A functional, analysis of televised U.S. Senate and gubernatorial campaign debates.

• Author: Benoit, William L.

Introduction

Televised political debates have emerged as an important part of our system of democracy. The first presidential primary debate, broadcast on radio, featured Dewey and Stassen in a primary campaign (see Benoit et al., 2002). The first presidential debate in a U.S. general campaign occurred in 1960 when Nixon and Kennedy participated in four nationally televised debates. Although general campaign debates in presidential elections were not held again until 1976, they have since become an accepted part of the American presidential campaign. Debates in campaigns for other, non-presidential, offices are becoming increasingly common as candidates for the U.S. Congress and for state governors (and other offices as well) use this message form to communicate with voters. In 2006, Meet the Press hosted six debates between candidates for the U.S. Senate, another indication of the importance of this campaign message form. Furthermore, broadcasts of debates within the state and the practice of digitizing non-presidential debates so they can be viewed or listened to on-line combine to increase their reach (see Minnesota Public Radio, 2002).

The fact that debates are typically organized around questions means that one can argue that political campaign debates function more like press conferences than debates (Auer, 1962; Jamieson & Birdsell, 1988; Zarefsky, 1992). Nevertheless, there can be no doubt that these messages are important for several reasons. First, debates feature the leading candidates discussing the same topics at the same time, helping citizens to compare the candidates as they make their vote choice. Second, non-presidential debates range from 30 to 90 minutes in length providing voters an extended opportunity to learn about the candidates, particularly compared with television spots (most of which are 30 seconds long). Candidates usually prepare for debates; however, the fact that candidates can be asked unexpected questions or be the target of unanticipated attacks from opponents, could mean that debates may provide a more candid view of the candidates than other message forms. Fourth, the direct confrontation affords candidates an opportunity to correct misstatements, accidental or otherwise, from their opponents. Such clash may give voters deeper insights into the issue at hand.

The Racine Group (2002) concluded that research supports the claim that watching political debates leads to voter learning. Meta-analysis reveals that presidential debates have important effects on voters, creating issue knowledge, influencing perceptions of the candidates' character, and at times altering vote choice (Benoit, Hansen, & Verser, 2003). These encounters provide an opportunity for interested voters to learn about the candidates (both directly and via press reports); they also provide a chance for candidates to present and clarify their platforms--as well as to alter the direction of their campaign if needed. Unquestionably, political debates merit scholarly attention.

Most political communication research focuses on presidential campaigns, particularly debates (for example, books on presidential debates include Benoit et al., 2002; Benoit & Wells, 1996; Bishop, Meadow, & Jackson-Beeck, 1980; Bitzer & Rueter, 1980; Carlin & McKinney, 1994; Friedenberg, 1994, 1997; Hellweg, Pfau, & Brydon, 1992; Hinck, 1993; Jamieson & Birdsell, 1988; Kraus, 1962, 1979, 2000; Lanoue & Schrott, 1991; Martel, 1983; Schroeder, 2000; or Swerdlow, 1987). Some scholars have investigated non-presidential political campaigns (e.g., Herrnson, 1998; Jacobson, 2001; Kahn & Kenney, 1999). However, Senate and gubernatorial political campaign discourse generally and campaign debates for these offices in particular deserve scholarly attention. Next to the president and vice-president, Senators should be considered the most powerful elected federal officials; governors are arguably the most powerful state officials. Campaign communication for these offices deserves scholarly attention.

Several studies suggest that non-presidential debates can influence viewers. Philport and Balon (1975) investigated the Democratic primary contest between John Glenn and Howard Metzenbaum for Ohio Senate in 1974, reporting that Glenn's image had been influenced by the debate. Lichtenstein (1982) reported that non-presidential debates were viewed as more useful than presidential debates. Just, Crigler, and Wallach (1990) indicated that viewers learned about the issue positions in a Connecticut House debate in 1984. A study by Bystrom, Roper, Gobetz, Massey, and Veal (1991) concluded that viewing an Oklahoma gubernatorial debate in 1990 influenced perceptions of the candidates' images and issue positions and, particularly for undecided viewers, assisted in the decision-making process. Hullett and Louden (1998) reported that viewers of a 1994 Congressional debate tended to recall more statements from their preferred candidate. Robertson, studying the 2004 South Dakota Senate race, found that these debates influenced viewer vote choice, perceptions of the candidates' character, and preferences for candidate policy positions. So, non-presidential debates have the potential to influence viewers.

Of course, fewer people watch non-presidential than presidential debates. However, far fewer votes are at stake in a given Senate or gubernatorial election than in a presidential election. Furthermore, as Graber (1989) explained non-presidential candidates are often overlooked. Similarly, Stempel (1994) observed that there is greater coverage of presidential than state or local campaigns. This could mean that non-presidential debates have a greater potential to inform and influence those who do view them because voters know less about non-presidential than presidential candidates (see also Bystrom, Roper, Gobetz, Massey, & Beall, 1991). It is possible that those who watch debates have greater interest in the election than non-viewers, which means debates have to potential to inform and influence those citizens who are most likely to vote.

For the most part research on non-presidential campaign discourse tends to focus on television advertising, so non-presidential debates have yet to receive much scholarly attention. McKinney and Carlin (2004) lament the paucity of scholarship on non-presidential debates. Pfau (1983), in a very early study of non-presidential debates, investigated format in debates. Ornstein (1987) presents a conceptual discussion of non-presidential debates instead of research on debate content or effects. A rhetorical analysis of narrative form in the 1984 Helms-Hunt Senate debate was offered by Conrad (1993). Johnson (1996) investigated intertextuality in a gubernatorial and a senate debate. Airne and Benoit (2005) content analyzed the 2004 Senate debates between Obama and Keyes: 59% of the statements were acclaims, 37% were attacks, and 4% were defenses; policy was discussed more often than character (65% to 35%). Banwart and McKinney (2005) analyzed two U.S. senate and two gubernatorial debates from 2000 and 2002, reporting that positive comments (79%) and policy discussion (82%) dominated these encounters. These studies are a useful beginning, but the sample of debates, candidates, and years is still very small.

This project will content analyze U.S. Senate and gubernatorial debates using Functional Theory in order to extend our understanding of the nature of U.S. Senate political campaign debates. Functional Theory investigates both the functions (acclaims, attacks, defenses) and topics (policy, character) of political campaign messages. The functions are important because they give voters reasons to prefer one candidate over another (e.g., acclaims can increase a candidate's apparent preferability, attacks may reduce an opponent's apparent preferability, and defenses may reduce the apparent drawbacks of a candidate). The topics are important as well, because voters need to have information about both what the candidates will do (and have done) in office as well as what kind of person or leader they are. Thus, Functional Theory investigates important aspects of the content of political campaign messages, such as debates.

Theoretical Underpinning

Benoit's (1999, 2007; Benoit, Blaney, & Pier, 1998; Benoit et al., 2003) Functional Theory of Political Campaign Discourse will provide the theoretical underpinning for this study. Political campaign messages have three functions: acclaims, which praise the candidate; attacks, which attack the opponent; and defenses, which refute attacks. Of course, third party candidates and some candidates in primary campaigns may run merely to champion a cause; Functional Theory focuses on candidates whose primary goal is to seek elective office. Such candidates must try to persuade voters that they are preferable to opponents and preferability can only be established by suggesting that one is desirable (acclaims), that an opponent is undesirable (attacks), or that alleged weaknesses are incorrect (defenses).

Campaign discourse can occur on two topics: policy (governmental action and problems amenable to governmental action) and character (the qualities and abilities of the candidates). Benoit and Wells (1996) argue that a candidate's position on an issue could influence perceptions of that candidate's character and a candidate's character could suggest that the candidate favors certain policy positions. For example, a candidate's statement advocating proposals (policies) to help the homeless may foster the impression that the candidate is a caring person (character); similarly, a candidate who stresses (non-governmental) service in the armed forces (character) could create the impression that he or she will support a strong defense (policy). Despite the fact that some voters could draw other inferences (i.e., forming impressions about a candidate's character from policy statements or vice versa)...