Cogent Social Sciences ISSN: 2331-1886 (Online) Journal homepage: www.tandfonline.com/journals/oass20 ‘We mean more than what we say on air’: interactional metadiscourse resources in television talk shows Emmanuel Nii Adama Mensah, Christiana Hammond & Albert Agbesi Wornyo To cite this article: Emmanuel Nii Adama Mensah, Christiana Hammond & Albert Agbesi Wornyo (2025) ‘We mean more than what we say on air’: interactional metadiscourse resources in television talk shows, Cogent Social Sciences, 11:1, 2472917, DOI: 10.1080/23311886.2025.2472917 To link to this article: https://doi.org/10.1080/23311886.2025.2472917 © 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published online: 11 Mar 2025. Submit your article to this journal Article views: 1493 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oass20 https://www.tandfonline.com/journals/oass20?src=pdf https://www.tandfonline.com/action/showCitFormats?doi=10.1080/23311886.2025.2472917 https://doi.org/10.1080/23311886.2025.2472917 https://www.tandfonline.com/action/authorSubmission?journalCode=oass20&show=instructions&src=pdf https://www.tandfonline.com/action/authorSubmission?journalCode=oass20&show=instructions&src=pdf https://www.tandfonline.com/doi/mlt/10.1080/23311886.2025.2472917?src=pdf https://www.tandfonline.com/doi/mlt/10.1080/23311886.2025.2472917?src=pdf http://crossmark.crossref.org/dialog/?doi=10.1080/23311886.2025.2472917&domain=pdf&date_stamp=11%20Mar%202025 http://crossmark.crossref.org/dialog/?doi=10.1080/23311886.2025.2472917&domain=pdf&date_stamp=11%20Mar%202025 https://www.tandfonline.com/action/journalInformation?journalCode=oass20 Cogent Social Sciences 2025, VOL. 11, NO. 1, 2472917 ‘We mean more than what we say on air’: interactional metadiscourse resources in television talk shows Emmanuel Nii Adama Mensah , Christiana Hammond and Albert Agbesi Wornyo Department of Communication Instruction, University of Education, Winneba, Ghana ABSTRACT This study explores the interactional metadiscourse resources employed on a Ghanaian television talk show and the communicative functions of these resources in persuading the target audience of the show. Five episodes of the television talk show prior to the 2020 presidential and parliamentary elections in Ghana were gathered as data and qualitatively analyzed using Hyland’s metadiscourse theory. The findings indicate that interlocutors on the talk show employed various interactional metadiscourse resources to indicate their stance toward propositions as well as their audience. The findings further revealed that the show is a persuasive arena for social commentators and representatives of political parties to exert rhetorical appeals of logos (rationality), ethos (credibility), and pathos (affection) on the audience. This study contributes to the literature on metadiscourse resources and concludes that television talk shows are permeating avenues for socio-political interlocutors to propagate their ideologies and assert their relevance in discussing issues of importance to their audiences. This study has implications for student teaching in communication and media studies. It recommends that interlocutors of television talk shows should be intentional in their on-air communicative engagements since ‘they mean more than what they actually say’ by their use of interactional metadiscourse resources. Introduction The goal of this study is to examine how metadiscourse resources are employed in television talk shows using a Ghanaian talk show to contribute to the literature on metadiscourse resources in nonacademic discourses. As a theory and framework, metadiscourse was originally introduced by the structural linguist Zelig Harris. However, in the mid-1980s, the theory gained attention in applied linguistics through the work of Vande Kopple (Kopple, 1985), Crismore (1983), and Williams (1981). Hyland (2005) developed a model which is being employed by the majority of related research. The core focus of this theory is that language is not only concerned with exchanging varied information but also concerns or comments on itself (Ädel, 2010). In effect, language not only presents ideas or propositional material but also presents readers with materials that help them organize, interpret, and evaluate what is being said. Metadiscourse ‘embodies the idea that communication is more than just the exchange of information, goods, or ser- vices, but also involves the personalities, attitudes, and assumptions of those who are communicating’ (Hyland, 2005, p. 3). Hyland (2019) divided metadiscourse resources into two categories: interactive and interactional. Interactive metadiscourse, on the one hand, relates to the producer’s (writer or speaker) awareness of the receiver, and the producer’s attempts to accommodate the receiver’s (reader or listener) interests and needs in order to make the argument satisfactory. This relates to composing discourses in ways that reveal only what the producer deems necessary to be recovered from the text by the reader. Interactive metadiscourse resources consist of transition markers, frame markers, endophoric markers, evidential, and © 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group CONTACT Emmanuel Nii Adama Mensah 200030738@st.uew.edu.gh Department of Communication Instruction, University of Education, Winneba, Ghana https://doi.org/10.1080/23311886.2025.2472917 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. ARTICLE HISTORY Received 18 September 2024 Revised 13 December 2024 Accepted 23 February 2025 KEYWORDS Metadiscourse resources; rhetorical appeals; persuasive communication; communicative functions; talk show SUBJECTS Media & Film Studies; Interpersonal Communication; Mass Communication; Persuasion; Rhetoric MEDIA & COMMUNICATION STUDIES  |  Research Article http://orcid.org/0009-0001-2873-3615 http://orcid.org/0000-0002-1142-658X mailto:200030738@st.uew.edu.gh https://doi.org/10.1080/23311886.2025.2472917 http://creativecommons.org/licenses/by/4.0/ http://crossmark.crossref.org/dialog/?doi=10.1080/23311886.2025.2472917&domain=pdf&date_stamp=2025-3-11 2 E. N. A. MENSAH ET AL. code glosses. Interactional metadiscourse, on the other hand, relates to the producer’s attempts to make their views explicit and engage receivers by anticipating their objections and responses to the text. It deals with the producer’s attempt to regulate how much of himself or herself is revealed in the text and to foster a relationship between the receiver and the producer through the text. Interactional metadis- course resources include hedges, boosters, attitude markers, engagement markers, and self-mention. Hyland (2005, 2019) posits that hedges are markers employed in the text to conceal or withhold pro- ducers’ commitments to the propositions and allow receivers to make inferences on their own. Examples of hedges include might, perhaps, likely, perchance, and possibly. Boosters are markers employed to indi- cate producers’ certainty in making claims or assertions. Boosters include expressions like in fact, as a matter of fact, undoubtedly, in no uncertain terms, I can state categorically, and definitely. Attitude markers are markers that reveal producers’ attitudes or stances toward propositional content. They include expres- sions like surprisingly, interestingly, fortunately, as claimed by…, I agree, and I concur. Engagement markers are markers employed by producers to build or foster relationships between the receivers, the text, and the producers. They include the expressions consider that imagine that you recall, you see, note that, and let’s elaborate on this. Self-mentions are markers that make explicit reference to the producers of text. Examples include I, we, my, our, mine, and us. The application of metadiscourse to text has been restricted by linguists and is often applied in studies bordering academic writing (Kawase, 2015; Kuhi & Behnam, 2011; Lee & Casal, 2014; Mur-Dueñas, 2011; Obeng, 2019). Hyland (2017) advocates for more diverse studies by commenting that ‘there is a serious danger that the [metadiscourse] approach might remain too closely associated with the description of a limited range of text types and fail to realize its potential as a systemic means of gaining insights into participant interaction more generally’ (p. 19). Furthermore, Sivanya (2019) echoed the need to explore the use of metadiscourse in spoken nonacademic discourse, particularly video content. Sivanya (2019) con- tends that ‘writing is not the only method people can use to approach others nowadays’ (p. 315). Notwithstanding the almost exclusive focus of metadiscourse studies on written academic texts, some strides have been made in studying how metadiscourse plays out in nonacademic spoken and written texts (Byro, 2017; Duwila & Probowati, 2021; Esmer, 2017; Herzuah, 2018; Lundell, 2014; Perez, 2014; Sivanya, 2019; Trajkova, 2014; Turiman et  al., 2018). Turiman et  al. (2018) study, which examined the use of metadiscourse resources in technical (one-on-one) and non-technical (panel interview) Malaysian English as a Second Language (ESL) job interviews, revealed that participants who were involved in non-technical interviews employed most metadiscourse resources to establish rapport with their inter- viewers. The panel nature of the discourse made it necessary for participants in such cases to seek common familiar grounds with their assessors. On the other hand, participants involved in technical interviews spent much effort organizing their responses for the comprehension of their assessors. Non-technical interviewees employed more interactional metadiscourse resources, while technical inter- viewees employed more interactive metadiscourse resources. Al-Anbar et  al. (2023) explored the use of interactional metadiscourse by first language (L1) and second language (L2) English editorialists. Hinging on Hyland’s (2019) model of metadiscourse to analyze 80 editorials published between 2020 and 2021 in two newspapers, the mixed-method study revealed that L1 editorialists featured more engagement markers in their writings than their L2 counterparts. Moreover, the L1 editorialists relied more on hedges, while L2 editorialists employed more boosters in their writing, with the different cultural preferences of the writers cited as a possible factor for the variation in the use of the interactional metadiscourse resources. In Alghazo et al. (2021) study on grammatical devices of stance in written English, the research- ers examined 120 research article abstracts in the areas of applied linguistics and literature. Their aim was to ascertain grammatical devices and semantic distinctions that are employed by English academic writers of English to express their stance in the research article abstract. Notable among their findings was the fact that both applied linguistics and literature writers express their stance (thus interact and engage with their readers) using the linguistic device of stance complement clause (grammatical con- structions controlled by a verb, an adjective, or a noun, or containing extraposed structures). Furthermore, Alghazo et  al. (2024) study on the construction of stance in English and Arabic newspaper editorials examined a corpus of 80 newspaper editorials using Hyland’s taxonomy of stance. The mixed-methods study revealed that the two languages differ in their construction of stance in the editorials. Whiles the Arabic editorials employed more attitude markers and boosters, the English editorials made more use of Cogent Social Sciences 3 hedges and attitude markers. Again, Alghazo et  al. (2023) investigation of the use of interactive metadis- course markers in L1 and L2 English editorials revealed that there is no statistically significant difference in the use of interactive metadiscourse between L1 and L2 editorialists, even though there is a slight variation in the use of individual interactive markers such as frame markers and evidential. The study maintains that L1 editorialists employed more frame markers and evidentials than their L2 counterparts, probably due to a number of factors including the nature of the genre and the status and identity of the writers. In a related study by Alghazo et  al. (2021) which explored how academic writers interact with their readers in both English and Arabic using 80 abstracts of research articles in the field of social sciences, the researchers used Hyland’s (2005, 2019) framework. Their findings indicate that there are clear differences in how academic writers express their stance and engage readers in the two languages; while the Arabic writers made use of more boosters and attitude markers, the English writers relied on more engagement markers and hedges. Esmer (2017) compared the use of interpersonal metadiscourse markers in Turkish election rally speeches delivered by Pro-Turkish and Pro-Kurdish leaders and concluded that metadiscourse markers were employed differently by the two groups of leaders based on their diverse ideologies. However, on the whole, the speeches from both groups of leaders made use of more instances of certainty and atti- tude markers than hedges; a revelation that Esmer (2017) posits is an affirmation of the view that stronger language presents a speaker as aggressive and competent. In a study on how metadiscourse was used to create effective and engaging EFL virtual classrooms during the Covid-19 pandemic, Rababáh et  al. (2024) examined 35 online lectures delivered by three university instructors in the UAE. With a quest to ascertain which metadiscourse markers – interactive or interactional – were employed more in the EFL context, the study maintains that more interactional metadiscourse resources were employed by the instructors. It further revealed that attitude and engagement markers were employed by the instructors for the purpose of involving students in the lessons and signaling instructors’ attitude toward their materials and audience. Trajkova (2014) argues that the choice and usage of metadiscourse resources depend on differences in the setting and genres of spoken discourse. A study that considered metadiscourse use in Macedonian and English talk shows and closing courtroom remarks maintained that the deliberate use of engage- ment markers in such spoken discourses aims to present speakers as less intrusive and more convincing. Trajkova’s (2014) study indicated that English talk shows and closing courtroom remarks were crafted to be more convincing to audiences than the corresponding genres in Macedonia. In effect, metadiscourse studies on spoken genres including job interviews, election rally speeches, as well as talk shows and courtroom remarks have established varying uses or functions of metadiscourse markers. This study explores the uses or functions of metadiscourse markers on television talk shows in the Ghanaian context. With regard to metadiscourse studies that have assessed media texts, the literature suggests that metadiscourse resources are consciously employed by producers of media texts in varying contexts and genres to achieve specific objectives. Thus, the context affects the use of metadiscourse resources. Lundell’s (2014) study considered interactions in cross-media formats (from traditional television to the web) of televised shows. The study, which was based on the premise that metadiscourse is all multi- modal (discursive and visual) communication that is used to comment on an upcoming show, concluded that speakers intentionally constantly adjust their speech to make room for audience involvement by using varying engagement markers that make talk explicit. It further adds that in live television broad- casts, the discourse is highly structured, following a discourse structure of a sequence of events. In a related study, Sivanya (2019) assessed the use of metadiscourse as a rhetorical strategy in YouTube review videos. The study concluded that the casual context that characterizes spoken nonacademic dis- courses (such as YouTube review videos) renders their use of metadiscourse resources less formal. Among the rhetorical metadiscourse strategies identified in this study are clarifying meaning, commenting on linguistic forms, managing terminology, managing topics and phonics, arguing and exemplifying, imag- ining scenarios, and managing comprehension and messages. Professional casters rely on more interactional metadiscourse resources to make their commentaries real and engage with viewers. In a study on how British speakers use personal metadiscourse in TED Talks, Duwila and Probowati (2021) reveal that the personal pronoun ‘I’ was used more than ‘You’ and ‘We’ in the talks. The British speakers used ‘I’ mainly to exemplify in their talks. It helped them present their personal experiences as buttresses to the audience and build a connection with the audience. Al-Ali 4 E. N. A. MENSAH ET AL. and Hamzeh (2024) also reveal that the first-person singular pronoun bound morpheme ‘I’ and ‘my’ are used to connect with addressees and the second-person singular pronoun ‘you’ is used as an engage- ment marker to acknowledge the presence of the addressee. In the domain of digital genres, such as websites, Perez (2014) studied how Spanish and US business websites utilize interactional metadiscourse. The study revealed a higher incidence of hedges, boosters, attitude, engagement, and self-reference markers on US websites, as opposed to the scarce presence of such markers on Spanish websites. Perez (2014) attributes this manifestation to the differences in the cultural values of both countries. Lee (2021) also observed differences in the use of attitude markers between US business communication and Chinese corporate communication and attributed the frequent and intense use of attitude markers in US corporate communication as compared to that of Chinese to differences in sociocultural factors. In addition to the foregoing, studies have also shown that metadiscourse resources contribute to impres- sion management and identity construction. Herzuah’s (2018) qualitative textual analysis study revealed that users employed the metadiscourse resources of self-mention (through personal and possessive pro- nouns) and several engagement markers (reader pronouns, questions, and directives) to communicate mes- sages and negotiate their identities on WhatsApp Messenger. The study found that self-mentions were used to intentionally center users’ communications on themselves, imply a unique relationship with other peo- ple, and entice readers to read status updates. Additionally, the use of reader pronouns, queries, and instructions signals the participation of readers as participants and assists users in meeting their expectations. Despite the fact that there has been an increase in the usage of social media platforms, there remains a quest for television programming as an alternative means for gathering information. As such, many socio-political agencies continue to employ television spaces to disseminate messages and persuade audiences. With heterogeneous target audiences in mind, media outlets design programs that appeal to their audiences, ranging from news, telenovelas, reality shows, talk shows, and documentaries (GeoPoll, 2019). Additionally, the discourse and linguistic resources featured in the programs on such media out- lets have different characteristics. These characteristics vary by way of the genres of communication, mechanisms for feedback, communication terminologies, and interlocutors involved in the discourse. To a large extent, the talk that occurs on television talk shows differs from the written discourses on other programs in terms of spontaneity. Except in cases where contextual references are made to external sources or materials, television talk show discourses are usually unscripted (Okoto, 2020). This sponta- neity of talk on television makes them suitable sites for persuasive communication in which interlocutors are likely to speak in a less pre-planned manner (Illie, 2006). Interlocutors on talk shows seize every opportunity on set to propagate their ‘ideologies’ assuming dif- ferent stances toward their lines of argument, and their audiences as well (Le, 2004). Interlocutors on tele- vision talk shows assume a defined stance toward their arguments in a way that presents the arguments in a good light for favorable consumption by the audience. To this effect, metadiscourse resources have proven useful for persuading audiences of television talk shows. By employing metadiscourse resources in line with various rhetorical techniques, interlocutors aim to achieve logical, credible, and affective appeals. The metadiscourse theory proves useful for the current study because it has been used in various studies to examine how producers of media texts engage their audiences. For instance, Lundell (2014) employed metadiscourse theory to study interactions in cross-media formats (from traditional television to the web) of televised shows and looked at the multimodal (discursive and visual) communication that is used to comment on an upcoming show. Sivanya (2019) used metadiscourse theory to examine YouTube review videos, which are rapidly becoming sources of information for many Internet users. Following this, the current study, which seeks to analyze how interlocutors on a Ghanaian television talk show employ metadiscourse resources in their ‘actual talk’ for the realization of persuasive appeals, lever- ages the conceptual foundations of metadiscourse theory. Persuasive communication Communication is done with a purpose in the minds of the parties involved. Ells (2017) asserted that the main purpose of communication is to inform, entertain, or persuade. Thus, parties in any communicative Cogent Social Sciences 5 event set out with a purpose they hope to achieve through the event. With communication that hinges on the aim of persuading others, producers (speakers or writers) aim to bring incontestable meaning to receivers that will win them over to the producer’s side on a given matter. According to Taillard (2000), persuasive communication has two goals: 1) the need to be understood, and 2) the need to be consid- ered. In other words, for persuasion to take place, there is a need for audiences to understand the mes- sages sent so that they can believe and respond appropriately. According to Higgins and Walker (2012), persuasion takes place only when meaning is shared and accomplished by receivers of discourse. Moreover, to accomplish meaning in receivers, speakers employ rhetoric; thus, wherever there is meaning, there is rhetoric (Higgins & Walker, 2012). Through his seminal work entitled Rhetoric, the Greek philosopher Aristotle postulated the basis for composing messages aimed at persuading audiences (Torto, 2020). In his postulations, Aristotle identified three means (appeals) for persuasion: logos (appeal to reason), ethos (appeal to credibility), and pathos (appeal to emotions). Over the years, persuasive appeals have been employed by many speakers in several communicative instances. Indeed, the ubiquity of usage has resulted in a plethora of studies that have been conducted on rhetorical appeals in different disciplines that use communication. For instance, studies have been con- ducted on rhetorical appeals in organizational and industrial communication (Murthy & Gosal, 2016), advertising studies (Yang et  al., 2018), and social/environmental communication (Higgins & Walker, 2012). Furthermore, extant literature has shown that various strategies are employed by producers of text to achieve persuasive goals in communication. Among the strategies identified in the literature for persua- sive communication are Higgin and Walker’s (2012) argumentation, justification, evidence/claims, simili- tude, deference, expertise, self-criticism, metaphors, and identification. Ting (2018) advanced the strategies of argumentation, warrants, examples, questioning, factual language, linguistic links, ingratiation, confess- ing, pleading, praising, and the use of emotive words for persuasive communication. Metadiscourse and persuasive communication Persuasive communication employs metadiscourse features. According to Hyland (2005), the concept of metadiscourse is pivotal in persuasive communication in that the resources help producers realize their rhetorical goals of ethos, pathos, and logos in producing text. Consequently, some researchers have delved into metadiscourse and the realization of rhetorical appeal. For instance, in a cross-cultural study, Mai (2016) assessed the use of metadiscourse for persuasion in selected American and Chinese speech. The study showed that the cultural differences of speakers made them employ varied measures of metadis- course resources for persuasion in their speech. The study further contends that the broad sociocultural contexts of American speakers make them employ the most metadiscourse for the realization of ethos far more than pathos and logos. The explanation is that the American speaker assumes responsibility for composing content to ensure that the meaning is interpreted as intended. On the other hand, since ‘in the Chinese culture, authority is not expected to be responsible for explaining’ (Mai, 2016, p. 216), Chinese speakers use the most metadiscourse resources to achieve pathos rather than ethos and logos. In a related study, Perez (2014) reveals that US companies employ hedges on their websites to refrain from imposing themselves on customers. This use of hedges allows customers freedom of decision, thus reducing negative face threats. Moreover, the use of hedges heightens emotional appeal and boosts the company’s reputation by presenting it as one that gives recourse to the decisions of consumers. In Esmer’s (2017) study, the findings revealed that speakers used high occurrences of commentaries (rhe- torical questions, direct address to listeners, inclusive expressions, and personalization) aimed at building solidarity between speakers and listeners. For instance, through the use of ‘Inclusive We,’ speakers created collective identities with their listeners and reduced rejection tendencies from listeners. On their part, Duwila and Probowati (2021) disclose that the personal pronoun (I) in text fosters speakers’ credible appeal, as it affords them the opportunity to present their experiences and expertise to the audience. Moreover, personal pronouns (You, We) help speakers build a connection with audi- ences, which fosters emotional appeal and helps speakers win audiences in their stance. On their part, Turiman et  al. (2018) revealed that participants used interactional metadiscursive elements to endear 6 E. N. A. MENSAH ET AL. themselves to their interviewers, thus building rapport with panel members. However, their findings posit that excessive use of metadiscursive elements such as sequencers (e.g. first, then, next, and then) impedes flow in communication, instead of making speakers more persuasive in their speeches. In effect, the wordiness and repetition that arise from certain instances of unwarranted use make talk laborious and sacrifice the persuasive effect of speech. Trajkova (2014) revealed that engagement metadiscourse resources play a role in enhancing the writer’s or speaker’s authority and fostering a relationship with readers or listeners. The foregoing studies echo Hyland’s (2005) position that metadiscursive resources change the language used to convey propositional content without changing the propositions them- selves. These elements provide useful channels that help the receivers consume messages. In all, the aforementioned studies give credence to the fact that metadiscourse resources have been employed in different contexts for the achievement of persuasive appeals. The present study, in turn, investigates the use of metadiscourse resources in the genre of television talk shows. Materials and methods This study employed a qualitative research approach to examine how interlocutors on a Ghanaian tele- vision talk show employ interactional metadiscourse resources in their talk and to realize rhetorical appeals. In tandem with the postulations of Lindlof and Taylor (2017) and Salvador (2016), this study drew subjective data from the talk show. Data were collected as interlocutors freely interacted on tele- vision. Subsequently, the data were analyzed. Lindlof and Taylor (2017) found that the raw materials analyzed in qualitative research were actual talk and gestures. As this study focused on data from a television talk show in the form of conversation, a qualitative approach was adopted. Again, the qualita- tive approach subjectively describes social phenomena as they occur naturally with no effort to influence the situation under study (Salvador, 2016). This study employed a qualitative content analysis design. The talk show videos were collected, tran- scribed, and read iteratively for recurring patterns in line with the tenets of metadiscourse. Qualitative content analysis entails the reduction of text by identifying core consistencies or patterns to categorize them as such and make sense of them (Patton, 2002). Moreover, the ultimate goal of qualitative content analysis is not to produce counts or statistical significance; rather, it is concerned with unearthing pat- terns or categories in the text that are relevant to social reality (Wildermuth & Zhang, 2014). As such, data from the talk show were diligently examined to categorize repeated forms into themes to identify the meaning of the patterns. Sampling The purposive sampling technique was used to sample five episodes of the Ghanaian television talk show. According to Lindlof and Taylor (2017), purposive sampling is a non-probability technique in which sites are selected based on the belief that what goes on there is essential for understanding a phenom- enon. The purposively selected sample should possess characteristics that enable the researcher to explore and understand the central themes being studied (Lindlof & Taylor, 2017). The Ghanaian televi- sion talk show was purposively sampled because it serves as a good ‘site’ for exploring interlocutors’ use of metadiscourse in the television space, owing to the fact that it ranks high in the list of current affairs talk shows with the most viewership (GeoPoll, 2019). Furthermore, it is telecast on a channel that has been selected for previous studies by reputable bodies such as the Media Foundation for West Africa based on its wide reach (audience). Moreover, the guests who feature on the show are experts in various fields that have a direct bearing on the particular issues being discussed at a time (Myjoyonline. com, 2020). The study sampled five episodes of television talk shows for an in-depth analysis of the phenomenon of metadiscourse use in television talk shows. Employing textual analysis as a data collection method, episodes of the talk show were collected from the YouTube channel of the media house that produced the show, transcribed, and closely read. Each episode of the talk show was approximately 2 hours and 15 minutes in length, with a total of three thousand, two hundred and forty-one utterances in all. The recurring patterns revealed through close reading were thematized and analyzed for latent meanings. Cogent Social Sciences 7 These episodes allowed the researcher to gather credible and relevant information on metadiscourse use by interlocutors on the television talk show. The episodes were selected over a period of one month (Saturday, November 7 to Saturday, December 5) before the 2020 general elections in Ghana. According to Ogbidi et  al. (2016), the period leading to elections in any country is usually critical in the media landscape, and matters that tend to affect voter decisions are rife on television and radio spaces during this period. Therefore, most programs aired around this time feature experts who are poised to garner support in matters of national interest. Moreover, in line with ethical considerations, the researcher masked the interlocutors by assigning each interlocutor a different pseudonym. The pseudonyms used indicate the number for identification as well as the episode number. For instance, a pseudonym like G1-E4 refers to guest one (G1) in episode four (E4), whereas the pseudonym G5-E2 refers to guest five (G5) in episode two (E2). The host of the program was assigned the pseudonym HS combined with a particular episode number. For example, HS-E4 refers to the host’s utterance in episode four (E4). Episode 1 featured four (4) guests, episode 2 featured five (5) guests, episode 3 featured five (5) guests, episode 4 featured four (4) guests, and epi- sode 5 featured four (4) guests. Additionally, with excerpts where interlocutors explicitly mentioned other persons, the names of the persons mentioned were replaced with befitting pseudonyms (as seen in the findings and discussion section of this paper). The masking of interlocutors and other persons in the talk follows Yin’s (2008) position that qualitative research employs pseudonyms in granting anonymity to participants and eliminating any possible linkage between participants and the pseudo-identity given by the researcher. Descriptive statistical tools such as simple frequency tables, bar graphs, and pie charts were employed to provide an overview of the indication of metadiscourse resources while providing further grounding for the discussions. Knupfer (2007) argues that the use of descriptive statistical tools in qualitative research reduces data into convenient formats for readers to understand, especially when large amounts of data are involved and there is a need to show patterns of occurrence and distribution. In a bid to adhere to accuracy, objectivity, and trustworthiness, the study employed rich thick descriptions, presen- tation of negative evidence, inter-coder reliability (involving the development of a clear coding scheme, training of two coders, pilot testing, iterative coding process, and regular debriefings), and spending prolonged time in the field as trustworthiness measures (Creswell & Creswell, 2017). Results and discussion This section of the study is divided into two parts. The first part presents the types of interactional metadiscourse resources employed on the television talk show, as well as their communicative impor- tance. The second part presents findings on how interactional metadiscourse resources are employed for the realization of rhetorical appeals in the talk show. Kinds of interactional metadiscourse resources used on the television talk show Based on Hyland’s (2005) metadiscourse theory, the analysis of data collected indicates that interactional metadiscourse resources accounted for 64% (two thousand and seventy-four occurrences) of the utter- ances made by interlocutors on the episodes of the talk show examined, as indicated in Table 1. These findings suggest that the main focus of interlocutors on the Ghanaian television talk show was conveying their beliefs and attitudes as well as engaging their audiences through utterances on the show. Rather than simply organizing their talk in a way that makes sense to their audiences, interlocutors employ varied interactional resources (64%) and thus give credence to their audience. By so doing, the audience is engaged and helps relate to the content of arguments. These findings affirm Turimanet al. (2018) position that speakers employ more interactional metadiscourse when they see the need to show respect, recognition, or allowance for their hearers. The findings also suggest that the use of interactional metadiscourse resources affirms speakers’ quest to connect with their hearers in ways that make the discourse content more realistic and easily relatable. Again, the findings presented above seem to affirm Lundell’s (2014) conclusion that the genre of television talk shows goes beyond merely organizing con- tent, representing speakers in the best light, and drawing audiences into the talk as participants. Lundell 8 E. N. A. MENSAH ET AL. (2014) adds that, since talk shows are broadcasts designed for ‘absent’ audiences, the shows must be organized in ways that make them welcoming and engaging to the audience. In the distribution of metadiscourse resources on the Ghanaian television talk show, as seen in the table above, the findings indicate that interlocutors on the show make the most mention of themselves and their expertise (self-mentions, 30%), use structures that invite their audiences to reason with them (engagement markers, 26%), make remarks that assert their claims (boosters, 20%), use linguistic devices that show their stance toward propositions (attitude markers, 15%), and make utterances that withhold their commitment to some claims made on the show (hedges, 9%). The next section of this paper focuses on determining the communicative functions that each resource performs on the talk show. Communicative functions of interactional metadiscourse resources used on the talk show Findings from the study indicate that interactional metadiscourse resources were put into several com- municative functions on the television talk show. The resources were employed mainly by interlocutors to make their views toward propositional material explicit and engage audiences in ways that foster a healthy relationship. Again, by using interactional resources, interlocutors also controlled the level of their personality in the text and guided their audiences to interpretations. Moreover, through interac- tional metadiscourse resources, the interlocutors showed their awareness of the socially determined posi- tions of their audience and crafted their arguments in ways that could reach the audience appropriately. The communicative functions of the interactional metadiscourse resources used in the talk show are also discussed. Self-mentions It is evident from the analysis of data in Table 1 that the interactional metadiscourse resource self-mention was the single most used linguistic element by interlocutors on the television talk show, with a fre- quency of 622 (30%) out of the 2074 (64%) of interactional metadiscourse resources employed in the episodes studied. According to Hyland (2005), self-mentions as metadiscourse resources are used to explicitly refer to the producer in communicative engagement. The study reveals that interlocutors on the talk show use the self-mention resources of first-person pronouns (such as I and exclusive we) and possessive determiners (such as mine, our, and ours) to protect themselves and signal their involvement in the discourse. Two main communicative functions were identified for self-mentions in the talk show: establishing a singular authorial presence and indicating shared credit. In establishing a singular authorial presence, the findings reveal that interlocutors use words like I, me, my, and mine to draw audience’s attention to themselves and show authorship of propositions made on the show. In effect, interlocutors project themselves according to how they say what they say on the show. ‘True’ self-mention markers that signaled singular authorial presence accounted for 72% (448 instances) of the total 622 self-mentions. He lived the ideals he sounds, and in leadership, it’s an extremely difficult, extremely complex process, and I think he managed to live true to himself until this was done untimely. G5-E2 …and in respect of that, I have seen some people asking that, I may be a lawyer but do I think I know so much? Well, this is the area where I did my LLM. My masters in law, I did it on article 181. That’s my project, so I think I know it fairly enough. HS-E2 Table 1.  Distribution of interactional metadiscourse resources on the talk show. Interactional metadiscourse resources Category Frequency Percentage Self-mentions 622 30% Engagement markers 539 26% Boosters 415 20% Attitude markers 311 15% Hedges 187 9% Total 2074 100% Cogent Social Sciences 9 The above excerpts of self-mention clearly depict interlocutors’ use of the first-person pronoun ‘I’ and the possessive determiner ‘My’ to bring to bear their stance on issues being discussed. Herzuah (2018) maintains that this use of self-mentions is a deliberate attempt to focus audiences’ attention on the producer of text and project the producer’s ideals. Duwila and Probowati (2021) affirmed this stance by indicating that the use of the first-person pronoun ‘I’ enables speakers to present their personal experi- ences in ways that build a connection with their audience. In indicating shared credit, the findings reveal that interlocutors on the talk show show a common agreement with other panelists and personalities on the show. Self-mention markers that signaled shared credit accounted for 28% (174 instances) of the total 622 self-mentions. Of course, we know that you cannot use the work of the Commission or Committee of Inquiry to prosecute. You need new investigations. It is what it is! HS-E3 We must also remember that this particular practice of refusing to provide information was so consistent that it is quite evident from the OSPs language that he was infuriated at the end of it. G2-E1 In the above examples, interlocutors used the pronoun ‘We’ to draw fellow interlocutors into agree- ment with propositions made. It turns out that in using self-mentions for indicating shared credit, inter- locutors on the show use the pronoun ‘We’ which Hyland (2005) terms as ‘inclusive ‘We’ to indicate that their utterances represent positions that are shared with, especially, other interlocutors in the studio. This communicative function of self-mentions presents interlocutors on the show as ‘one’ and waters down the ‘controversial’ nature (Illie, 2006) that talk shows have been identified to have. Engagement markers Analysis of the data in Table 1 also indicates that, next to self-mentions, interlocutors on the television talk show make much use of engagement markers, recording a frequency of 539 which forms twenty-six percent (26%) interactional metadiscourse use in the data analyzed. Hyland (2005) argued that engage- ment markers ‘engage’ audiences in discourses by beckoning them as discourse participants, or steering their attention to specific details. The findings of the study reveal that interlocutors employ engagement markers for a studio audience (host or other guests) as well as an absent audience (viewers). Two com- municative functions were identified for the use of engagement markers in the talk show, inclusion and direction. With the communicative function of inclusion, it turns out that interlocutors use engagement markers to create a unification of speaker and audience, in that the markers project interlocutors and their audi- ence in a common light as participants in ongoing discourse. Interlocutors achieve this communicative function mainly by the use of ‘inclusive we’ (Hyland, 2005), thus bonding with their audiences. Engagement markers that signaled inclusion accounted for 33% (178 instances) of the total 539 self-mentions. The office of the Special Prosecutor is something we all advocated for, and it went through the process. You see, apart from being a party manifesto pledge, that once it went to Parliament and both sides decided to deal with it in a bi-partisan way, the party act was off. G2-E3 At the very least, we all have agreed that when it comes to integrity, he is 200%. Thus, once he says he suspects that the investigators are compromised, we expect a transparent investigation of that matter. G4-E4 From the excerpts above, it is evident that the expression ‘we all’ is used to refer to interlocutors as well as audiences of the show. By using this expression, the speakers suggest that the information they are presenting is common to the interlocutors and the audience. With the communicative function of direction, the findings reveal that engagement markers are employed in the show to steer the audience’s attention to specific details. In this regard, interlocutors use viewer references like the pronoun ‘You’; directives like ‘See’ or ‘Look’; and questions like ‘Can you imagine?’ to focus audiences’ attention on propositions. Engagement markers that signaled direction accounted for 67% (361 instances) of the total 539 self-mentions. You recall the misadventure by my learned senior NA relative to the ex-party application to injunct Afari Gyan… G4-E1 10 E. N. A. MENSAH ET AL. Do that horizontally because that, you know, improves your chances of not staining the ballot. G4-E5 See, that matter was dealt with in the Parliament okay. That is why I would have loved if he had continued in office and probed further, so that we all have some education and clarification on those issues. G3-E4 In the excerpts above, interlocutors use the expressions ‘You recall,’ ‘You know,’ and ‘See’ to beckon their audiences to reason with them and concentrate on propositions being made. According to Duwila and Probowati (2021), when speakers desire to carry their audiences along in the arguments being made in any discourse, they resort to the use of engagement markers, as confirmed by the results of this study. Boosters Apart from self-mentions and engagement markers, the analysis of data from the study indicates that interlocutors use boosters, with a frequency of 415 (20%). Hyland (2005) mentions that boosters are employed to present the voice of text producers as assertive, thus leaving no room for ambiguity or varied opinions. In tandem with this assertion, this study reveals that interlocutors on the Ghanaian television talk show employ boosters to assert their arguments and indicate certainty and completeness in their assertions. The communicative function identified for boosters in the talk show asserts assert- ing claims. In asserting claims, boosters are employed by interlocutors to affirm their personal claims against those from other speakers on the show. The data further reveal that working closely with evidentials and endophoric markers, this function of boosters seeks to cover all argumentative grounds and present the speaker as an authority in a given subject matter. You come to the facts upon which you mount a prosecution, but you get the sign here that the Special Prosecutor has his view of the matters in terms of some of the statements. The Official Government One is this person. As a matter of fact, you have not interacted with the person you claim to be Government Official One, so there might be issues. G3-E2 …and certainly, we don’t want to signal as a country that, you know, the government of the day is using prosecution to harm its leading rivals in an election. G3-E4 We definitely would know that Martin Amidu’s integrity and credibility are not in doubt, but his conduct, with regard to the operational difficulties he refers to, is. G4-E3 In the examples cited above, interlocutors use expressions (italicized and boldened) that signal finality to assertions made. The expressions indicate certainty in the views of speakers, as opposed to counter- arguments made by other speakers. The findings affirm Perez’s (2014) assertion that boosters indicate speakers’ commitment to propositions made and their willingness to defend their stance by all means. Again, the findings align with Hyland’s (2005, p. 53) assertion that boosters present a single voice to a speaker’s assertions, with little to no allowance for varied opinions or viewpoints. Attitude markers Findings from the study revealed that interlocutors on the talk show employed attitude markers 311 times, representing (15%) of interactional metadiscourse use on the show. Attitude markers generally reveal the feelings or stances that producers hold toward claims made in a discourse (Hyland, 2005). The findings of this study reveal that interlocutors employ attitude markers to provide signals of their affec- tive stance toward arguments made. In addition, attitude markers are employed to indicate speakers’ stance toward other speakers on the show. It turns out that attitude markers perform the communicative function of demonstrating a positive stance and a negative stance. These functions are performed through the use of deontic modals (like have to, should, must), attitudinal adjectives (like desirable, unfortunate), affective adverbs (like interestingly, surprisingly) and other expressions conveying a stance (Wei et  al., 2016). In demonstrating a positive or affirmative communicative function, interlocutors use expressions that indicate a good outlook at propositions and other speakers on the show. In effect, interlocutors use expressions that confirm, approve, or condone claims or people. In the examples cited below, interlocu- tors use expressions that indicate fascination and delight at propositions. Cogent Social Sciences 11 Trump is not thinking about conceding, even though, in the key states that are now involved, Joe Biden clearly is in the lead. G4-E1 What is even more fascinating is that Donald Trump’s votes have actually been increasing among minorities, and obviously, you know, there is a monolithic voting bloc. …So, I think it is a really exciting lesson for all of us. G2-E1 Good morning again, samson. I’m delighted to join your show this very sad day. To share thoughts about a charismatic leader. I do not think that, in Africa, we have found more charismatic leaders. G2-E2 In demonstrating negative or contradictory communicative function of attitude markers, interlocutors’ disposition is presented through expressions that indicate bad taste or bad light in relation to subjects and individuals under discussion on the show; in opposition and contradiction or deviation from the ‘norm’: I’m so disappointed. Look at the document from the Parliament. This was an aspect of the criminalization of unexplained wealth. A special prosecutor revealed that as part of his corruption prevention program in 2020, it plans to push the agenda for the criminalization of unexplained wealth. This would place a burden of proof on public servants and private citizens to explain their excess earnings in the same pursuit. G2-E3 Unfortunately, this was not the case, and now we have 0.01% or something in Ashanti Gold International. G1-E2 Hedges Analysis of the data gathered through the study indicates that the least used interactional metadiscourse resource is hedged, with a frequency of 187 (9%). Wei et  al. (2016) found that hedges are employed to indicate speakers’ recognition of the possibility of varied opinions or viewpoints on propositions. Findings from the study further show that by using hedges, interlocutors on the talk show withhold their com- mitments to propositions and allow audiences to make inferences on their own. Through hedges, inter- locutors vary the weight they place on their assertions and provide room for negotiations with other speakers. In so doing, speakers indicate that their claims are not finite, thus allowing for plausible inputs from other speakers. From the analysis, it turns out that interlocutors on the talk show use epistemic modal verbs (like might, could, could be, may, may be), lexical verbs (like suggest, think, appear, claim, presuppose), adjectives, adverbs (like probably, quite, possibly), and nouns (like a possibility, probability, likelihood) as hedges. Two main communicative functions were identified for hedges on the talk show: suppressing speaker commitments and reducing imposition. In suppressing speaker commitments, interlocutors on the show withhold their commitment to claims using expressions that indicate that the claims made are not conclusive in themselves. Such expressions are employed by speakers in instances where they are not absolutely certain of the facts, details, or claims being made at a time. In the excerpts below, expressions like ‘quite difficult, ‘‘I think, ‘‘presupposes, ‘‘it appears, ‘‘so to speak, ‘‘quite clear’ and ‘I am not sure’ are hedges that end up suppressing the speak- ers’ commitments to utterances made at a point. This use of hedges also positions speakers as people who are not willing to approve or deny claims made by other speakers outrightly, thus creating a pseudo common ground for speakers despite their opposing views. In a close election, it is quite difficult and I think any country with such a system would have faced such significant challenges. G3-E1 We have gone far ahead of that, but that presupposes that the US is living on past glory. G1-E1 Thus, it appears that the markets reward America when politicians do not act. G2-E1 I just looked at the document and it’s like even with this minimum exposure. I find that the whole transaction is riddled with bad things, so to speak. G3-E3 However, if you look at what has happened in the Agyapa transaction, it is quite clear that someone who was very skilled had set out to circumvent the process without doubt, and I am not sure that any technical mea- sures, so to speak, would have been enough if we had. G5-E3 12 E. N. A. MENSAH ET AL. In reducing imposition, it turns out that hedges are used by interlocutors to reduce all possible occur- rences of forcefully advancing arguments or making the demands of interlocutors against their will. The use of modal verbs in the excerpts below shows interlocutors’ attempts to reduce their imposition on their hearers. Let us consider KK in this case. KK, if you may start from the last point. HS-E4 …but maybe shall we look at it. Section 17 clause 3 states that the deputy Special Prosecutor shall act in the absence of the special prosecutor? G1-E3 The findings of the study align with postulations by researchers such as Herzuah (2018) and Sivanya (2019), who maintain that metadiscourse resources are employed in spoken media texts for the main purpose of revealing the persona of speakers as well as indicating their stance toward propositions and other speakers. In effect, resources add life to propositions in the discourse. However, the findings further extend existing postulations from the said scholars by revealing that interactional metadiscourse resources are employed for the additional purposes of establishing a singular authorial presence, indicating shared credit, and providing direction for audiences. According to Hyland (2005), the employment of metadiscourse resources in discourse is correlated with speakers’ quest to fulfill rhetorical demands in persuasive communication. The analysis of how inter- actional metadiscourse resources are employed for the realization of rhetorical appeals on television talk shows is the focus of the next section. Interactional metadiscourse resources and the realization of persuasive appeals Once meaning is accomplished in the receivers of any discourse, persuasion is said to have taken place, and rhetoric is pivotal in realizing persuasion (Higgins & Walker, 2012). This means that at the forefront of any communicative act that would be successful in the end are the rhetorical appeals of ethos, logos, and pathos. Hyland (2005), in turn, posits that Aristotle’s three means of persuasion are of great interest in metadiscourse research (ethos, pathos, and logos). Hyland (2005) argued that metadiscourse resources con- tribute to the realization of persuasion in any discourse. Indeed, studies conducted by Mai (2016) and Biri (2017) have concluded that metadiscourse resources are employed in various measures and techniques toward the end of persuading audiences, in line with the demands of ethos, logos, and pathos. Analysis of the data gathered for this study indicates that interactional metadiscourse resources are employed on the television talk show for the major purposes of realizing credible (ethos) and affective (pathos) appeals. Interactional metadiscourse resources and the realization of credible appeal (ethos) Ethos is an aspect of rhetoric that invites audiences to trust and accept a speaker’s propositions, usually based on the persona projected for the speaker (Ting, 2018). Relating to the credibility of the speaker, ethos revolves around the audience’s perception of the speaker and the resultant willingness of the audience to accept the message being conveyed to them by the speaker. Torto (2020) maintains that the receiver’s perception of the producer’s persona determines whether the producer’s message is accepted by the receiver or audience. As far as Hyland (2005) is concerned, the use of metadiscourse resources to project a producer’s authority or competence lies in the domain of ethos. As such, Hyland posits that boosters, hedges, and self-mentions are the resources used to convey ethos. In confirmation of Hyland’s (2005) postulations, analysis of the data gathered for this study reveals that the interactional metadiscourse resources of boosters, hedges, and self-mentions are employed in specific ways for the realization of credible appeal (ethos). The findings revealed that interactional meta- discourse resources are employed for credible appeal through similitude, deference, and expertise. Similitude The rhetorical strategy of similitude borders projects similarities between speakers and their audiences (Ting, 2018). As such, similitude projects the speaker as one who is ‘one of’ the people being addressed Cogent Social Sciences 13 (Higgins & Walker, 2012). It creates a sense of shared identity between the speaker and audience and thus fosters trust between the two parties. It is this trust evoked in the audience that makes the speaker credible, and that his message is acceptable to the audience. From the data, it is evident that interlocutors on the talk show use pronouns like ‘we’ (inclusive we), and expressions like ‘you and I,’ which are self-mentions, to indicate similitude with their audience. The office of the Special Prosecutor is something we all advocated for, and it went through the process. You see, apart from being a party manifesto pledge, that once it went to Parliament and both sides decided to deal with it in a bi-partisan way, the party act was off. G2-E3 At the very least, we all have agreed that when it comes to integrity, he is 200%. Thus, once he says he suspects that the investigators are compromised, we expect a transparent investigation of that matter. G4-E4 The excerpts above provide evidence of interlocutors’ use of expressions that indicate similitude with their hearers. The findings further corroborate Trajkova’s (2014) assertion that the use of ‘Inclusive We’ projects both speaker and listener onto the same plane for solidarity, thus building a relationship between speakers and audiences who may never have direct physical contact. In other instances, the ‘exclusive we’ is used to indicate similitude among the interlocutors on the show. In such cases, speakers refer to themselves and other speakers on set, exempting the audience. By doing so, interlocutors indicate likeness or parallelism with other experts on the show and thus boost their ethos, as in the example below: …And here on this show, we put Ghana first. HS-E3 Deference The rhetorical strategy of deference in persuasive communication indicates the speaker’s recognition of the opinions, rights, and feelings of the audience (Higgins & Walker, 2012). With deference, the speaker shows the utmost respect for the audience’s needs by restraining them from imposing the opinions of the speaker on their audience. Using the strategy of deference, audiences are given the freedom to decide whether to accept propositions by speakers. The data reveal that interlocutors of the television talk show mainly employ the interactional metadis- course resource of hedges to portray deference. The use of hedges presents speakers with accommodat- ing and unimposing audiences. The examples presented below show the interlocutors’ use of hedges for the purpose of deference. But if you look at what has happened in the Agyapa transaction, it is quite clear that someone very skilled had set out to circumvent the process without doubt, and I am not sure that any technical measures, so to speak, would have been enough if we had. G5-E3 In a close election, it is quite difficult and I think any country with such a system would have faced such significant challenges. G3-E1 We have gone far ahead of that, but that presupposes that the US is living on past glory. G1-E1 …Maybe; shall we look at it? … G1-E3 Expertise The rhetorical strategy of expertise involves honing speakers’ qualifications, experience, and knowledge (Higgins & Walker, 2012). The strategy of expertise projects the speaker’s unique position as an authority on the subject under discussion. According to Torto (2020), expertise refers to the credentials of the speaker and presents them as the best suited to speak on specific issues. Analysis of the data gathered for this study indicates that interlocutors use self-mentions and boosters to hone their expertise in the talk show. In effect, self-mentions were used on the show to focus audi- ences’ attention on the interlocutors’ credibility and experience by mentioning speakers’ own personal expertise in relation to others and the peculiar issue being discussed. The excerpt below provides evi- dence of an interlocutor’s use of self-mention to turn the spotlight on himself. 14 E. N. A. MENSAH ET AL. …and in respect of that, I have seen some people asking that, I may be a lawyer, but do I think I know so much? Well, this is the area where I did my LLM. My masters in law, I did it in Article 181. That is my project, so I think I know it fairly enough. HS-E2 The analysis further showed that the interactional metadiscourse resources of boosters were also employed to show expertise. Interlocutors’ use of boosters on the talk show presented them as people who know what they are about and stand by their claims as assertive speakers. For instance, in the excerpt below, boosters are used to project the talk show as one that has the most credible content for which audiences can ask. Good morning and welcome to the show. It is the most authoritative news analysis platform. HS-E3 The data further reveal that the added significance of expertise is achieved by indicating the sort of persons or sources from which discussions are drawn on the show. This technique for honing expertise is usually featured during the introduction of guests at the beginning of the show, and through the host’s remarks, guests take their turn to make submissions: My guest this morning: BS, who is an honorary executive of IA. In the studio, DYG, Coordinator, TWNA, NA, NPP Communications Team, MD, and STCL. In addition, EGT is a lawyer and member of the NDC National Communication and Legal Team. HS-E1 . or you want to read for me because you are the lawyer? G1-E4 Hyland’s (2005) postulations indicate that a balanced blend of boosters (assertive statements) and hedges (tentative statements) presents interlocutors in a positive light. The blend presents interlocutors as both firm and flexible personalities who would not just assert their views on audiences but also make allowances for others’ views. The resultant persona created for the speakers enhances their ethos. As such, interlocutors present themselves to their audience as ‘expert-knowers’ who are yet humble ‘disci- plinary servants’ (Hyland, 2005, p. 91). Wei et  al. (2016) confirm Hyland’s (2005) position by indicating that producers of text construct the ‘dual authorial identity of humble servants of their disciplines and originators of new knowledge’ (p. 197) by using hedges in balance with boosters. Interactional metadiscourse resources and the realization of affective appeal (pathos) The rhetoric that invites audiences to take action by affecting their emotions and cognition through propositions is pathos (Torto, 2020). Pathos aims to get the audience to feel what the speaker feels about the issues being discussed. Consequently, Hyland (2005) posits that when metadiscourse resources signal respect for the receivers’ viewpoints or show direct relevance to the audience, they convey pathos. As such, Hyland (2005) maintained that attitude and engagement markers are interactional metadis- course resources that convey affective appeal. In affirmation of Hyland’s (2005) postulations, analysis of data gathered for this study reveals that the interactional metadiscourse resources of attitude and engagement markers are employed in specific ways in the talk show for the realization of affective appeal. The findings revealed that interactional metadis- course resources are employed for affective appeal through identification/sociality, imagery and implorations. Identification/sociality As a rhetorical strategy, identification or sociality portrays the speaker as one who understands the needs of their audience. Through identification, speakers show recourse to the audience’s need to belong on a com- mon level as expert interlocutors, not inferior in any way, thus endearing audiences to the speakers. Identification draws audiences into discourse by signaling shared beliefs and attitudes (Higgins & Walker, 2012). According to Higgins and Walker (2012), identification arises from the rhetorical techniques of similitude. The analysis of the data gathered for the study indicates that interlocutors use engagement markers to identify with their audience. Engagement markers explicitly refer to audiences in interlocutors’ utterances and invite the audience to reason and feel together with the interlocutors, as seen in the excerpt below: Cogent Social Sciences 15 Forgive me that I am holding you to the text because I believe that if we go deeper into the text, we will do a lot more justice to our viewers and listeners… HS-E1 Data analysis further revealed that an added function of engagement markers for affective appeal on the show is their use for phatic purposes. Guests usually employ the expression ‘our viewers’ in rendering salutations to the audience before making submissions on the show. It turns out that this expression ‘beckons’ audiences as participants in a less formal or conversational manner, as cited in the example below: SS, good morning, and good morning to our viewers and listeners. G3-E1 Imagery As a rhetorical strategy, imagery borders the use of strong, emotive expressions that create profound emotional effects on audiences (Ting, 2018). This rhetorical technique aims to adequately portray speak- ers’ feelings and stance toward propositional content and audiences in a way that draws audiences into discourses using language that arouses experiences that are persuasive to the speakers’ cause. An analysis of the data gathered for the study indicates that interlocutors employ attitude markers to con- vey imagery and achieve affective appeal, as shown in the excerpts below. The excerpts provide evidence of interlocutors’ use of expressions that reveal how they feel about propositions and persons being discussed: What is even more fascinating is that Donald Trump’s votes have actually been increasing among minorities, and obviously, you know, there is a monolithic voting bloc. …So, I think it is a really exciting lessons for all of us. G2-E1 Good morning again, samson. I’m delighted to join your show on this very sad day. To share thoughts about a charismatic leader. I do not think that, in Africa, we have found more charismatic leaders. G3-E2 Implorations As a rhetorical strategy, implications refer to expressions that signal confessions, regrets, or pleas (Ting, 2018). The technique of imploration gives recognition and respect to audiences and other interlocutors by indicating that it is not the speaker’s intention to offend the audience or other interlocutors. An analysis of the data gathered for the study indicates that interlocutors employ implorations for affective appeal as well. This finding is an extension of the extant literature on the use of metadiscourse resources for the realization of persuasive appeals. Here, though reference is being made to the self (self-mention), an added expression of confession, regret, or plea is attached to it. It turns out that inter- locutors mostly use apologies and pleas for signal recognition of the freedoms of other interlocutors: Forgive me that I’m holding you to the text… HS-E1 In a nutshell, the findings of this study present extensions or add-ons to the extant literature on the subject of metadiscourse and persuasion. There is a substantive extension on postulations by scholars like Hyland (2005), Ting (2018) and Torto (2020) who aver that persuasion in discourses relates not only to the rational presentation of claims, but also to the trustworthy personas of producers (as revealed through the text), and the charged language which is able to arouse audiences appropriately. Conclusion and recommendations The study concludes that the concept of metadiscourse is not restricted to written discourses only, nor is it concerned only with discourses in academia. Indeed, there is extant research into metadiscourse use in various genres (e.g. Al-Anbar et  al., 2023; Alghazo et  al., 2021, 2024; Byro, 2017; Esmer, 2017; Rababáh et  al., 2024; Sivanya, 2019). Metadiscourse is present wherever there is a need to convey messages. The study reveals the existence of interactional metadiscourse in persuasive communication that occurs on television talk shows. This is evident in the plethora of communicative functions of interactional meta- discourse resources employed in Ghanaian television talk shows, as reported by the findings. Indeed, interlocutors make use of interactional metadiscourse resources with or without recourse to what the resources imply and their effects on the content of propositions and the audience at large. Again, it is the conclusion of the study that television talk shows serve as rife avenues for engaging in persuasive 16 E. N. A. MENSAH ET AL. communication, thus acting as permeating avenues for socio-political interlocutors to propagate their ideologies and assert their relevance in discussing issues of importance to perceived audiences. There are implications for the findings of the study for the instruction of communication and media students. Conscious efforts ought to be made in exposing students to the communicative functions of interac- tional metadiscourse markers, especially in oral interactions. Moreover, instead of focusing exclusively on written academic and nonacademic genres, as has predominantly been the case, in the teaching of metadiscourse at the tertiary level, attention should be turned toward how the said markers function in other avantgarde genres including television talk shows, as this study has done. There may be possible limitations to this study. Due to time constraints, the study considered a single talk show within the Ghanaian context without comparing it to other talk shows and for a much longer period. Moreover, the unavailability of literature on metadiscourse and television talk shows in the African sub region and Ghana in particular posed some limitations to the study. Nevertheless, this study will serve as a reference to subsequent researches in the domain. The study recommends that interlocutors of talk shows be intentional in their on-air communicative engagements because ‘they mean more than what they actually say’ often availed by metadiscourse resources. Furthermore, future studies on meta- discourse and persuasion in nonacademic spoken discourses may attempt to comparatively study meta- discourse use by hosts and guests of a number of talk shows over a longer period. Again, since this study has examined interactional metadiscourse resources in the domain of television talk shows, a future study may attempt to study interactive metadiscourse resources in a similar domain. Authors’ contributions The corresponding author, Emmanuel Nii Adama Mensah, was responsible for the conception and design of the study. This included the acquisition, analysis and interpretation of data for the study, and the drafting of the paper. The coauthor, Christiana Hammond was involved with revising the paper critically for intellectual content and providing constructive feedback on the methodology used. The coauthor, Albert Agbesi Wornyo was also involved with revising the paper critically for intellectual content and the final approval of the version to be published. All authors agree to be accountable for all aspects of the work. Disclosure statement T No potential conflict of interest was reported by the author(s). Funding No funding was received for this study. About the authors Emmanuel Nii Adama Mensah holds an MPhil in Communication and Media Studies (Communication Instruction) and a Post Graduate Diploma in Education (PGDE), both from the University of Education, Winneba. He is currently a doctoral candidate in the Department of Communication Instruction at the University of Education, Winneba, Ghana. His research areas are Communication Pedagogy, Intercultural Communication, Discourse Analysis, Public Relations, and Pragmatics in Human Communication. Christiana Hammond is an Associate Professor in Intercultural Communication and Pragmatics at the School of Communication and Media Studies of the University of Education, Winneba. She has published widely on identity studies, pragmatics and intercultural communication including studies on politeness, metadiscourse, genre analysis, online dating, onomastics, cyber trolling and hate speech, and cultural shock and navigation process among others. Her research areas are Intercultural Communication, Pragmatics, Social Media Research and Academic Writing. Albert Agbesi Wornyo is an associate professor in English Rhetoric and Communication Skills in the University of Education, Winneba. With a PhD in English Communication Skills from the University of Venda, South Africa; an MPhil in Teaching English as a Second Language from the University of Ghana; and a first degree, B.A Hons in Linguistics and Psychology, also from the University of Ghana. His research area is Intercultural Rhetoric. Cogent Social Sciences 17 ORCID Emmanuel Nii Adama Mensah http://orcid.org/0009-0001-2873-3615 Albert Agbesi Wornyo http://orcid.org/0000-0002-1142-658X Data availability statement The data that support the findings of this study are available in YouTube at YouTube.com. These data were derived from the following resources available in the public domain: https://drive.google.com/drive/folders/15Sz1NQbjHDLo4 ODIKJM0E3eMVBxNRRFW?usp=drive_link. The data that support the findings of this study are available from the corresponding author, E.N.A.M., upon reasonable request. References Ädel, A. (2010). Just to give you kind of a map of where we are going - A taxonomy of metadiscourse in spoken and written academic English. Nordic Journal of English Studies, 9(S2), 69–97. https://doi.org/10.35360/njes.218 Al-Ali, M. N., & Hamzeh, S. M. (2024). Extra cues extra views: A multimodal detection of Arabic clickbait thumbnail verbo-visual cues. Discourse & Communication, 18(1), 3–27. https://doi.org/10.1177/17504813231190332 Al-Anbar, K., Alghazo, S., Jarrah, M., & Altakhaineh, A. R. M. (2023). First language and second language English edito- rialists’ use of interactional metadiscourse. Discourse and Interaction, 16(2), 5–28. https://doi.org/10.5817/DI2023-2-5 Alghazo, S., Al Salem, M. N., Alrashdan, I., & Rabab’ah, G. (2021). Grammatical devices of stance in written academic English. Heliyon, 7(11), e08463. https://doi.org/10.1016/j.heliyon.2021.e08463 Alghazo, S., Al-Anbar, K., Altakhaineh, A. R. M., & Jarrah, M. (2023). Interactive metadiscourse in L1 and L2 English: Evidence from editorials. Topics in Linguistics, 24(1), 55–66. https://doi.org/10.2478/topling-2023-0004 Alghazo, S., Al-Anbar, K., Rabab’ah, G., Abusalim, N., & Rayyan, M. (2024). The construction of stance in English and Arabic newspaper editorials: A case study. Humanities and Social Sciences Communications, 11(1), 1–10. https://doi. org/10.1057/s41599-024-03418-2 Alghazo, S., Al Salem, M. N., & Alrashdan, I. (2021). Stance and engagement in English and Arabic research article abstracts. System, 103, 102681. https://doi.org/10.1016/j.system.2021.102681 Biri, Y. (2017). Metadiscourse in online opinion texts: Exploring variation within a genre [MA thesis]. University of Helsinki. ]. Byro, J. (2017). Beautiful forcefields: Promotional Metadiscursive Language in eSports Commentaries. http://urn.kb.se/res olve?urn=urn:nbn:se:kau:diva-64776 Creswell, J. W., & Creswell, J. D. (2017). Research design: Qualitative, quantitative, and mixed methods approaches. Sage Publications. Crismore, A. (1983). Metadiscourse: What it is and how it is used in school and non‑school social science texts. Center for the Study of Reading Technical Report; no. 273. Duwila, S., & Probowati, Y. (2021). Personal metadiscourse features on TED talks by British speakers. Advances in Social Science, Aducation and Humanities Research, 546, 520–526. https://doi.org/10.2991/assehr.k.210427.079 Ells, K. (2017). Inform, persuade, other: Exhuming epideictic for speech and writing instruction. Florida Communication Journal, 45(2), 1–12. https://www.researchgate.net/publication/339434934 Esmer, E. (2017). Interpersonal metadiscourse markers in Turkish election rally speeches delivered by Pro-Turkish and Pro- Kurdish Leaders. Athens Journal of Social Sciences, 4(4), 367–384. https://doi.org/10.30958/ajss.4-4-2 GeoPoll. (2019, July 10). Ghana Media Ratings, 2018. Geopoll. https://www.geopoll.com/blog/ghana-media-measurement- top-tv-radio-2018. Herzuah, P. A. A. (2018). Enacting personal identity through language: An exploration of some profile statuses of WhatsApp messenger. Journal of Communications, Media & Society, 5(1), 99–119. http://journals.gij.edu.gh/ wp-content/uploads/2018/08/JOCMAS-Enacting-Personal-Identity-through-Language-An-Exploration-of-Some-Prof ile-Status-of-WhatsApp-Messenger.pdf Higgins, C., & Walker, R. (2012). Ethos, logos, pathos: Strategies of persuasion in social-environmental reports. Accounting Forum, 36(3), 194–208. https://doi.org/10.1016/j.accfor.2012.02.003 Hyland, K. (2005). Metadiscourse: Exploring Interaction in Writing. Continuum. Hyland, K. (2017). Metadiscourse: What is it and where is it going? Journal of Pragmatics, 113, 16–29. https://doi. org/10.1016/j.pragma.2017.03.007 Hyland, K. (2019). Metadiscourse: Exploring interaction in writing. Bloomsbury Publishing. Illie, C. (2006). Talk shows. In K. Brown (Editor-in-chief) Encyclopedia of Language & Linguistics. (2nd ed.), 12, (pp.489– 494). Elsevier. Kawase, T. (2015). Metadiscourse in the introductions of PhD theses and research articles. Journal of English for Academic Purposes, 20, 114–124. https://doi.org/10.1016/j.jeap.2015.08.006 Knupfer, N. N. (2007). Descriptive research methodologies. McLellan Wyatt Digital. Kopple, W. V. (1985). Some Exploratory Discourse on Metadiscourse. College Composition & Communication, 36(1), 82–93. https://doi.org/10.58680/ccc198511781 http://orcid.org/0009-0001-2873-3615 http://orcid.org/0000-0002-1142-658X https://drive.google.com/drive/folders/15Sz1NQbjHDLo4ODIKJM0E3eMVBxNRRFW?usp=drive_link https://drive.google.com/drive/folders/15Sz1NQbjHDLo4ODIKJM0E3eMVBxNRRFW?usp=drive_link https://doi.org/10.35360/njes.218 https://doi.org/10.1177/17504813231190332 https://doi.org/10.5817/DI2023-2-5 https://doi.org/10.1016/j.heliyon.2021.e08463 https://doi.org/10.2478/topling-2023-0004 https://doi.org/10.1057/s41599-024-03418-2 https://doi.org/10.1057/s41599-024-03418-2 https://doi.org/10.1016/j.system.2021.102681 http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-64776 http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-64776 https://doi.org/10.2991/assehr.k.210427.079 https://www.researchgate.net/publication/339434934 https://doi.org/10.30958/ajss.4-4-2 https://www.geopoll.com/blog/ghana-media-measurement-top-tv-radio-2018 https://www.geopoll.com/blog/ghana-media-measurement-top-tv-radio-2018 http://journals.gij.edu.gh/wp-content/uploads/2018/08/JOCMAS-Enacting-Personal-Identity-through-Language-An-Exploration-of-Some-Profile-Status-of-WhatsApp-Messenger.pdf http://journals.gij.edu.gh/wp-content/uploads/2018/08/JOCMAS-Enacting-Personal-Identity-through-Language-An-Exploration-of-Some-Profile-Status-of-WhatsApp-Messenger.pdf http://journals.gij.edu.gh/wp-content/uploads/2018/08/JOCMAS-Enacting-Personal-Identity-through-Language-An-Exploration-of-Some-Profile-Status-of-WhatsApp-Messenger.pdf https://doi.org/10.1016/j.accfor.2012.02.003 https://doi.org/10.1016/j.pragma.2017.03.007 https://doi.org/10.1016/j.pragma.2017.03.007 https://doi.org/10.1016/j.jeap.2015.08.006 https://doi.org/10.58680/ccc198511781 18 E. N. A. MENSAH ET AL. Kuhi, D., & Behnam, B. (2011). Generic variations and metadiscourse use in the writing of applied linguists: A com- parative study and preliminary framework. Written Communication, 28(1), 97–141. https://doi.org/10.1177/07410883 10387259 Lee, J. J., & Casal, J. E. (2014). Metadiscourse in results and discussion chapters: A cross-linguistic analysis of English and Spanish thesis writers in engineering. System, 46, 39–54. https://doi.org/10.1016/j.system.2014.07.009 Le, E. (2004). Active participation within written argumentation: Metadiscourse and editorialist’s authority. Journal of Pragmatics, 36(4), 687–714. https://doi.org/10.1016/S0378-2166(03)00032-8 Lee, W. L. (2021). Emotions in business communication: A comparative study of attitude markers in the discourse of U.S and mainland Chinese corporations. Discourse & Communication, 15(6), 629–649. https://doi.org/10.1177/175048 13211026541 Lindlof, T. R., & Taylor, B. C. (2017). Qualitative Communication Research. Sage. Lundell, A. K. (2014). Cross-platform television: Superliveness, metadiscourse and complex audience orientation in a sports journalism production on the web. Northern Lights: Film & Media Studies Yearbook, 12(1), 11–27. https://doi. org/10.1386/nl.12.1.11_1 Mai, H. (2016). An intercultural analysis of meta-discourse markers as persuasive power in Chinese and American political speeches. International Journal of Language and Linguistics, 4(6), 207–225. https://doi.org/10.11648/j.ijll. 20160406.13 Mur-Dueñas, P. (2011). An intercultural analysis of metadiscourse features in research articles written in English and in Spanish. Journal of Pragmatics, 43(12), 3068–3079. https://www.researchgate.net/publication/251586757. https:// doi.org/10.1016/j.pragma.2011.05.002 Murthy, M., & Gosal, M. (2016). A study on Aristotle’s rhetoric applied to industrial communication. British Journal of English Linguistics, 4(1), 64–74. https://doi.org/10.37745/bjel.2013 Myjoyonline.com. (2020). June 29). Full list of 64 award winners at 25th GJA Awards. Myjoyonline. https://www. myjoyonline,com/full-list-of-64-award-winners-at-25th-gja-awards/ Obeng, B. (2019). A comparative analysis of the rhetorical moves and metadiscourse elements in abstracts of science conferences [MPhil thesis]. University of Education. Ogbidi, J., Basil, G., & Bassey, A. E. (2016). Marketing communication strategies and voters’ behaviour in electioneer- ing process in Cross River State. Nigeria. Global Journal of Management and Business Research (A), 16(11), 52–62. https://www.researchgate.net/publication/328230366 Okoto, E. (2020). Discourse approach to talk shows in Ghana: A study of Newsfile and Agenda [MPhil thesis]. University of Education. Patton, M. Q. (2002). Qualitative Research and Evaluation Methods. (3rd ed.). Sage. Perez, F. M. (2014). Cultural values and their correlation with interactional metadiscourse strategies in Spanish and US Business Websites. Atlantis, 36(2), 73–95. https://www.jstor.org/stable/43486661 Rababáh, G., Yagi, S., & Alghazo, S. (2024). Using metadiscourse to create effective and engaging EFL virtual class- rooms during the Covid-19 pandemic. Iranian Journal of Language Teaching Research, 12(1), 107–129. https://doi. org/10.30466/ijltr.2024.121421 Salvador, J. T. (2016). Exploring quantitative and qualitative methodologies: A guide to novice nursing researchers. European Scientific Journa, 12(18), 107–122. 7 https://doi.org/10.19044/esj.2016.v12n18p107 Sivanya, N. (2019). Metadiscourse as rhetorical strategy in YouTube review videos. 3rd English Language. And Literature International Conference (ELLiC) Proceedings, 3. 315–323. https://jurnal.unimus.ac.id/index.php/ELLIC/article/view/ 4726/4253 Taillard, M. (2000). Persuasive communication: The case of marketing. Working Papers in Linguistics, 12, 145–174. https://www.phon.ucl.ac.uk/home/PUB/WPL/00papers/taillard.pdf Ting, S. H. (2018). Ethos, logos and pathos in university students’ informal requests. Journal of Language Studies, 18(1), 234–251. https://doi.org/10.17576/gema-2018-1801-14 Torto, R. T. (2020). Aristotelian rhetorical theory as a framework for analysing advertising texts in the print media in Ghana. Theory and Practice in Language Studies, 10(3), 269–283. https://doi.org/10.17507/tpls.1003.02 Trajkova, Z. (2014). Active participation within written and spoken argumentation: The use of engagement markers across different genres. In A. Akbarov (Ed.). Linguistics, Culture and Identity in Foreign Language Education. (pp.1909– 1915). IBU. Turiman, S., Abdullah, N., & Noor, N. M. (2018). Spoken metadiscourse in Malaysian ESL Job interviews. GEMA Online® Journal of Language Studies, 18(3), 135–152. https://doi.org/10.17576/gema-2018-1803-08 Wei, J., Li, Y., Zhou, T., & Gong, Z. (2016). Studies on metadiscourse since the 3rd Millennium. Journal of Education and Practice, 7(9), 194–204. https://core.ac.uk/download/pdf/234638613.pdf Wildermuth, B. M., & Zhang, Y. (2014). Qualitative analysis of content. In B. Wildermuth (Ed.), Applications of Social Research Methods to Questions in Information and Library Science, (pp. 308–319). Libraries Unlimited. Williams, J. M. (1981). Style: Ten lessons in clarity and grace. Scott, Foresman and Company. Yang, S. B., Lee, H., Lee, K., & Koo, C. (2018). The application of Aristotle’s rhetoric to the sharing economy: An em- pirical study of Airbnb. Journal of Travel & Tourism Marketing, 35(7), 938–957. https://doi.org/10.1080/10548408.2018. 1455622 https://doi.org/10.1177/0741088310387259 https://doi.org/10.1177/0741088310387259 https://doi.org/10.1016/j.system.2014.07.009 https://doi.org/10.1016/S0378-2166(03)00032-8 https://doi.org/10.1177/17504813211026541 https://doi.org/10.1177/17504813211026541 https://doi.org/10.1386/nl.12.1.11_1 https://doi.org/10.1386/nl.12.1.11_1 https://doi.org/10.11648/j.ijll.20160406.13 https://doi.org/10.11648/j.ijll.20160406.13 https://www.researchgate.net/publication/251586757 https://doi.org/10.1016/j.pragma.2011.05.002 https://doi.org/10.1016/j.pragma.2011.05.002 https://doi.org/10.37745/bjel.2013 https://www.myjoyonline,com/full-list-of-64-award-winners-at-25th-gja-awards/ https://www.myjoyonline,com/full-list-of-64-award-winners-at-25th-gja-awards/ https://www.researchgate.net/publication/328230366 https://www.jstor.org/stable/43486661 https://doi.org/10.30466/ijltr.2024.121421 https://doi.org/10.30466/ijltr.2024.121421 https://doi.org/10.19044/esj.2016.v12n18p107 https://jurnal.unimus.ac.id/index.php/ELLIC/article/view/4726/4253 https://jurnal.unimus.ac.id/index.php/ELLIC/article/view/4726/4253 https://www.phon.ucl.ac.uk/home/PUB/WPL/00papers/taillard.pdf https://doi.org/10.17576/gema-2018-1801-14 https://doi.org/10.17507/tpls.1003.02 https://doi.org/10.17576/gema-2018-1803-08 https://core.ac.uk/download/pdf/234638613.pdf https://doi.org/10.1080/10548408.2018.1455622 https://doi.org/10.1080/10548408.2018.1455622 We mean more than what we say on air: interactional metadiscourse resources in television talk shows ABSTRACT Introduction Persuasive communication Metadiscourse and persuasive communication Materials and methods Sampling Results and discussion Kinds of interactional metadiscourse resources used on the television talk show Communicative functions of interactional metadiscourse resources used on the talk show Self-mentions Engagement markers Boosters Attitude markers Hedges Interactional metadiscourse resources and the realization of persuasive appeals Interactional metadiscourse resources and the realization of credible appeal (ethos) Similitude Deference Expertise Interactional metadiscourse resources and the realization of affective appeal (pathos) Identification/sociality Imagery Implorations Conclusion and recommendations Authors contributions Disclosure statement Funding About the authors ORCID Data availability statement References