For over a generation, social scientists have tried to categorize the relationship between journalists and politicians. Which side holds power and influence over the other? Some scholars propose “active” theories: journalists have preferences and the power to impose them on anyone seeking media attention. Other scholars argue journalists are essentially “reactive,” dutifully writing down what politicians say with little ability to add alternate perspectives. In this dissertation, I propose both camps are extremes based on a fundamental misunderstanding of how journalists can apply their preferences on news content. Politicians and other sources provide information to reporters, bloggers and other new media writers. Each writer then chooses how to respond to this information. Journalistic power – whether we are discussing traditional media outlets or newer partisan media organizations – is best understood as a set of if : then propositions.

The empirical sections of the dissertation consist of three separate studies, each of which focuses on one set of inputs and the output from a particular set of news organizations. The first study focuses on how presidents schedule press conferences at particular times and places. I find scheduling influences how much attention journalists give a conference, which in turn influences the balance of opinion found in stories. The second study shows how journalists resist but may ultimately give in to evasive responses, by examining quotations on a statement-by-statement basis. The third study examines some of the most popular phrases from the 2008 election, comparing how a wide range of media organizations responded to the same set of political and non-political ideas. Put together, these studies offer a common theoretical framework for comparing traditional and new media organizations, allowing for commonalities as well as differences.

In the last twenty-five years the media has undergone major changes, moving fully from a broadcast model to cable and beyond. One effect of this change has been shifts in how politicians reach voters, particularly when campaigning. This has lead to politicians seeking out voters in venues that would have previously been considered beneath the dignity of their office. This, along with other shifts, has lead some critics to suggest that news is becoming corrupted, but others have suggested that news in merely evolving along with technology and culture. This project seeks to understand some of these shifts by investigating both entertainment talk shows, which have become a locus of political interviews, as well as interviews on cable news. Using qualitative and quantitative methods this project seeks to chart the norms of the talk show interview, how these norms shift in the face of a political interviewee, and how cable news norms differ from or not broadcast news norms. Using conversation analysis, chapter three of this dissertation finds two norms drive the talk show: personalization and congeniality. These contrast with the norms that drive news interviewing: neutralism and adversarialness. In the chapter four interviews are coded and compared statistically, showing that when a political guest appears on a talk show a mix of norms are used. The norm most closely associated with the genre, personalization, remains strong, while the norm more associated with the treatment of the guest, congeniality, is more likely to be dropped in favor of its news counterpart, adversarialness. In the fifth chapter, the same measures are used to compare broadcast and cable news, finding that there are few significant differences between the two.

Grand corruption occurs when high-level officials take advantage of government positions to benefit themselves or their cronies. Corruption scandals sometimes give us a glimpse into these acts and associations. Yet due to their illicit nature and secretive arrangement by members of political and economic elites, little is known about the everyday procedures that situatedly organize grand corruption in practice. The dissertation applies conversation-analytic methods to investigate the role of language and social interaction in framing and shaping allegedly corrupt acts exposed in Peru’s 2018 “CNM Audios” scandal, a notorious case of grand judicial corruption. Using audios from police wiretapping circulated by investigative journalists as a sui generis source of observational data, the dissertation describes a constellation of speaking practices - a “pragmatics” - whereby alleged conspiracy and influence trafficking are talked about and concertedly organized as favor-doing between friends. By analyzing how allegedly corrupt behavior is situatedly embedded in mundane practices and structures, like gift and favor exchanges across informal networks, the dissertation shows the work of everyday talk in both realizing and obfuscating illegitimate deals and associations.

The dissertation is organized in five chapters. Chapter 1 situates the research project among extant sociological scholarship on corruption obfuscation and everyday morality in social interaction. The chapter also offers an overview of Peru’s “CNM Audios” scandal and a description of the data and methods. Chapter 2 explores the category amigo (“friend”) in describing and referring to persons in associates’ networks, showing the practical operation of the ordinary category to obscure ties and talk in allegedly corrupt undertakings. Chapter 3 describes “doing ‘being close’” as a practice for displaying a relational and affective stance when greeting associates over the phone and explores its work in furnishing a communicative situation that is conducive to making personalized arrangements and requesting favors. Chapter 4 examines a set of phone calls in which associates request help with institutional matters to showcase the delicacy of doing and exchanging allegedly corrupt favors in the scandal. Lastly, chapter 5 offers a summary of findings, discussing the obfuscatory work of the constellation of speaking practices as a “pragmatics” of corruption, and highlights the dissertation’s contributions to corruption scholarship, economic sociology, and conversation analysis.

Studies examining US prenatal care have identified problem screening and information giving as the doctor’s main agenda, and receiving information as patients’ main goal, however few studies have examined the interactional practices of prenatal care to determine how physician and patient goals and orientations become manifest in the visit and how they are implemented in practice. This study applies the methods of conversation analysis to 30 prenatal consultations to uncover how patients’ desires, views, and concerns manifest in discussion, and how the doctor navigates the balance between patient concerns and a medical agenda. An analysis of ultrasound introductions shows how the doctor and her patients index a secondary orientation to the ultrasound as a valid means to satisfy patient curiosity, while simultaneously maintaining a primary orientation to the ultrasound as a medical tool. In addition, both parties may exploit patient curiosity as a cover for medical concerns. Examining different points of future decision making reveals that the doctor implicates varying levels of patient agency in determining the course of care. These differing levels of agency appear to vary systematically with the type of decision under discussion, apparently owing to patient experience, professional recommendations, and larger cultural attitudes. Finally, the doctor and patients’ regular use of particular interactional strategies creates an environment in which worry is obscured or circumvented. This allows for the doctor to monitor the patient and share information with her without triggering alarm. Taken together, these chapters help to elucidate how important prenatal care objectives are pursued with due sensitivity to patient interests and concerns.

This dissertation aims to discover the systematic properties of ‘knowledge-testing’ questions (KQs) to broaden our understanding of the action of knowledge-testing accomplished via questions. The data of this study consist of 190 KQs obtained from cross-examinations among political candidates in 116 Korean political campaign debates (155 hours and 23 minutes in total). The defining features, emergence, and compositional forms of KQs are explored in great detail using the methodological and analytic framework of conversation analysis (CA).

To clearly define KQs, this study examines the interactional functions of KQs, and their production and recognition as KQs during the cross-examinations. The findings show that candidates use KQs to discredit their opponents and expose their ignorance of the subject of inquiry while promoting their own knowledgeability. Candidates also utilize linguistic, sequential/interactional, and broad social resources for action formation and ascription of KQs and can retrospectively transform the action import of KQs.

The investigation of the emergence of KQs during the cross-examinations show that they occur in three distinctive but potentially co-existing contexts: (a) where candidates have good reason to believe that the respondents are vulnerable to KQs, (b) where candidates are involved in undermining their opponents’ policies/promises/claims, and (c) where candidates fail to provide the correct answers to the KQs issued by their opponents earlier and subsequently attempt to show that their opponents have the same weakness as themselves regarding a domain of knowledge. Candidates therefore find KQs to be a useful debate tool to primarily expose their opponents’ ignorance, which can serve as a basis for undermining their opponents’ promises/claims or doing damage control via counterattack. The different sequential contexts contribute to diversifying the import of KQs as a political campaign debate strategy and point to the close relationship between sequential position and action import.

The compositional forms of the KQs are analyzed by identifying the parameters (or dimensions) that explain linguistic variation in KQ design and the import delivered by each KQ format in testing knowledge. This study identifies two parameters of KQ design and their respective formats: the epistemic parameter accounting for nine types of KQ formats, and the precision parameter accounting for three types of KQ formats. The nine epistemic KQ formats convey different degrees of expectation towards the likelihood of the opponents’ knowing the answers, which range from strongest pessimism to strongest optimism. The three precision formats realize to varying degrees how precise an answer the KQs will accept. Acceptable answers as conveyed by these formats range from approximate to precise and accurate. The sequential and frequency analyses of the epistemic and precision formats show the variation in their surface forms to be an accountable and systematic phenomenon geared towards maximizing the effectiveness of KQs as a linguistic device for testing the opponents’ knowledge and proving their ignorance.

As a rare conversation analytic study providing a detailed and comprehensive analysis of KQs used in political campaign debates—a type of question crucial in constituting various forms of institutional interactions such as pedagogical interactions, news interviews with political candidates, job interviews, and legal proceedings, this dissertation adds to our understanding of the institutional uses of KQs, linguistic strategies for political campaign debates, and question design in Korean, thereby contributing to the fields of conversational analysis, political communication, and linguistics.

The purpose of this study is to investigate, through analyses of videotaped records, how presidents of the United States interact with members of the White House press corps during presidential news conferences. The study is divided into two main parts. The first part examines practices of interaction in this institutional setting. One main objective is to determine when and how the president of the United States starts reacting to the propositions and presuppositions contained in the journalists questioning turns, or put slightly differently, what features of a journalist's questioning turn trigger non-vocal responses from the president that show his stance on the question he is listening to. In addition, this part of the dissertation also analyses refusals to answer without an account. More specifically, it studies how a president manages a refusal to answer a question without an account by defusing the fact that he is not being accountable. The president uses various semiotic resources to accomplish this task and the chapter provides a detailed analysis of these practices. The second part of this dissertation builds on studies that have documented a rise in aggressiveness in presidential news conferences (Clayman and Heritage, 2002b; Clayman et al., 2006, 2007, 2010) and zeroes in on a particular strategy the White House employed starting in the George H.W. Bush presidency to counteract this trend: holding joint press conferences--where the president appears with another head of state and answers only two or three questions. These types of conferences have surpassed solo press conferences in frequency. Yet, a key question about the effectiveness of this strategy has yet to be answered: how has this change in participation framework affected micro-level journalistic practices? In other words, do journalists address less aggressive questions to the president in joint press conferences than in solo conferences? With these two major objectives, this study contributes to the growing literature on stance-taking in interaction and the research on president-press relations. Employing qualitative and quantitative methods, this dissertation seeks to study press-state relations to show how these two bodies relate to one another.