA and (im)politeness research (Chevalier, 2009; Karafoti, 2021). For example, Lerner (1996) discusses the ways in which the anticipatory completion of a turn could

transform a structurally dispreferred action such as a disagreement into its preferred alternative (i.e., agreement) and recognized this kind of completion as a potential locus for the study of face work in interaction. Hutchby (2008) outlined the contributions that CA may make to politeness research, as well as its natural limitations. The discursive turn in politeness research have also begun to recognize politeness as a social order that penetrates the here-and-now of conversation (Kàdàr & Haugh, 2013). Here, (im)politeness, face-saving, face-threats or face aggravation is situationally, institutionally, or contextually bound and revealed in the discourse (Bucholtz & Hall, 2005). This chapter contributes to this line of research by engaging in a sequential analysis of a specific linguistic practice (i.e., requests performed through main clause omission) to examine how face work may be displayed and co-constructed in and through interaction in the Korean medical context.

The current study argues that main clause omissions in Korean are a patient's resource oriented toward the face-threatening nature of requests and employed in a collaborative, sequential accomplishment of facework. Because treatment decisions are in the realm of the doctors' power, patients' request for a preferred treatment plan must be carefully constructed in line with how the doctor responds. In the following section, the key aspects of main clause omissions in Korean are based on previous research, after which instances are taken from the data set and the possible connections between the sequential deployment of main clause omissions and facework in the Korean language are discussed. The study was driven by the following question:

How is facework collaboratively performed through the use of main clause omissions in patients' request for treatment plans during Korean medical encounters?

#### **2 Main Clause Omissions in Korean Talk-in-Interaction**

Main clause omissions during interactions have been frequently observed in Korean. In a typical main clause ellipsis (Rhee, 2020; Sohn, 2003), which is this study's main focus, the main clause in the latter part of the sentence is withheld, leaving the subordinate clause ending with sentence final particles or different connectives (e.g., -*ntey, -se, -nikka*). Because ellipsis is a term that is used to refer to a wide range of syntactic, semantic and pragmatic structures (Stainton, 2006), the term "omission" is used in this study to focus attention on the deliberate interactional withholding of turns completed by the current speaker.

 When the main clause is missing, the hearer must reconstruct the missing information from the omitted main clause (Heine et al., 1991). These deliberately invited inferences or reinterpretations are often employed in delicate or problematic interactions, as shown in example 1 above (i.e., challenges to requests). This study complements previous research on Korean main clause ellipsis by sequentially analyzing these turns to reveal whether the recipients understood the speaker's actions and were sensitive to the trouble in the talk.

Rhee (2008) claimed that interactional ellipsis has resulted in the grammaticalization of Korean complementizers (e.g., -*lako, -tako*), which have gradually evolved into sentential endings. Ending forms such as "*lako"* have developed from elliptical structures involving complementizers that emphasize a statement, a question, or a command that was presented before the current turn. For example, in *ppalli o-lako*, "Come quickly-*lako*," instead of completing the sentence by stating the main clause as in *ppalli o-lako mal-hay-ss-e*, "I have already said this, come quickly," the speaker simply stops the sentence midway after the complementizer "-*lako*" to indicate the unstated information which is "I have already said this before" (Rhee, 2008, p. 14). Similar

instances have been examined in other languages and have been given various labels, such as "suspended clauses" (Ohori, 1995), "inconsequential clauses" (Haiman, 1988), and "insubordination" (Evans, 2007, 2009, Evans & Watanabe, 2016). Insubordination refers to a type of local ellipsis in which a structural change from [S[Sub.Clause-CONN Main.Clause]] to [S[Clause-CONN]] and further to [S[Clause-SFP] occurs. The process involves the connective (CONN) being reinterpreted as a sentence final particle (SFP) in the final stages (Rhee, 2020). Because it is unclear whether the connective function of connectives has survived or not, the structure may be ambiguous because of the opposing functions of connective clauses and ending sentences. For the parties in interaction, however, there does not seem to be an ambiguity because to project what the turn seeks to accomplish, the main clause omissions in Korean are accompanied by interactional resources, such as sequential positions, connectives/sentence final suffixes, sound stretches, and non-verbal actions (e.g., laughter, gaze).

The following example (Example 2) shows how a patient employs main clause omission as a challenge to the doctor's diagnosis, which is accompanied by laugh particles. Here, the patient is questioning the doctor's proposal that this (Tuberculous lymphadeniti) is not a rare disease (lines 1-3) by employing the target practice (line 6, *cey cwuwi-ey amwuto epse-kaciko*, "nobody around me has it so hhhe hhh").

#### Example 2. SS\_01\_TL (56-67)



When the doctor delivers the diagnosis along with an explanation that tuberculous lymphadentitis commonly occurs when one's lymph node grows in size (line 03), the patient challenges the doctor's use of the term "commonly occurs" by repeating the doctor's words using questioning intonation (line 4, *cal sayngki-nun kemnikka*, "(It) commonly occurs?"). After the doctor responds to this challenge with a reassurance, the patient acknowledges the doctor's response with a change of state token *ah* "oh" (Heritage, 1984) and provides an account for having questioned him earlier (line 6, *cey cwuwi-ey amwuto epse-kaciko hhhe hhh*, "nobody around me has it so hhhe hhh"). In place of the main clause which may have conveyed the actual opposition in the lines of "Nobody around me has it (so I didn't believe you)" and an indirect challenge, the patient inserts laugh particles (hhhe hhh). The injection of laugh particles into talk is often associated with the reporting of misdeeds of some kind (Jefferson et al., 1987), especially in medical consultations (Haakana, 2001; Heritage & Robinson, 2006). In the next turns, the doctor grunts (line 8, "hm"), and then provides additional evidence for the claim that this is not an uncommon disease by providing a contrast with tuberculosis occurring in other areas of the body. Therefore, the withholding of the main clauses does not create ambiguity here; rather, it has an important interactional purpose related to facework in which the patient minimizes possible threats to the doctor's face by not fully articulating his doubts. It is worth noting that facework is also performed via nonverbal resources such as laughter along with the employment of main clause ellipsis.

Ellipsis theories differ widely on the degree to which the ellipsis material is recovered. For example, Quirk et al. (1972, p.536) restricted the use of the term "ellipsis" to describe words that were uniquely recoverable, that is, when there is no doubt about the missing words. Evans (2006) defined ellipsis as involving "some recoverable elements that are grammatically acceptable," and then defined a range of situations from "uniquely recoverable to non-uniquely recoverable" (p.370). Following Evan's definition, when main clause ellipses occurred in the data set, it was assumed that a grammatically compatible main clause would be reconstructed by the hearer. Evans (2006) stated that when determining if the regular ellipsis is the best analysis for a given language, it is necessary to conduct sensitive language-specific tests. For example, there may be syntactic evidence for the underlying presence of the main clause, such as the presence of negative polarity items like "ever" or "any" in an English clause as in "Like I'll ever give you any money?" the

presence of which can only be accounted for by an ellipsed negative matrix clause like "You don't believe." In Korean, the syntactic evidence for an underlying main clause is most effectively displayed using connectives at the end of the subordinate clause (subordinating conjunction), such as *"-lako,*" *"-nuntey,*" and *"-se.*" Due to the scope of this paper, the conventionalized usage of insubordinate clauses or grammatical forms that emerge through ellipsis are not considered; however, the importance of this line of research (Evans & Watanabe, 2016; Rhee, 2020) in understanding the target phenomenon is acknowledged. By examining patient requests without a main clause using CA and assessing the relationship with facework, this study sought to fill a gap in previous research.

#### **3 Method**

The present study adopts a conversation analytic approach as applied to institutional interactions, (Heritage & Maynard, 2006; Pilnick et al., 2009) to examine politeness as a social practice in the collected interactions (Kàdàr & Haugh, 2013). Discursive theories of politeness have considered it important to make native speaker assessments of politeness "the basis of a discursive, datadriven, bottom-up approach to politeness" (Locher & Watts, 2005, p. 6) by building theories of politeness that rely on the participants' perspectives. Eelen (2001) argued for the importance of using naturally occurring data in politeness research by stating that "due to the situational embeddedness of politeness, they would have to derive from natural settings and occur spontaneously, as elicited evaluations and/or an experimental setting introduce particular social aspects and motivations that warrant their classification as separate social practices" (p. 255).

The current data were drawn from a corpus of sixty videotaped primary care visits that had been collected from Korean medical practitioners. Two community-based practices and two university-based practices located in two Korean cities (Seoul and Chonan) allowed the researcher to collect videotapes of the doctor-patient interactions that had occurred in the family medical departments and internal clinics. All data collection was approved by the university human-subject protection committee, and the participants provided informed consent to being recorded before the study, were aware of being recorded, and permitted the recordings to be published. Of the 60 cases, five encounters in which a patient requested a specific treatment plan were selected for analysis. In the following extracts, three instances are given that involved visits to the family medical department in two separate hospital settings. All data were transcribed by the author and all names were changed to maintain confidentiality. The Korean was romanized using the Yale system, which represents the actual sounds rather than the standard Korean orthography. The three lines used in the transcripts respectively represent the sound, the morpheme-by-morpheme gloss, and the English translation.

Rather than restricting analysis to isolated sentences or phrases, CA examines the social actions that people seek to accomplish through their interactions. CA was used to search for interactional patterns in the interaction that gave some evidence of systematic usage to accomplish a particular social action (Heritage, 1984; Schegloff, 1968). For example, previous CA research on medical encounters has identified various practices used for opening and closing an encounter (Robinson, 1998; Park, 2017) and for delivering diagnoses and treatment (Peräkylä, 1998; Stivers, 2002). This paper assessed the main clause omission practice used by the patients to request the desired treatment plan. The analysis proceeds by addressing 2) how patient requests' (performed through main clause omissions) can be either successful or unsuccessful by investigating its sequential development and 2) how the focal practice illustrates the co-constructive nature of politeness and facework in this setting.

#### **4 Analysis: Facework and Main Clause Omissions in Patient Requests**

This section argues that patients frequently employ main clause omissions when they seek to advocate for a specific treatment plan or when delivering requests for a prescription that is delicate or problematic. The analysis centers around the question of how patients' requests (in the form of main clause omission) become (un)successful and how the employment of this practice is related to politeness and facework. Examples 3 and 4 show cases in which patients' requests are unsuccessful (doctors reject the need for the treatment being requested) whereas examples 5 and 6 display cases in which patients' request for treatment are granted. All target turns (arrowed lines) that lead up to the request do not contain a main clause and instead end with a connective that allows the speaker to avoid stating the actual request. In this way, the main clause omissions are used to hint at trouble in the talk (during the FTA) and leave it to the recipient to figure out the main import of the action. Doctors' responses to the patients' main clause omission based requests as well as the participants' use of non-vocal resources are also examined.

In Example 3 below the patient reported that she had been bitten by her dog several days ago (lines 01-03). When the doctor asked for the relevance of this report (line 06, *yey kuntey?* "Yes, so?"), the patient inquired about the need for a shot by repeating that she had been bitten by her dog (line 07) followed by an indirect request (line 09, *cwusalul macaya tweyna,* "get a shot or (not)"). This request is incomplete as the patient ends the turn with a questioning final suffix -*na*  INTERR without articulating the main clause (i.e., *kekcengtweyseyo, "(I) was worried"*). When the doctor appears reluctant to grant the request for a shot (lines 23, *Ani mwe i-cwu-na twey-ssney:,* "No, it's been two weeks already"), the patient builds a turn without a main clause to appeal her case once again (line 27, *kangaci paley ancase* "(I) sat on the puppy's foot so::"). This extract displays the extensive amount of work that both parties put into both performing a request and rejecting the request in a less face threatening manner by not being direct.

Example 3. KU#3\_42


**By a puppy**? 05 PAT: 예. yey. yes:POL **Yes.** 06 DOC: 예 근데? yey kuntey? Yes so **Yes so?** 07 PAT: 강아지한테 며칠 전에 물렸는데요? kanagaci-hantey myech-il ceney mwullye-ss-nuntey-yo? puppy-from several-day before bit-PST-but-POL **(I) got bit by a puppy a few days ago?** 08 DOC: 음[:.] **mm[:.]**  09 PAT: -> [그:] 그 주사를 맞아야 되나.= [ku:] ku cwusa-lul mac-aya toy-na.= DM DM shot-ACC get-should IMP-INTERR  **-> [the:] (should I) get a shot or not.=** 10 DOC: =헤파티누스? **=Hepatinus?** 11 **(0.5) (22 lines of history taking omitted regarding the name of the shot)** 23 DOC: 아니 뭐 이주나 됐네:. 이주 전에 그랬고, Ani mwe i-cwu-na twey-ss-ney:. i-cwu-cen-ey kelay-ss-ko, no what two-week-even been-PST-ASSIL two-week-ago so-PST-CONN **No what (it's) been two weeks:. That happened two weeks ago and,** 24 PAT: 또 어제요. tto ecey-yo. Another yesterday-POL **The other (case) was yesterday.** 25 DOC: 어제? 어디. ecey? eti. yesterday where **Yesterday? Where.** 26 PAT: .hhh @아하@ .**hhh @aha@ ((stands up and sits down again))** 27 -> 침대가 있는데 강아지 발에 앉아서::, chimtay-ka iss-nuntey kangaci pal-ey anca-se::, Bed-TOP be-but puppy food-at sit.down-so **There's a bed and (I) sat on the puppy's food so::,**


Through repetition and main clause omission the patient indirectly insists that she wants to get a shot which constitutes a FTA. The doctor on the other hand indirectly rejects the need for a shot by asking for clarification and emphasizing the lapse of time since the incident. Although the patient reiterates that she was bit by a puppy twice in lines 01-03 and line 07, the doctor does not immediately register this as a legitimate problem. The doctor performs facework by declining to acknowledge the doctorability of the presented problem (Robinson & Heritage, 2002) by 1) employing a repair initiation (line 04, *kangacihantey?,* "By a puppy?') and 2) requesting for clarification (line 06, *yey kuntey* "Yes so?"). When her description of the problem fails to receive the desired outcome, the patient first builds her request for a (rabies) vaccine shot in line 09 using a subordinate clause ending with the connective -*toyna* IMP-INTERR*.* The doctor reassures the patient that if no symptoms had presented in the two weeks since the bite, the patient probably did not have to worry, that is, the doctor indicates that the proposed shot is unnecessary without directly rejecting the proposal. She employs the sentence-initial particle *ani* "no" and emphasizes the amount of time passed (line 23, "No what (it's) been two weeks:") which resists the term of the prior request (Kim, 2015). In the light of rejection, the patient provides another case, saying that she was bitten by the dog the previous day as well. When building this turn, the patient does not mention the consequences of this action, which is hinted at by the omitted main clause. The use of the connective *-se::,* "so:::" at the end of this turn (line 27) and the sound stretch hints at the unarticulated consequence of being bitten by her dog. What is omitted may be twofold: "the puppy bit me" and "I want a shot." This layered intended message makes her main clause omission more complicated. This complexity is constructed because her request was already repeated and rejected earlier but she still wants a shot and at the same time has to appear polite. The doctor also works hard to make her rejection less face threatening by employing a variety of interactional practices. Disagreeing actions such as rejections are dispreferred as it can threaten the hearer's face (Karafoti, 2021). It appears that the doctor in this example employs clarification practices which asks for the the agent of the episode (line 04, *kangacihantey? "*by a puppy?" and line 28, *cakiney kangaci?*  "your puppy?"), instead of directly acknowledging the patient's concern. This enables the doctor to perform the FTA without employing a dispreferred action (i.e., rejecting a request). In this context, the request for clarification functions as a pre-sequence leading to the actual rejection of the patient's request (Schegloff, 2007) by creating an opportunity for the patient to retract or mitigate her request before it can be openly rejected.

The amount of work the patient and doctor puts into making and rejecting a request less face-threatening is noteworthy in this example. It shows how facework and preference organization are intertwined actions with both parties orienting to requests and rejections of requests, which are both dispreferred actions, as requiring extensive interactional work. While the patient wishes for the doctor to accept her unfinished statements (via main clause omissions) to be accepted as a request, the doctor avoids rejecting the request by asking for clarification so that the actual request may be delivered in attenuated form. Main clause omission is used as a repair action (line 27), while also saving face, after the patient's unsuccessful first attempt at receiving the desired treatment plan. Both requests are unsuccessful, however, and the patient ultimately fails to receive the desired shot.

The following example also displays an unsuccessful request from the patient delivered via main clause omission. Similar to the prior example, rather than directly rejecting the request, the doctor performs face saving work by using non-verbal cues, accounts and hesitation markers that show the dispreferred nature of his own actions. The patient was visiting a family medical department at a university hospital to complain of a skin rash that had (re)appeared on his hand. The following example was taken from the history taking phase of the medical encounter in which the doctor asks questions related to the symptoms/problems that the patient have earlier presented. The patient was indirectly advocating for an antibiotic shot by first appealing to a non-medical contingency – he had a (knee) surgery scheduled on the same day so it was convenient for him to make the quick visit to the family medical department for a shot before the problem worsened (lines 66–67). In the following, the lines containing the patient's main clause omission based requests are marked with arrows (lines 66–67).

Example 4. CU\_#5\_63


```
68 지금. 주사 한대 맞을까 하고(hhhh).
          cikum. cwusa han-tay mac-ul-kka ha-ko(hhhh).
          now shot one-shot get-ACC-INTERR do-CONN
          now. (I thought I might as well) get one shot and (hhh). 
69 DOC: mm.
70 (0.2)
71 DOC: .hhh 이- 이런 경우: 이런 식으로 도진 거는::,
          .hhh i- ile-n kyengwu: ile-n sik-ulo toci-n-ke-nu::n,
           this- this-ATTR case this-ATTR type-INS reappear-ATTR-thing-TOP
72 .hhh 원인을 찾기는 쫌 어려워요. 실제로는.=
          .hhh wenin-ul chatki-nun ccom elyewe-yo. Silceylo-nun.=
           reason-ACC find-TOP a.bit difficult-POL reality-TOP
          .hhh This- cases like this: when (rashes) reappear like this::
          .hhh (It's) a bit difficult to find the reason. In reality.=
73 PAT: =이: 좀 더 전에는 쫌 많이 났을 때는
          =i: com te ceney-nun ccom manhi na-ss-ul ttay-nun 
          This a.bit DC before-TOP a.bit many appear-PST-ACC-when-TOP
74 이게 가려워가지고? [이게=
          i-key kalyewo-kaciko? [i-key=
          This-thing itchy-so this.thing
          =this- um a while back when (this) was much bigger this
          was itchy so? [this=
75 DOC: [mm.
```
In this particular extract, the patient produces the request after appealing to the contingency of the visit. Therefore, the location of this request and the preference statement (for an antibiotic shot) had occurred before the physician's treatment recommendation, which made it more problematic in terms of the sequential context. In general, a patient rejects a treatment recommendation after one is provided by a doctor rather than proposing their own treatment plan in this early location (Stivers, 2002). A patient's request for antibiotics provides grounds for extensive facework to occur as such requests can threaten the hearer's negative face (Brown & Levinson, 1999) and indicate that the patient is not willing to avoid impeding the doctor's freedom of action. In this type of situation, dispreferred actions such as requests or rejections are generally delayed (Pomerantz, 1984). Advocating for antibiotics is a dispreferred action, which patients generally do not resort to unless the doctor does not recommend antibiotics (Stivers, 2012). However, in this example, the patient uses a variety of practices to request an antibiotic shot before the delivery of the treatment recommendation. First, he refers to the visit as being motivated by a surgery scheduled on the same day and being a contingent one rather than being the main or sole purpose (line 66, *sikan-i ccokum naki-l-la::y* "I had a bit time left (before the surgery) so::"). Second, he presents the onset of a more serious problem (line 67, *talen-tey tto sayngkye-se* "It appeared in another place again so"), which hints at the potential of the rash spreading to other

areas. Both of these turns omit the main clauses *cwusa macule wasseyo* ("that's why (I) visited for a shot"). When there is no uptake after these hints, the patient produces the actual request accompanied with laughter which shows the delicate nature of his request (line 68, *cikum cwusa han-tay mac-ul-kka ha-ko(hhhh).* "(now (I might as well) get one shot and (hhh)"). The turns with main clause omissions are incrementally produced to provide multiple opportunities for the doctor to enter the patient's turn and provide a granting response. The mention of *cwusa* "shot" in line 67 only occurs after an appeal to consider their previous experiences (being treated by antibiotic shots before) and non-medical contingencies. In both examples 3 and 4, the patients first build their requests indirectly using main clause omission and only when these turns fail to receive uptake from the doctor, do they produce the actual request ("I want a shot").

Because the initial accounts (lines 63–67) are restricted to non-medical and personal matters, the turns minimize possible threats to the doctor's face by respecting the doctor's medical expertise. However, when the patient mentions "getting a shot," which appears to be his main purpose for this visit and a request that imposes upon the doctor's authority, the doctor minimally accepts the terms of the turn ("mm"), pauses, and implicates a denial/rejection by referring to the difficulties associated with finding the causes of skin rashes (lines 70–71). The doctor does facesaving work without directly rejecting the request using non-verbal cues (line 69, a pause of 0.2 seconds and line 70, an inbreath) and not addressing the request directly but instead mentioning how it is difficult to find the cause of rashes instead.

In both examples, not completing the turns (and ending with a subordinate clause and a connective) is employed as one way in which facework is accomplished by the patient. The placement of these turns without main clauses merits attention as they both occur in sequential environments that are dispreferred; during the request or after the patients fail to secure the doctors' confirmation that a shot would be prescribed. As doctors have the authority to prescribe and treat and generally have greater health knowledge, they know when a shot is appropriate or necessary. With this asymmetry present, not completing their turn enabled the patients to pursue the matter and insist without being over-insistent, and to attempt to have the terms of this matter jointly constructed by leaving what is unsaid understandable by providing enough context in the subordinate clause. The sentence final connectives contribute to this co-construction process by hinting at the main purpose of the turn without imposition. In both examples, the patients' requests were unsuccessful and the doctors employed a variety of face-saving practices during the rejection of the requests. The doctor's withholdings and the use of accounts and questions, in particular, projected their rejection/disagreement to the patients' actions.

The following two examples include cases in which the patients' indirect requests though main clause omission format are successful and the requests are accepted. In Example 5, which was taken from the final part of the same encounter in Example 4, the patient continues to persuade the doctor by articulating his understanding of the doctor's withholdings and the assessment of the situation using reported speech. The patient referred to the undesirable nature of antibiotic shots (line 112) in preparation for the upcoming request that omitted the main clause (line 113, *an maculye-to hay-ss-nuntey,* "I wasn't going to get shots but,"). However, there are several possible readings for the omitted main clause, such as "I came here because it works well" or "I want you to prescribe me shots because I came all the way here," as any grammatically compatible main clause could have been "reconstructed" by the doctor following the connective -*nuntey* in Korean (Park, 1999). However, they all reveal the delicate action of an agent performing an action of visiting a doctor to obtain a specific treatment plan. This example also exemplifies how non-verbal cues can signal whether the patient's request has been accepted or not.

#### Example 5. CU\_#5


117 **\*(9.0) ((The doctor types on the computer and takes out his notes))**


```
e::: ney. ehem
DM yes
Okay:: yes. ehem ((throat clear))
```
124 **(60.0) ((The doctor engages in a range of embodied actions-he searches his notes and types into the computer and then writes on his chart))**

In the extract, the doctor's rejection of the attempt to negotiate the terms of the need for an antibiotic shot (see extract 4 above) results in the patient producing line 112 ("doctors also said that it's not good to get shots like this"). The patient's use of reported speech indicates his understanding of the doctor's withholding. By citing the words of other doctors, the patient assumes the role of a reasonable person reporting what a doctor would be likely to say. He treats his request for antibiotic shots as undesirable and in the doctor's hands while communicating his position as strongly favoring antibiotics. The sentence final *-ntey* "but" (line 113) indicates that what was unsaid is retrievable as a contrastive action to what the doctor had said (Park, 1999). Therefore, the patient orients to the doctor as the authority with greater knowledge regarding the consequences of taking antibiotics but at the same time foregrounds their identity as a patient who knows their own pain. The target turn (line 113) includes an unarticulated main clause that hints at the undesirable action ("visiting the doctor for an antibiotic shot") and may have accounted for the doctor's withholdings concerning a prescription.

It is noteworthy that immediately after the patient acknowledges the inadequacy of receiving antibiotic shots, the doctor looks up at the patient and establishes mutual eye gaze (line 114). This gaze is followed by laughter which provides a non-verbal cue that he is going to accept the request even though he considers it problematic. Acknowledging the reluctancy, the patient produces mutual laughter in overlap. Both parties orient to the request as an FTA and a dispreferred action by employing different types of laughter throughout the interaction during both request making and accepting. Through nonverbal actions, the doctor appears to concede and confirm his understanding of the import of the prior turn. By employing pauses, averting their gazes (not shown here), and explaining that ointments should be enough to treat skin rashes, the doctor has indicated that the patient's request (i.e., requesting for antibiotic shots) is problematic. However, after the patient's turn and establishment of mutual eye gaze (line 114), the doctor aligns with the patient by laughing. The doctor and patient's short laughter (hhh) (lines 115-116) conveys their orientation to the delicacy of the patient's action (Haakana, 2001). Laughter indicates both that the request the patient fell short of producing in a full sentence was indeed problematic and that the patient has been successful in persuading the doctor to accept the request. Ultimately, the doctor proceeds to fill out the request for the antibiotic shots. After a nine-second pause during which the doctor engages with the computer and notes (line 117), the doctor asks questions related to the possible side effects of the shot (i.e., drowsiness), after which the interaction closes. Throughout the entire visit, there is no direct patient request for an antibiotic shot, such as "Can you prescribe antibiotic shots for me?," nor is there any openly articulated disagreement/rejection of this request by the doctor. Instead, the patient manages the divergence primarily by building requests using turns that omit the main clause (e.g., "I had a bit of time so::"). This linguistic resource enables the patient to orient toward and hint at a request while holding back from fully articulating it. It could be surmised that these turns put pressure on the doctor to prescribe the shots being indirectly asked for by the patient without threatening either of the participants' face. Research has shown that when doctors perceive a patient to expect antibiotic medication, they are more likely to inappropriately prescribe antibiotics (Mangione-Smith, McGlynn, Elliott, Krogstad, & Brook, 1999). As building requests may be face threatening, these examples have shown one way in which patients can deliver their wants and save face at the same time. However, after these practices, doctors can pretend that a request has not been performed by engaging in a different course of action in the following turns (e.g., asking clarification questions, providing an account).

Patients also build turns that do not have a main clause in requests that do not concern a specific prescription. In the following excerpt, a patient has visited a family medical department to request an obstetrics referral. At line 08, she builds a turn that is missing a main clause as a request for a transfer (*sanpwuinkwa ccokulo kassemyen cohkeyssnuntey::* "(I) want to go to the obstetrics but::"). Although a request in the earlier part of the medical encounter is not common and constitutes a dispreferred action, here the request is immediately granted by the doctor during the problem presentation phase.

Example 6. SS1\_C10(10:26:13) obstetrics

```
01 DOC: 예. 어떻게 오셨어요 오늘.
         yey. ettehkey o-sy-ess-eyo onul. 
         Yes how come-HON-PST-POL today
         Yes. what brought you in today.
02 (1.0)
03 PAT: 여- 갱년기가 오구요::,
         ye- kayngnwenki-ka o-kwu-yo::, 
         Here menopause-NOM come-CONN-POL
         Here- (I) had menopause::,
04 DOC: 예.
         yey.
         Yes:POL
         Yes.
05 PAT: 저기 (0.2) 화기가 막 나구
         ceki (0.2) hwaki-ka mak na-kwu 
          DM hotflask-NOM DM appear-CONN
         There (0.2-moves arm toward face) hot flashes occurred and
06 DOC: 예.=
         yey.=
         Yes:POL
         Yes.=
07 PAT: 더워갖고:: 막 어- 이것때문에
         =tewe-kacko:: mak e- i-ket-ttaymwuney 
          hot-CONN. DM. um thin-thing-because
         =(I) get so hot:: um so because of this
08 -> 어떻게- 산부인과 쪽으로 갔으면 좋겠는데::
         ettehkey- sanpwuinkwa ccokulo ka-ss-emyen cohkey-ss-nuntey::
         how obstetrics to go-PST-if like-PST-but
         How- (I) want to go to the obstetrics but::
```

```
09 DOC: 네::.
          ne::y.
          Yes:POL
          Yes::.
```
10 **(1.0) ((writes down on chart))**

After explaining the circumstances concerning menopause and hot flashes (lines 03—07), the patient produces a turn without a main clause in line 08 ("(I) want to go to the obstetrics but::"). Unlike in the previous cases, the doctor accepts this request with an immediate *ne::y* "yes::" and begins to ask history taking questions possibly because the patient's request is well warranted based on the background information that she had provided. It is worth noting that in both Examples 5 and 6, the patient builds their requests using main clause omission ending with a sentence final *-ntey* and both are successful (i.e,. the doctor grants the request). According to Park (1999), the Korean final suffix *-ntey* without main clauses can frame interactionally delicate actions such as requests and allow the speakers to avoid explicitly stating their intentions. It is described as a practice that speakers employ to appear indirect by inviting the interlocutor to infer the speaker's intention. *-ntey* may be distinguished from other Korean sentence endings such as  *yo* or -*e* which does not convey an invitation to infer the speaker's projected action. Not completing the main clause and ending with turn final *-ntey* is a resource that enables the patient to be polite while orienting to the possible trouble that is associated with producing a request for a transfer rather than asking for the doctor's help. She establishes the background for her request by telling her doctor that she wants to go to the obstetrics because she is experiencing hot flashes (lines 7-8). The patient performs the task of making the request by stating her preference or wish. Therefore, the deployment of the main clause omission may be viewed as an attempt to preserve face, enabling the patient not to mention the agent who can grant the request and leaving it for the doctor to interpret the main import by providing an accountability relevant point (Park, 1999).

In sum, in addition to the use of hedges or nonverbal resources (e.g., laughter, looking away) that encode dispreferred actions, the main clause omission can be seen as a way in which patients orient to trouble and engage in negative politeness that does not impinge on the doctor's authority. These turns tend to occur in problematic or delicate medical encounter talk environments, such as patients requesting specific prescriptions or treatment plans. Since main clause omissions concern an unarticulated action and an avoidance mechanism, this practice may be effectively accounted for by using the notions of face and facework (Goffman, 1967). Interestingly, there were no cases in the data set in which main clause omissions were employed by the doctor when making requests (i.e., directing the patient to the examination room, requesting the patient to change life styles). This skewedness in numbers suggests that doctors may not engage in facework as much as patients do when performing FTAs. This is in line with prior research that evinced a systematic pattern of asymmetry in doctor-patient interaction. For example, analysis of recordings of doctorpatient interaction have shown that doctors ask more questions that patients (West, 1984), interrupt patients more than the reverse (West, 1984), decide which topics are relevant to the consultation (Davis, 1988) and so on (see ten Have, 1991 and Heritage & Maynard, 2006 for further discussion on this topic). The lack of main clause omission employed by doctors may be due to the epistemic imbalance or a socially sanctioned knowledge asymmetry characterizing doctor-patient interaction (Heritage, 2012; Pilnick & Dingwall, 2011). However, this does not mean that doctors do not engage in face work during their interaction with patients. Examples 3-4 showed how doctors may engage in politeness practices such as being indirect or providing an account when rejecting

requests made by patients. In sum, the analysis has shown how FTAs are collaboratively performed in doctor patient interaction by focusing on the use of main clause omissions in patient requests and their responses.

#### **5 Discussion and Conclusion**

This study examined how face was constructed and negotiated by patients and doctors as discursive practices in real-time Korean primary care interactions by focusing on patient's request speech acts conveyed through main clause omissions. The findings illustrated that the omission of main clauses displayed an orientation to the notion of face and facework by hinting at a problematic action rather than fully articulating the request. Similar to the politeness strategy 'hedging' described in the politeness literature (Carter & McCarthy, 2006), patients used main clause omission to avoid sounding too assertive and to weaken his/her commitment to the request. Politeness has often been associated with indirectness and since omitting the main clause are used to hint at an action that is delicate or troublesome, it may constitute one form of practice that patients use to portray that a dispreferred action is being performed in an indirect way. Furthermore, the patients' use of main clause omission preserved the doctor's face and respected the doctors' authority as the deliverer of diagnoses and treatment plans by being indirect. The avoidance of an outright request provided a sequential context for patients and doctors to collaborate in establishing the consequences of the delicate action. Patients were able to hint at a request being performed by ending their turn with a subordinate clause and the doctor could either choose to address or ignore that a request action had been performed in the next turns. The results are in line with CA research that showed how asymmetry or epistemic imbalance is co-constructed by both patients and physicians (Peräkylä, 2006; Stivers, 2002). The role of non-verbal resources in requests and accepting requests has also been contemplated concerning facework in several of the extracts.

In extracts 4 and 5, the patient built a turn without a main clause to request antibiotic shots. By designing the turn as unfinished and leaving it pending at the point where the intention of the speaker could be formulated, the patient left it to the doctor to formulate exactly what the consequences were (i.e., prescribing antibiotic shots) without being too imposing. Avoidance is a commonly used mechanism to manage delicacy/trouble and save face (similar to off record FTAs). In extract 3, the turns ended without the main clause, which hinted at both the doctor's authority to prescribe and an unarticulated concern of being bitten by a puppy after sitting on its foot. When a request did not concern an antibiotic shot or prescription, as in extract 6, the patient provided the problems/concerns before articulating their request without the main clause. In all of these extracts, the delicate or problematic business was indirectly handled through the subordinate clause construction. In this context, the hinting (request for shots or request for a transfer) enabled an understanding of the unarticulated items to be the joint work of the participants.

What was interesting about the target forms was that enough was produced to project a recognizable action. The participants knew what the turn embodied even though the turn was only produced halfway. At the same time, not completing the turn enabled the speaker to hold back, to hint rather than fully articulate. By omitting the main clause, the patients signaled an orientation to trouble and managed the potential threats to face; therefore, this practice could be viewed as one way the interactants engaged in facework. If politeness is equated with facework, the main clause omission may be viewed as a politeness mechanism, that is, as one way the participants were exercising politeness. However, these turns did not inherently index politeness or face preservation. Rather, the analysis indicated that only an interactional and sequential account permitted a contextualized understanding of what the main clause omission was seeking to accomplish, which reflects a shift away from both Goffman's notion of face as a motivation for interaction and Brown and Levinson's concept of face as being individualistic (Brown & Levinson, 1987) and displayed a shift to a conceptualization of face as being interactional (Schegloff, 1998). Following Karofoti (2016, p. 13), polite behavior is defined as an element of social order that may penetrate in the here-and-now of a conversation. In the present paper, details of talk displayed how participants themselves orient to structural preferences, recognizing the relevance of their social relationship at these points in interaction. In other words, instead of treating main clause omission mechanistically, which identify main clause omission as polite or face saving, the participants' orientations to the normative character of both politeness and main clause omission was pursued.

The analysis also evinced the importance of the two inter-related analytical notions – "incrementality" and "sequentiality" in understanding politeness in interaction (Kàdàr & Haugh, 2013). Incrementality refers to the way in which speakers adjust or modify their talk in the light of how the progressive uttering of units of talk is received by other participants. Examples 3 and 4 showed patients producing their requests incrementally in light of the doctors' rejection, with each turn subject to ongoing evaluations as they were being produced and adjusted accordingly in real time. Sequentiality refers to the way in which current turns or utterances are always understood relative to prior and subsequent talk. Next turns are a critical resource for participants in reaching intersubjectivity, including understandings of each other's actions. When examining understandings of politeness, next turns provide us with a record of "publicly displayed and continuously up-dated intersubjective understanding" (Heritage, 1984, p. 259). In the examples shown, the patients formulated a request using main clause omission and showed low entitlement to having the request granted. The requests were structured as a dispreferred action through hesitations (e.g., mm), hedges (e.g., I mean), qualifications (e.g., doctors said antibiotics are not good), and laughter. The patients progressively build their request using main clause omissions and sequentially structured their request following the doctors' (dis)preferred next turn actions.

This study was limited as it only discussed instances from collected medical encounters located in large cities in Korea. Future studies could examine the same practice as employed in ordinary conversations among friends and family members in the Korean language to determine whether main clause omission is commonly used when performing a delicate or face threatening action across interactional contexts (cf. Park, 1999). The analysis also focused on syntactic resources and their relationship to facework and did not engage in a full analysis of the use of the nonverbal actions and paralinguistic resources, which could be very important in understanding politeness practices as observed in a number of examples in this chapter. The relevance of laughter and eye gaze when indirectly rejecting the request or when involved in face-saving work have been noted in Extracts 4 and 5. Non-verbal modes include gesture (i.e., an expressive movement that has a clear boundary of onset but does not result in any sustained change of position), facial expressions (i.e., movements of parts of the face), gaze (i.e., the organization, direction and intensity of looking), proxemics (i.e., the distance that individuals take up with respect to others and relevant objects) and posture (i.e., the way participants position their bodies) (Norris, 2004). In example 4, for instance, the doctor averts his gaze from the patient, moves his chair away from the patient, and gazes down at his notes while the patient utters his request for antibiotic shots (lines 63-70) which may show his orientation to the patient's possibly impolite action. Finally, in future studies, the role of epistemic imbalance could be contemplated in the construction of these sequences (Heritage, 2012) using a wider sample of conversations as it is possible that not finishing one's turn may also be closely related to how much knowledge or authority participants holds in the interaction in addition to performing facework.


#### **Appendix A. Abbreviations**

#### **Appendix B. Transcription Conventions**


#### **References**

Atkinson, J. M. & Drew, P. (1979). *Order in court*. Oxford Socio-legal studies. Atlantic Highlands: NJ.

Austin, J. L. (1962). *How to do things with words.* Harvard University Press, Cambridge: MA.

Blum-Kulka, S. & Olshtain, E. (1984). Requests and apologies: A cross-cultural study of speech act realization patterns (CCSARP). *Applied Linguistics*, *5*(3), 196–213.

Brown, P. & Levinson, S. (1987). *Politeness: Some universals in language usage*. Cambridge University Press, Cambridge: New York.

Byon, A. S. (2006). The role of linguistic indirectness and honorifics in achieving linguistic politeness in Korean requests. *Journal of Politeness Research*, *2*, 247–276.

Chevalier, F. H. G. (2009). The facework of unfinished turns in French conversation, *Discourse Studies*, *11*(3), 267–284.

Cordella, M. (2007). No, no, I haven't been taking it doctor: noncompliance, face-saving, and face-threatening acts in medical consultations**.** In

M.E. Placencia, C. Garcia (Eds.), *Research on politeness in the Spanish speaking world* (pp. 191–212). Lawrence Erlbaum, Mahwah, NJ.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Challenges in Managing Korean Online Service Requests and Complaints via Business Chat**

#### **Mary Shin Kim, Sujin Kang, and Tyler Miyashiro<sup>1</sup>**

**Abstract** This study examines how service providers and their business clients in South Korea manage and negotiate face-threatening encounters when making and receiving requests through online chats. An analysis of a collection of these online chats illustrates that unsatisfied communications lead to repeated requests and complaints from clients. The research will focus on examining how service providers adapt to these repeated requests by (re)formulating their response turns to satisfy clients. The study will reveal how their response turns differ by agency, certainty, and immediacy in executing the request, and how certain response types are preferred by clients. According to the findings, what is considered appropriate or polite in person may not be the same in online business-to-business communications.

**Keywords** Agency Client Complaint Compliance Online chat Request Service provider

1 M. S. Kim ()

 Department of East Asian Languages and Literatures, University of Hawaiʻi at Mānoa, Honolulu, Hawaiʻi, USA. e-mail: maryskim@hawaii.edu

 S. Kang The Prompt Company, Seoul, South Korea e-mail: sujin@theprompt.tech

 T. Miyashiro Department of East Asian Languages and Literatures, University of Hawaiʻi at Mānoa, Honolulu, Hawaiʻi, USA. e-mail: tylerbm@hawaii.edu

#### **1 Introduction**

Transactions and interactions through online apps and chats have become a daily routine for many Koreans, with food ordering and delivery being particularly prevalent. The online food service industry has undergone rapid expansion in recent years. According to the National Bureau of Statistics in South Korea, online food service transactions surged to 26.4 trillion won in 2023, a substantial rise from 2.7 trillion won in 2017. There is a need for empirical research drawing on these interactions to provide new insights on Korean (im)politeness. This study focuses on online interaction between agents of a delivery application company and their clients, restaurant or shop owners. As food orders require quick delivery service, with delays, clients will contact companies with complaints about slow deliveries. This research investigates how clients and service agents navigate and resolve such face-threatening situations during the course of making and receiving requests through online chat interfaces.

This research is in line with growing research on Korean (im)politeness from multidisciplinary and multimodal perspectives across a broad range of different interactional contexts and communication platforms. Korean speakers construct, negotiate, and utilize politeness or impoliteness as discursive practices during daily interaction, online and offline. As we communicate through an increasing number of platforms on a daily basis, recent studies examine these real situations by not only looking at face-to-face interaction, but also interaction through instant text messages, chat boxes in livestreaming sites, social media account messages, as well as other communication channels.

Studies dealing specifically with particular contexts of business negotiations and/or transactions are still rare. Despite the limited number of studies, they provide important insights in understanding complaints across various settings and languages. Vásquez (2011) analyzed online complaints in English on TripAdvisor concerning negative hotel reviews and discovered that most complaints were indirect or third-party, although some complaints blurred the direct/indirect dichotomy. In contrast to prior investigations, which revealed a tendency for complaints to coincide with speech acts like warnings or threats, in this study, complaints were found to more frequently accompany advice and recommendations. On the other hand, studies about complaints concerning the provision of customer services, such as Orthaber and Marquez-Reiter (2011), have made noteworthy findings about the sheer directness and almost confrontational nature of complaints. Their analysis of business-to-consumer service calls to a Slovenian public transport company discovered that complaints were often made explicit from the very beginning, often incorporating intensifiers that escalated utterances into outright criticisms, insults, and even threats. Once the facts and details of the complaints have been deliberated, the service agents conclude the calls by either providing or not providing a solution. Throughout the complaint interactions, both the complainants' and agents' faces are found to be at risk. Another study by Stalpers (1995) that examined oral business negotiations in French and Dutch found comparatively fewer mitigators in comparison to casual conversations, which potentially reflected a lesser concern for personal feelings and/or perceptions. This overall tendency toward more direct forms of complaints and disagreement has been attributed to transactional goals that are more prominent in business interactions, which also carries different face considerations in comparison to more informal contexts of interaction. In general, a type of "professional face" (Charles, 1996) becomes prioritized in which people's transactional roles (e.g., customer, employee, service provider, etc.) dictate behavioral expectations that also warrant more explicit and clear language.

Furthermore, Decock and Spiessens (2017) examined a multilingual business context. Their investigation on French and German business-to-business email exchanges for the sales department of a Belgian international company made interesting observations. The customer's language shifts from initially neutral, issue-focused, and standardized expressions in initial complaints to increasingly confrontational, individual-focused, and spontaneous expressions in rebuttal emails following complaint rejections. The study also noted a difference between German and French customers. Germans tend to be more direct, whereas the French exhibit a more assertive approach. The study underscores the need for a reassessment of our evaluation of directness and underscores the pivotal role of cultural comprehension in business practices.

Studies of Korean online business communications are even more scarce. One exception is Kim and Brown's (2019) research examining text messages between an online food business vendor and a customer on a blog. In contrast to studies which focus on customers' complaints and/or the language of complainers, the authors investigated how the recipient of the complaints, the business vendor who owns the blog, perceives and agentively evaluates customers' language from a (im)politeness perspective.

Building upon prior research on business transactions and online interactions, the current study will investigate how service providers respond to business clients' requests and complaints. Recent studies have contributed to understanding how Koreans formulate here-andnow requests and complaints in ordinary conversations by utilizing particular linguistic choices, such as *wh*-interrogatives or imperatives (M. S. Kim 2020; S. H. Kim & M. S. Kim 2020; Yoon 2006). The responses to these requests and complaints have not been given enough attention, particularly in the context of business transactions and online interactions. No matter how many times clients make their complaints or repeat their requests, service providers have the obligation and goal to fulfill their clients' demands. The receivers of these requests and complaints (service agents) do not respond with countercomplaints, disagreements or denials; rather, they seek to resolve matters and appease their clients.

A collection of these online chats reveals how agents' responses can be seen in a variety of patterns which consequently results in an array of possible sequential developments that involve recurrent exchanges between repeated requests and complying responses. Depending on how two parties respond to each other, the initial request can expand and multiply with added frustration and criticism. Such extended sequences between agents and clients provide us an opportunity to examine beyond the initial pair of request and response, leading to third, fourth, and even seventh pairs with different turn designs and strategies. Overall, the study will reveal how service agents adapt to their clients' demands and make strategic choices or shifts of choice during their turn-by-turn online interaction.

#### **2 Data and Methodology**

This research deploys the method and framework of conversation analysis (CA) in analyzing online chat interactions. The synchronous aspect of live online chat mirrors how people talk, even though online chat is delivered in a written format. Time stamps aid researchers in understanding when the interaction occurs synchronously or with a delay in time. Through CA, researchers can investigate the intricate interplay between language and interaction: how speakers deploy certain (non-)linguistic resources to formulate particular actions, such as requests, complaints, compliances (action formation) and how they are understood by the

recipients in interaction (action ascription). CA takes an emic perspective which helps us observe how the speakers' (non-)linguistic choices or shifts of choice are made relevant in the unfolding turn. The CA perspective shows how actions are not constructed in isolation but reflexively formulated and oriented to by the recipient. What speakers actually do in interactions is made observable in CA through detailed transcription practices and the analysis of data (Clift, 2016; Schegloff, 2007).

The data for this study consists of a collection of 65 online chats between a delivery application company called *Kanata* (pseudonym) and restaurant or shop owners that are their clients.<sup>2</sup> The interactions are transcribed in four tiers consisting of (a) the original utterance in Korean, (b) its corresponding romanization, (c) morpheme-by-morpheme gloss, and (d) English idiomatic translation. The transcript shows agents' and clients' own spelling, spacing, and use of emoticons. The time stamps are also marked to help us understand the temporal gap between the turns. The restaurants or food businesses are identified as clients (C) while Kanata's chat operators are referred to as agents (A) who respond to the clients' queries and requests through live chat.

Before analyzing the online interactions, we need to briefly explain the structure of food delivery service in Korea. The structure and the different stages involved in delivery ordering provide an understanding of the roles, relationship, and interactions between the delivery application company and their clients — restaurants or food businesses (e.g., bakery, cafes, fast food chains).

Overall structure of online food ordering and delivery:


Kanata creates and provides the application platform used by restaurants and regional delivery companies. The app company works with the delivery companies who actually recruit and manage drivers in each region. In online chats, delivery companies are referred to as dispatch offices or dispatch hubs. The app company also directly deals with restaurants and other food businesses by assisting in uses of the app service and aiding or addressing delivery issues. The online chats are initiated by the restaurants or food-related businesses who request that Kanata resolve delivery issues or delays caused by delivery companies or drivers.

#### **3 Formulating Requests and Compliances**

The way the clients formulate requests are by and large consistent in terms of their composition and position across the collection of chats. They make an explicit request for a quick delivery dispatch (Excerpt 1) or just provide the reason for their contact, such as trouble with the delivery (Excerpt 2). In essence, clients indicate their need for assistance from the agent. As they are

<sup>2</sup> The names and places of the business have been anonymized.

often in a hurry to make the customers' delivery on time, clients sometimes rush to convey their message to the agents even before the agents can start the chat with their self-introduction (Excerpts 1 and 2). We will first examine an excerpt in which the client's request is successfully fulfilled within a relatively short period of time that results in a brief chat interaction (approximately one minute between 16:48 to 16:49, as shown in the time stamp).

Excerpt 1


YEAH NOT.EXIST-DEF:DC **'Yeah, (we) don't have any more.'**

As soon as agents identify themselves (line 3), they respond to the client's request with a compliance, which is to send a request to the dispatch office (line 4). The agent shortly after confirms the dispatch to the client (line 5), and the client displays appreciation for the agent's help (line 6). As their transaction reaches closure, the agent initiates closing the chat by confirming whether the client does not have any further inquiries (lines 7–9). Such a short interaction is uncommon to find in the data collection.

Excerpt 2 shows another example of a successful request and compliance transaction between the client and agent, but there is a hint of a complaint in this interaction. The chat begins with the client's report of delivery trouble (line 1).

Excerpt 2


swuhayng cisa-lo paycha yocheng centalha-keyss-supnita. DISPATCH OFFICE-TO DISPATCH REQUEST SEND-WILL-DEF:DC

#### **'We will send the dispatch request to the dispatch office.'**


As in Excerpt 1, the client in Excerpt 2 makes a negative observation that something expected did not happen (Schegloff 1988). Compared to Excerpt 1, here the negative observation in Excerpt 2 is accompanied with an explicit negative stance, highlighting the excessiveness of the delay, over 30 minutes, (lines 1, 3, 6) while displaying the client's affective stance through the crying emoticon ㅠ (line 6). As Schegloff (1988) notes, negative observations often accomplish the work of complaining. In response to this complaint, the agent quickly acknowledges the delay (line 4), and immediately responds with a promise to make the dispatch request (lines 5, 7). Despite the client's initial complaining stance, the client finds the agent's responses satisfactory and displays gratitude (line 8).

These two chat examples are uncommon cases found in the data collection. Most chats are prolonged and not smooth, as disalignments and/or disaffiliations emerge and require more negotiation on both ends to manage potentially face-threatening situations. We will examine more challenging cases in Section 4.

#### **4 Managing Clients' Repeated Requests and Demands**

One of the recurrent patterns we observe in troublesome online interactions is when a client repeats a request throughout the interaction.<sup>3</sup> The client modifies the request turns to display the urgency of the request to accomplish the goal of getting the agent to take actions promptly. In response to repeated requests, what we see is that the agent redesigns the responses sensitive to the shifting demands or requests from the client.

In this section, we will examine excerpts which show sequential developments involving recurrent exchanges between repeated requests and complying responses. In Excerpt 3, after the agent offers to help, the client then requests quick delivery dispatch for two orders from different locations (lines 1-6).

Excerpt 3

01 A: 가나다 컨택센터입니다. 무엇을 도와드릴까요? 19:28 kanata khenthaykseynthe-i-pnita. mwues-ul towatuli-lkka-yo? NAME CONTACT.CENTER-BE-DEF:DC WHAT-ACC HELP.FOR-Q-POL **'This is Kanata Contact Center. How may we help you?'** 02 C: 저희 능동건 배달시간 이미 지났는데 cehuy nungtong ken paytalsikan imi cinass-nuntey OUR AREA.NAME CASE DELIVERY.TIME ALREADY PASS-BUT 03 아직도 라이더 배차가 안돼서요 19:28 acikto laite paycha-ka antwayse-yo YET RIDER DISPATCH-NOM FAIL.SO-POL **'The delivery time for our Neungdong (delivery) case has already passed, but the delivery person has not been yet dispatched.' 04** 얼른 좀 부탁드릴게요 19:28 ellun com pwuthaktuli-lkey-yo FAST DM ASK.FAVOR-WILL-POL **05** 그리고 같이있는 신사동건도 kuliko kathi iss-nun shinsadong ken-to AND TOGETHER EXIST-RL AREA.NAME CASE-ALSO **06** 30 분 밖에 안남아서 배차 얼른 부탁드려요 19:28 30 pwun pakkey an-nama-se paycha ellun pwuthaktulye-yo 30 MIN ONLY NOT-LEFT-SO DISPATCH FAST ASK.FAVOR-POL **'Please hurry. And, please hurry up with the dispatch for the Shinsa neighborhood (delivery) case as well since there's only 30 minutes left.' 07** A: 수행 지사에 빠른 배차 요청드릴 수 있도록 하겠습니다. 19:28

swuhayng cisa-ey ppalun paycha yochengtuli-l swu isstolok ha-keyss-supnita.

<sup>3</sup> This shows a stark contrast with the previous two excerpts. In excerpt 1, the client issued the request once (line 1). In Excerpt 2, the client did not explicitly make any requests but only repeated the reason for contacting the agent (lines 1, 3) as their turn proceeded the agent's self-introduction.

DISPATCH OFFICE-TO FAST DISPATCH REQUEST-BE.POSSIBLE.TO DO-WILL-DEF:DC **'We will see to it that the request is possible to the dispatch office for a swift dispatch.'**

**08** C: 얼른 올 수 있게 지속적으로 요청 부탁드려요 19:29 ellun o-l swu isskey cisokcekulo yocheng pwuthaktulye-yo FAST COME-BE.POSSIBLE.TO CONTINOUSLY REQUEST ASK.FAVOR-POL **'Please continue to make the request so (the delivery person) can arrive quickly.'**

**09** A: 요청 드리겠습니다. 19:29 yochengtuli-keyss-supnita. REQUEST-WILL-DEF:DC **'We will make the request.'**

After the agent provides a response in compliance to the request (line 7), we expect the chat to reach its closure. However, as seen in line 8, the client immediately repeats the request. The repeated request shows that the client is not yet satisfied or convinced by the agent's response. It also exerts a sense of urgency (Mondada, 2017). Consequently, the agent is urged to respond to the repeated request again. In line 9, the agent again pledges to comply. What is worth noting is that when the agent complies the second time, the turn is designed differently. The first response type uses indirect language ('see to it that the request is possible') rather than a direct one ('will request'). In general, indirect language in Korean is considered polite (Sohn, 1999). This response does not violate any politeness standard. The response also displays the agent's intention (*keyss* 'will') to comply to the client's request.

However, the agent's first response in line 7 is designed with a construction which portrays the agent as a passive actor (요청드릴 수 있도록 하겠습니다 'We will see to it that the request is possible to the dispatch office….') who merely passes on the clients' message to the dispatch office. Additionally, as shown in the *swu iss-* 'be possible to' phrase in this construction, the agent does not promise or guarantee the delivery of the message, but states only the possibility. This ambiguous construction in line 7 contrasts with the construction used in the agent's second response in line 9. The second construction conveys a stronger sense of agency to the agent who will be directly involved with the matter (요청드리겠습니다. 'We will make the request.'). We see a recurrent pattern of how agents redesign their responses to respond to their clients' repeated requests.

Excerpt 4 exhibits a series of repeated requests and compliance responses throughout an entire interaction between a client and an agent. The client repeats the request three times (lines 2, 4, 6-7).

Excerpt 4


#### DISPATCH ASK.FAVOR-DEF:DC **'Please dispatch (a delivery vehicle).'**


**'Just a while ago the delivery vehicles were running fine, but … it seems that (the drivers) dislike coming (here) due to the far distance.** ㅠㅠ **so, we're sad.** ㅠㅠ **Please (make the request).'**


While the first request in line 2 is articulated in a simple manner ('Please dispatch (a delivery vehicle)'), the repeated requests are accompanied with more details which show the agent's

desperation and frustration (lines 4, 6–7). As the agent repeats the requests and displays more frustration, the agent also modifies the response. Similar to Excerpt 3, the agent first responds with the same construction which displays passive uncertainty (line 3, 요청드릴 수 있도록 하겠습니다 'We will see to it that the request is possible to the dispatch office….'). Then, the agent modifies the response with more agency, albeit with continued uncertainty (line 5, 요청 드려보겠습니다. 'We will try to request'). The 드려 보- 'try to' construction indicates an attempt. After a third request, the agent finally replies with stronger agency and certainty (line 8, 요청드리겠습니다. 'We will make the request.'). With this firm complying response, the client finally wraps up the request with a closing formulaic phrase, 잘 부탁드립니다 ('We would appreciate your help.'). This long sequence of repeated requests and responses closes with the agent's promise to continuously make the request (line 10).

It is interesting to note that these different types of response turns by the agent is consistent across a collection of the chat data. Excerpt 5 shows the same pattern by which the agent consistently redesigns the response from weaker to stronger agency and with a gradual increase in certainty. In this interaction, the agent repeats the complying response seven times.

Excerpt 5



**'Yes, because it's Friday peak time, all cases are delayed.'**



The following four types of responses are deployed through the entire interaction:

(a) 요청드릴 수 있도록 하겠습니다 'We will see to it that the request is possible.': line 3 (b) 요청 드려 보겠습니다. 'We will try to make the request.': lines 7, 14, 16 (c) 요청드리겠습니다. 'We will make the request.': lines 9, 18 (d) 요청드리고 있습니다. 'We are (right now) making the request.': line 20

In addition to the three (a), (b), (c) response types, the agent deploys response type (d) at the end of their interaction. The agent reassures the client that the agent is complying with the request and is concurrently making the request while chatting with the client.

The consistent pattern and order that the agents use to redesign their responses demonstrate that clients prefer type (c) and (d) over type (a) and (b) responses. The higher level of directness and certainty helps bringing the conversation to its end point. In such service interactions, clients prefer direct language which guarantees their service request will be met swiftly rather than indirect, euphemistic expressions which display an agent's diplomatic approach between the client and delivery dispatch office.

Repeated requests and responses that take a lot of time during peak delivery time hours are counterproductive because they are inefficient and time-consuming on both sides. The client's frustration can escalate into a face-threatening situation, as illustrated in the following interaction in Excerpt 6. Much like Excerpts 1 and 2, in Excerpt 6, the client is eager to initiate the conversation with a reason for contacting before the agents can even begin their selfintroduction.

#### Excerpt 6


#### **'Right now, is it the peak time?'**

10 저희 매장이 요즘 계속해서 cehuy maycang-i yocum kyeysokhayse OUR STORE-NOM LATELY CONTINUOUSLY **'Our store these days continues to**

11 유독 기사님 배정도 안되고 늦는데요 20:06 yutok kisanim payceng-to antoy-ko nuc-nuntey-yo UNIQUELY DRIVER DISPATCH-ALSO NOT.WORK-AND LATE-BUT-POL **experience difficulties with driver assignments, causing delays.'**


**'The delivery time was supposed to have been by 7:36, but not being able to assign drivers for more than an hour, isn't this excessive?'**



**We apologize for any inconvenience.'**

After the agent issues the first compliance in line 6 with the (a) type response, the client continues to reveal the difficulties of dealing with an angry customer (line 7). In response, the agent responds with a more reassuring response, type (c) in line 8. However, issues resurface as the client picks up the chat four minutes later from line 9. The inquiry posed by the client in line 9 regarding peak time and the four-minute gap suggest that the delivery has not yet been sent out. This situation prompted to voice additional complaints about the recurring delays the client has been experiencing lately (lines 9-11). In response to this complaint, the agent responds with a (b) type response, which displays uncertainty ('We will continue to try and make a request.,' line 14) and reverses the level of certainty to a lower one. This reversal of response type brings trouble in the interaction. Despite the agent's apology (line 15), the client expresses further dissatisfaction. The client points out the issues with repeated and extreme delays in delivery dispatch (i.e., hour delay) and even difficulties in communicating with the Kanata company (lines 16–21). In response to such an agitated client who reiterates a fast dispatch (line 22), the agent no longer responds with uncertainty; rather, the agent confirms that they are currently making continuous requests (line 23).

As seen in Excerpt 6, the type and order of response used by the agent is significant. Consequently, if the agent deploys the preferred type of response in the right order, the interaction is less likely to develop into an unnecessarily long confrontational chat. In Excerpt 7, as the client requests a swift dispatch even after the agent provides a complying response (lines 3–4), the agent immediately responds in the preferred format (lines 5). With this preferred response, the client does not raise any further issues and the chat closes.

#### Excerpt 7


**dispatch office.'**

Similar to Excerpt 7, in Excerpt 8 when the agent's initial compliance (line 3) is met with the client's urgent request for action (line 4), the agent responds in the preferred format (line 5). As the client informs the agent that the delivery rider has been dispatched, the agent closes their chat (lines 6-7).

#### Excerpt 8


**04** C: 배달시간 임박이어서 빠르게 부탁드려요 20:22 paytalsikan impak-i-ese ppalukey pwuthaktulye-yo DELIVERY.TIME IMPENDING-BE-SO FAST ASK.FAVOR-POL **'The delivery time is coming up soon, so please hurry.' 05** A: 네 요청드리겠습니다. 20:22 ney yochengtuli-keyss-supnita. YES REQUEST-WILL-DEF:DC **'Yes, we will make the request.'** 06 C: 됐습니다 20:22 tway-ss-supnita BECOME-PST-DEF:DC **'It's done.'** 07 A: 종료합니다 20:22 conglyoha-pnita CONCLUDE-DEF:DC **'We will conclude (the chat).'** 08 C: 감사합니다. 20:22 kamsaha-pnita. THANK.YOU-DEF:DC **'Thank you.'**

The importance of the agent's complying response type is clearly shown in these excerpts. If the agent initiates the preferred response from the outset, we can expect their interaction to reach their goals and conclude in a brief manner. Recall that in Excerpts 1 and 2, the agent responded with direct formats: Excerpt 1 line 4, 배차 요청 전달하겠습니다 'We will send the dispatch request.', Excerpt 2 line 5, 배차 요청하겠습니다 'We will request the dispatch. The preference for providing direct and clear assurances about the prompt resolution of delivery problems is prioritized over the usual practice of using subtle and indirect language, typically associated with politeness. The research results indicate that what is considered suitable or courteous in in-person interactions may vary in the context of online business-to-business communication.

#### **5 Conclusion**

The findings of this study carry important and interesting implications for several areas of study. First, it provides explicit insight into the conversational conventions of workers in online service industries that have seen exponential growth in recent years. Second, the interactions of the current study differ from previous studies which investigated communications between business vendors and customers, who buy or consume the goods (Kim & Brown 2019; Orthaber & Marquez-Reiter 2011; Vásquez 2011). This study specifically addresses business-to-business relationships: between food industries that create a product for consumers and a delivery application company which outsources delivery service to local dispatch offices. The application company holds a unique position of being the middleman, who must answer to its food industry

clients, while also sending the clients' requests to local delivery offices. This business-tobusiness relationship and the application company's middle position influence transactional interests, ultimately determining the language used by clients in the food industry and agents of the application service provider during online business chats.

 The linguistic choices of the aforementioned conversational interactants ultimately have notable implications on previous precepts and understandings of politeness as they relate to the preferred sequence of exchanges between co-interlocutors in an ongoing online chat. Due to the time-sensitive nature of delivery service issues, clients often incrementally increase the intensity, earnestness, and urgency of their requests when the agent's response was found to be unsatisfactory. This need for explicit reassurance that the delivery issues would be promptly addressed seems to outweigh putative expressions of restraint and indirectness that are conventionally viewed as more "polite."

 In the most extreme cases, online chats became confrontational as they unequivocally voice criticism about the inadequate handling of service delay issues. In spite of this, the receivers of these chats (agents) maintained a level of professionalism expected of someone working in a business that relies on the patronage of another (food industry). As such, the escalated requests and/or complaints made at the end of an extended sequence of turns between client and service agents serve as empirical evidence of conflicting expectations from both parties. While diplomatic and accommodating platitudes may serve as a default for the service providers, their clients who deal directly with the demands of dining consumers indicate a preference for direct and confident language that provide a sense of reassurance.

 The findings indicate that what is considered appropriate or polite may not be the same in different communication platforms. In online business-to-business chats, clients demand prompt responses and resolutions, and accordingly service providers strategically choose or shift their choice of words which display higher agency, certainty, and immediacy in executing the request. This research focuses on online interactions within the context of food delivery service complaints and requests. Expanding research efforts to encompass diverse online business interactions involving cross-linguistic and cross-platform communications has the potential to enrich our understanding of effective communication strategies and outcomes in the digital age.

#### **Acknowledgement**

This research was funded by the University of Hawaiʻi Endowment for the Humanities 2023 Summer Research Award.

#### **Abbreviations**


#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Index**

#### **A**


#### **C**


#### **D**

Digital discourse, 36, 37

## **E**


#### **F**

Facework, 6, 40, 223–226, 228–230, 232–235, 239–241 Family instant messages, 3, 35, 36, 39, 40, 48, 50 Fan discussions, 122 Fan subtitles, 125, 126 Footing, 3, 35, 37, 39, 42, 44–49 Frames, 3, 5, 39, 43, 46, 65, 71, 85, 128, 135, 141, 207, 208, 212–216, 239

#### **G**

Gestures, 29, 72, 113, 159, 160, 167, 197, 241

#### **H**

Honorifics, 1, 3, 12, 13, 17, 18, 22, 25, 36–38, 43, 48, 49, 55–62, 65, 71, 81, 84–86, 92, 123, 124, 134, 155–159, 161, 162

Honorific speech level markers, 43, 48, 49

M. S. Kim (ed.), *Exploring Korean Politeness Across Online and Offline Interactions*, Advances in (Im)politeness Studies, https://doi.org/10.1007/978-3-031-50698-7 © The Editor(s) (if applicable) and The Author(s) 2024 267

### **I**

Identity, 4, 28, 30, 35–37, 39, 47–49, 84, 91, 93–95, 100, 102, 103, 106, 110–112, 114, 118, 119, 124, 142, 143, 151, 174, 182, 195, 214, 237 Identity negotiation, 91, 94, 95, 110, 111, 119 (Im)politeness, 1, 2, 4, 11, 13, 26, 29, 40, 49, 50, 121, 122, 124–127, 129, 132, 135–137, 142, 144, 150, 151, 159, 160, 225, 226, 248, 249 Interpersonal pragmatics, 121, 122, 124, 133

#### **K**

KakaoTalk, 3, 35, 36, 41

#### **L**

Language learning, 5, 36, 126, 155–157, 159, 162, 163, 175, 177, 178 Lay translations, 135

#### **M**

Main clause omission, 6, 223–227, 229, 230, 232, 233, 235, 239–241 Medical interaction, 6, 223, 224 Meta-discussion, 125, 137 Metalinguistic knowledge, 29 Meta-pragmatic marker, 143, 149 Metapragmatics, 10, 11, 14, 15, 23, 28, 29, 125, 129 Multimodality, 29, 134, 146, 157

#### **N**

Negativity, 5, 205, 206, 208, 210–212, 214–217 News, 5, 93, 100, 163, 205–208, 210–213, 215–217, 224 Non-verbal expressions, 156, 157, 161, 172, 173

#### **O**

Online chat, 2, 6, 247–250, 265

#### **P**

Panmal (half-talk), 93

Politeness, 1, 2–6, 9–16, 18, 20–30, 33, 35–37, 39, 40, 42, 44, 47–50, 55, 58, 60, 85, 86, 91–93, 109, 122, 124, 125, 134–136, 139, 150, 155–162, 164–177, 182, 200, 201, 209, 214, 217, 223–226, 229, 230, 239–241, 248, 255, 264, 265 Politeness metaconcepts, 2, 10–13, 15, 21, 26, 30 Power and solidarity, 3, 35, 36, 37, 39, 40, 42, 45, 49, 50, 92, 118, 119

#### **Q**

Qualitative interviews, 9

#### **R**

Relational work, 4, 121, 124–127, 129–147, 150, 151 Request(s), 6, 22, 38, 44, 45, 47, 49, 60, 76, 82, 111, 139, 140, 146, 166, 186, 188, 197, 223–226, 229, 230, 232–235, 237–241, 247–265

#### **S**

Service provider, 6, 247–249, 265 Social awareness, 3, 55, 85, 86 Socialization, 3, 36, 55, 56, 59–61, 65, 71, 76, 82, 85, 86 Social responsiveness, 3, 55, 85, 86 Speech level markers, 3, 35–40, 42–50 Speech style, 1, 3, 4, 37, 91–93, 95, 100, 105–107, 109, 110, 113, 114, 117–119, 158, 181, 182, 200 Speech style shift, 95, 100, 110, 114, 118

#### **T**

Talk shows, 1, 4, 5, 91, 93–96, 98, 102, 113, 118, 181, 182, 187, 188, 192, 201 Treatment plans, 6, 223, 224, 226, 229, 230, 233–235, 239, 240

#### **V**

Viki, 4, 121–123, 125–127, 132, 134, 135, 147–151