CC BY
640
Benoit W.L.
Штат Миссурий
THE FUNCTIONAL THEORY OF POLITICAL CAMPAIGN DISCOURSE
Many countries have political campaigns for elective office. For example, political campaign debates have been held in many countries, including Australia, Canada, France, Germany, Greece, Holland, Israel, New Zealand, Scotland, South Korea, Sweden, Poland, Taiwan, the Ukraine, and the United States. Political television advertisements have been broadcast in countries around the world, including Britain, France, Germany, Greece, Israel, Italy, Korea, Poland, Russia, Taiwan, Turkey, and the United States. Of course, political systems and election campaigns have some differences. For example, some nations cast votes for candidates; others have a parliamentary system and vote for a political party rather than directly for a candidate. Rules about who can vote, registering to vote, as well as when and how candidates, political parties, and others can campaign also vary from country to country. For example, in South Korea the president is restricted to one term in office and there is no vice president. This aspect of that system of government the importance of incumbency in that country (see Lee & Benoit, 2005). These cross-national differences can be important and we must remain aware that such variations exist. Nevertheless, the basic situation -political candidates, parties, and organizations try to convince enough voters to select me or my political party - ensures some commonalities. Furthermore, as differences in the media systems around the world diminish [Hallin & Mancini, 2005], the comparability of many campaign message forms increases. For example, political debates and television advertising are regulated by laws, but such events can be considered more media events than inherent aspects of the governmental system.
This essay explains the Functional Theory of Political Campaign Discourse, which has been used to analyze political campaign messages in the United States, and, more recently, in other countries (for a more complete discussion, see Benoit, in press). Obviously, this is not the only possible approach to understanding campaign communication, but it is one approach worth considering. There can be no question that political campaign message are instrumental, means to achieve a particular end (obtaining political office) or functional, in nature. Of course, a few candidates who have no realistic change of winning do run for office to publicize an issue, but they are in the mi-
nority. Most candidates use their campaign messages to persuade voters to choose them over opponents. Political campaigns matter -because presidents, prime ministers, and chancellors propose and implement vital policies - and Functional Theory can help understand the messages from viable candidates for office.
AXIOMS
Functional Theory of Political Campaign Discourse is based on six key assumptions. Each of these assumptions will be explained next. Those who are interested in understanding political campaigns can consider these assumptions to decide how well they fit the circumstances in the political systems of interest.
1. Voting Is a Comparative Act
When citizens cast their votes, have a fairly straightforward choice: For which candidate (or political party) shall I vote? This is a choice between two (or more) competing options and it clearly involves a comparative judgment. It is unreasonable to expect any candidate or party to be perfect: It is possible that a voter would agree with a candidate or party on many, but not all, issues. Candidates and political parties may have both good and bad points. So, each person is essentially deciding which candidate (or party) seems to be preferable on whatever factor matters the most to that person.
2. Candidates Must Distinguish Themselves from Opponents
The first assumption of Functional Theory, that voting is a comparative act, entails the second one: Because voting is a comparative action, candidates must tell voters how they differ from other candidates. If a citizen does not see any difference between two candidates, he or she has no reason to prefer one candidate over the other. This means that political candidates who run in contested races must develop and disseminate clear distinctions between themselves and viable opponents.
This does not mean that candidates for elective office must differ on every possible point of comparison. Surely all candidates would want to support a strong national defense, plentiful jobs, low inflation (although they might disagree on the means of achieving some of these goals). For this reason we should anticipate some similarities in the issue positions of the candidates. Still, if voters are to have a basis for choosing one candidate instead of another, the candidates must make clear to voters some points of distinction.
3. Political Campaign Messages Are Important Vehicles for Distinguishing between Candidates
The first two assumptions lead directly to the third: Candidates
use campaign messages to convey the distinctions among competitors that they chose to emphasize to voters. Some campaign messages reach voters directly, as they listen to a candidate’s speech, view a television spot, read a political pamphlet, watch a televised debate, or visit a candidate’s webpage. On the other hand, some information from candidate messages (including their press releases) reaches voters through the news media, as journalists report on the campaign (and, of course, other interested people and organizations also disseminate information about the candidates). However, we must realize that research makes it clear that voters cannot always depend on the news media to provide voters with this information. Patterson and McClure (1976) concluded that in the United States in 1972 “during the short period of the general election campaign, presidential ads contain substantially more issue content than network newscasters” (p. 23). It is important to keep in mind three important facts about news coverage of campaigns: (1) the news enacts a gate-keeping function, deciding which ideas from the candidates to pass along to voters and which to ignore, (2) the news media has the opportunity to provide commentary on the candidates’ ideas, evaluating the candidates positively or negatively, and (3) the news has a strong tendency to focus on horse race issues (who is ahead, campaign strategy, and campaign events) rather than the candidates’ policy positions or qualifications for office (see Farnsworth & Lichter, 2003; Patterson, 1980; Robinson & Sheehan, 1983). The news, in addition to candidate messages, is a way for voters to learn about the candidates and their positions on the issues.
Third, many voters learn about the candidates and their policy positions from other voters in political discussion [Lenart, 1994]. I want to make clear that I do not subscribe to the theory of opinion leaders (Lazarsfeld, Berelson, & Gaudet, 1948), in which a small number of people (the “opinion leaders”) take information from candidate messages and the news and then pass that information on to other citizens. Rather, each person in a conversation may have information about the campaign that is not known to the other and they exchange that information in political discussion. Or, perhaps today I tell my friend something he did not know about the campaign and then tomorrow he will tell me something else that I did not know. So, many people learn about the political campaign from interacting with other citizens.
So, voters acquire information about the candidates and the campaign from several sources: candidate messages, news cover-
age of the campaign, messages from other people and groups, and political discussion. Keep in mind that all of these sources are very complex: candidates may give speeches, participate in debates, broadcast television commercials, appear on talk shows, create webpages, and use other messages as well (and note also that candidates can give many different speeches, participate in more than one debate, broadcast multiple TV ads, and so forth). Voters have access to television news (on multiple networks, broadcast, cable, and Internet) and various newspapers and news magazines. A citizen may discuss the campaign with several different people, each with his or her own information and point of view. We must realize that messages from different sources can reinforce each other, contradict one another, or treat different topics. Clearly, the modern political campaign occurs in a highly complex media environment with hundreds of thousands of different messages. Each voter cannot attend to each message, but the information available on candidates for elective office can be very complex (particularly campaigns for higher levels of office, where candidates tend to have more money for messages and the news is more interested in the campaign. I also want to make it clear that I do not assume that all voters (or even most voters) are keenly interested in the campaign, actively seek out information about the candidates and their positions on the issues, critically evaluate information (and the sources of information) and work hard to make a rationale voting decision. Clearly, campaign interest and information seeking about the candidates and parties varies widely from citizen to citizen. Some people never seek out information about the candidates; if they vote, they do so on the basis of whatever information they happen to encounter during the campaign. Nevertheless, campaign messages can inform and persuade voters and clearly merit scholarly attention.
4. Candidates Establish Preferability Through Acclaiming, Attacking, and Defending
Of course, it is not enough for political candidates to be distinctive in their messages; they must strive to appear to be different from opponents in ways that will attract voters. Popkin (1994) explains that “candidates manage to get a large proportion of the citizenry sorted into opposing camps, each of which is convinced that the positions and interests of the other side add up to a less desirable package of benefits” (p. 8). For example, a statement like “I am the only candidate who wants to eliminate our national defense” would definitely differentiate that candidate from opponents, but this distinc-
tion is not likely to win many votes for that candidate. So, candidates must be distinctive from opponents in ways that make them appear preferable to other contenders.
This need for a political candidate to appear preferable to voters means that campaign messages have three potential functions. First, a candidate’s message may acclaim, or tell voters about his or her good points (see Benoit, 1997). Stressing a candidate’s desirable attributes or policies can make that candidate appear preferable to opponents, particularly for citizens who value that attribute or policy. So, one way to increase the impression that one candidate is preferable is to send messages which acclaim, or emphasizes his or her desirable features. Of course, candidates often proclaim the same goals: creating jobs, reducing inflation, protecting the country. An acclaim can make a candidate appear preferable, but it may not do so.
A second potential function of political campaign messages is to attack or criticize the opponent. Identifying an opponent’s weaknesses or disadvantages is likely to make that candidate appear less enticing to voters (again, particularly with voters who believe the topic of the attack is important). This means that an attack can improve a candidate’s net preferability by reducing his or her opponent’s apparent desirability: The worse my opponent looks, the better I look by comparison (and remember voting is a comparative decision).
Some evidence indicates that voters tend to consider policy attacks more acceptable than character attacks (Johnson-Cartee & Copeland, 1989), so some attacks might be more likely to backfire than others. Other research reports that positive advertisements are just as persuasive as negative ones (Allen & Burrell, 2002; Lau, Sigelman, Heldman, & Babbitt, 1999). Clearly, attacking is an option in a political campaign and it is capable of persuading voters that candidate is preferable. However, research does not support the notion that negative advertising is generally more effective than positive advertising.
Many voters say that they do not like mud-slinging (Merritt, 1984; Stewart, 1975). This does not mean that attacks do not work: Voters may be persuaded to downgrade the target of an attack even though they do not like mudslinging. However, the fact that most voters dislike mudslinging means that attacks have a potential disadvantage which does not apply to acclaims. There could be a backlash from an attack, in which voters dislike the attacker more than they dislike the
target of the attack. This possibility may lead candidates to attack less than they acclaim. Research on presidential television spots (Benoit, 1999) found that one group of candidates who attacked more were those who had trailed their opponents throughout the campaign. So, presidential candidate may attack more when they start to get desperate, when nothing else has allowed them to overtake their opponents.
The third and final possible function of political campaign messages is defense, a statement from a campaign message that refutes an attack (see Benoit, 1995). A timely and suitable defense has the potential to prevent further damage from an attack and it may help restore a candidate’s preferability from damage caused by an attack. However, it is important to keep in mind that defenses have several possible disadvantages. First, a response to an opponent’s attack could appear to place the candidate on the defensive. Candidates may prefer to appear proactive rather than reactive. Second, attacks are most likely to to occur, of course, in areas where a candidate is weakest, so responding to attacks are likely to take a candidate “off-message.” Generally it is a good idea for a candidate to spend the most time on topics where he or she is strongest. Third, a candidate must identify an attack before refuting it. Mentioning the attack might remind or inform voters of an alleged weakness of the candidate.
It is worth mentioning that politicians and their advisors understand the basic idea that campaign messages can perform these functions. For example, in 1972, one of President Nixon’s aides, H.R. Haldeman, gave this advice for Nixon’s reelection campaign: “Getting one of those 20 [percent] who is an undecided type to vote for you on the basis of your positive points is much less likely than getting them to vote against McGovern by scaring them to death about McGovern” (Popkin et al., 1976, p. 794n). So Haldeman realized that Nixon could seek votes by praising himself (acclaiming Nixon’s positive points) or by attacking his opponent (frightening them with attacks against his opponent). Another advisor, Vincent Breglio, who worked on Ronald Reagan’s successful 1980 presidential campaign, explained that “It has become vital in campaigns today that you not only present all the reasons why people ought to vote for you, but you also have an obligation to present the reasons why they should not vote for the opponent” (1987, p. 34). Clearly, candidate messages can establish preferability through the functions of acclaims, attacks, and defenses.
This means it is reasonable to analyze political campaign discourse into statements that acclaim oneself, attack one’s oppo-nent(s), and defend oneself from an opponent’s attacks. Although these three message functions do not occur with equal frequency, they are all options that candidates have available for use, they each have the potential to increase a candidates preferability, and all three functions are found in political campaign messages.
5. Campaign Discourse Occurs on Two Topics: Policy and Character
There are only two broad topics that are useful for candidates to try to distinguish themselves from their opponents: They can discuss their character (often referred to as “image”) and/or their policy stances (“issues”). One candidate might try to show himself or herself as a competent and forceful leader. Another might attempt to create the impression that he or she is honest or compassionate. Candidates can also discuss their positions (proposals) or past accomplishments on policy, such as education, jobs, national defense, or crime. In the 2004 Ukrainian political debates, for example, candidates Yanukovych and Yushchenko discussed such policy topics as the economy, wages, prices, and pensions. Policy and character are the two general topics on which candidates can try to establish their distinctiveness.
Some scholars have argued that policy is more important than character. Patterson and McClure (1976) explained that “of all the information voters obtain through the mass media during a presidential campaign, knowledge about where the candidates stand on the issues is among the most vital” (p. 49). Similarly, Hofstetter (1976) observed that “Issue preferences are the key elements in the preferences of most, if not all, voters” (p. 77). Public opinion poll data from 1976 through 2000 indicate that more American voters report that the most important determinant of their vote for president is issues rather than character (Benoit, 2003). Furthermore, Benoit (2003) found that presidential candidates who won, as a group, discussed policy more and character less in their campaign messages. This does not mean that character is unimportant; a substantial number of voters do believe that character is the most important factor in their presidential vote choice. Nevertheless, the evidence suggests that of the two possible topics, policy is more important in American presidential elections than character. Of course, if voters in the U.S. - or in other countries - believed that character was more important than policy, presumably this situation would change and political candi-
dates would be more inclined to emphasize character over policy.
6. A Candidate Must Win a Majority (or a Plurality) of the Votes Cast in an Election
Although this sounds obvious, this proposition has important implications for political campaigns generally and presidential campaigns in particular. First, it is important to realize that a political candidate does not need to persuade everyone that he or she is preferable to opponents. It is neither practical nor necessary to win the support of every citizen in an election. One reason this is important is that some policy positions (e.g., on abortion) will simultaneously attract some voters and repel others. That is, support for many issues is dichotomous. For instance, in 2000 Governor Bush wanted to permit some citizens to invest part of their Social Security in the stock market; Vice President Gore opposed this proposal. Voters were also split on this suggestion, so the presidential candidates could not hope to win the support of all voters on this issue. So, it is important to realize that political candidates do not have to obtain all of the votes cast to win the election.
Second, we must realize that in a very real sense the only citizens who really matter in an election are those who actually vote. So, a candidate does not need the support of most citizens, but of most citizens who actually vote. Candidates need not try to appeal to citizens who are not eligible to vote or those who choose not to vote (although, of course, one cannot know for certain which citizens will actually vote in the election). For this reason, some public opinion polls report results for only registered voters or likely voters.
Third, American presidential elections are unusual because candidates only need to persuade enough people to win 270 electoral votes. The importance of the Electoral College, and this idea, became obvious in the 2000 American presidential election because Al Gore won the popular vote by over 500,000 votes but lost the election, because George Bush received a majority of the Electoral College votes (Duchneskie & Seplow, 2000). The importance of the Electoral College, and the “winner-take-all” rules that operate in all but two states, has led American presidential candidates in recent elections to campaign more vigorously (e.g., spend more on airing political advertising; schedule more speeches and political rallies) in some states than others. Of course, rules for parliamentary governments, in which citizens vote for parties rather than candidates, are different.
Summary
So, presidential candidates disseminate their campaign messages through various channels or media (such as speeches, television spots, debates, direct mail advertising, radio spots, web pages) hoping to provide information to whoever happens to be paying attention to that medium at the time of the message. Acclaiming, or providing information about a candidate’s advantages (desirable issue stands or qualities) works to enhance that candidate’s perceived preferability. Attacking, or providing information about an opponent’s disadvantages (policy or character) has a tendency to reduce the opponent’s perceived preferability. Defending or refuting an attack should help restore a candidate’s perceived preferability. These three functions work as an informal form of cost-benefit analysis: When persuasive to the audience, acclaims increase a candidate’s benefits, attacks increase an opponent’s costs, and defenses reduce alleged costs. Of course, I want to make clear that I do not believe voters assign numerical values to acclaims or attacks or that they perform mathematical operations to make a vote choice (as in traditional cost-benefit analysis). I believe the three functions contribute to global evaluations of candidates. Acclaims, when persuasive, ought to increase the candidate’s apparent benefits. Attacks, if accepted by the audience, should add to the target’s perceived costs. Defenses, if effective, should reduce the candidate’s apparent costs. Finally, the cumulative effects of the messages encountered by voters (from candidates, news, and other sources) containing these three kinds of messages, along with the voters’ existing beliefs and values, should eventually determine their vote choice.
This view explains why basic ideas are, and should be, repeated throughout the campaign. For those voters who pay attention throughout the campaign, repetition serves to reinforce the candidate’s message with those auditors. On the other hand, candidates disseminate a relatively constant message in hopes that voters who rarely pay attention to the campaign will sooner or later notice their key campaign themes.
ADVANTAGES OF THE FUNCTIONAL APPROACH
The Functional Theory enjoys several important advantages over other approaches to studying political campaign communication. The assumptions of the theory allow it to make predictions about the nature of campaign messages, as we shall see. Political television spots are clearly the most intensely studied form of presidential campaign message, so I will begin by explaining the advantages of
the Functional Approach compared with other research on political advertising. I conceptualize TV ads as performing two functions, as most other research does in studying the nature of political advertising: acclaiming (positive) and attacking (negative). However, functional theory recognizes a third function, defenses, which is ignored in other research. For example, a TV spot for Vice President Richard Nixon in the 1960 campaign began by explaining that the ad contains a clear example of a defense, responding to an attack from the opponent: “President Eisenhower answers the Kennedy-Johnson charges that America has accomplished nothing in the last eight years.” This message identifies an attack against Nixon and notes that President Eisenhower will answer that attack. The spot then displayed a video clip from Eisenhower, who observed that “My friends, never have Americans achieved so much in so short a time,” clearly denying the accusation identified earlier. This message cannot be really understood by considering it either as a negative or a positive ad; clearly it contains something else. It identifies an opponent’s attack and then explicitly rejects that attack. Thus, by looking for defenses as well as attacks and acclaims the Functional Approach provides a more comprehensive understanding of political campaign messages.
Second, unlike most other research, because so many TV spots include multiple statements, we do not classify entire ads as either positive (acclaiming) or negative (attacking). Some political ads are completely positive or entirely negative, but many other messages are mixed, and that mix is not always 50/50. For example, Kaid and Tedesco (1999) offer this spot from Bill Clinton’s 1996 reelection campaign to illustrate of a negative spot:
America’s values. The President bans deadly assault weapons; Dole/Gingrich vote no. The President passes family leave; Dole/Gingrich vote no. The President stands firm: a balanced budget, protects Medicare, disabled children; No again. Now Dole resigns, leaves gridlock he and Gingrich created. The President’s plan: balance the budget, protect Medicare, reform welfare. Do our duty to our parents, our children. America’s values (p. 213; emphasis added).
I italicized the attacks and left the acclaims in plain type in this text. In point of fact, even though this messages is presented as an example of a negative ad, twice as many words are devoted to positive than negative ideas (44 to 19). Accordingly, we analyze and classify each utterance in a given commercial, providing a more pre-
cise picture of the degree to which a political spot is positive, negative, or defensive. This means that a method that classifies this spot as either positive or negative provides an inaccurate picture of the content of this spot.
Notice that some scholars include a third option for classifying television spots: positive, negative, and comparative (ads that are both positive and negative). However, we know that not every ad which combines acclaims and attacks divides them evenly (e.g., some have 25% or 10% acclaims and 75% or 90% attacks). Using three categories is a little more accurate than using two categories, but my approach of categorizing each remark (theme) as acclaiming, attacking, or defending is still superior.
The same problem, of course, arises with topics as well. Although some commercials are just about policy (issues), or only about character (image), many spots contain a mixture of these two topics. Some past research classifies the topic of ads according to the “dominant focus” of the ad. Other research counts the number of ads that “mention” an issue. Classifying each theme or statement in the ad as either policy or character provides a more accurate measurement of the content of these messages than the approach taken in other research.
Third, much research stops after classifying political spots as either policy (issues) or character (image). Our analysis extends this work by analyzing both policy and character into finer subdivisions than does most current research (I divide policy into past deeds, future plans, and general goals; I split character into personal qualities, leadership qualities, and ideals). Analysis of the 1996 presidential campaign, for example, revealed that Bill Clinton’s television spots were more comprehensive than Bob Dole’s TV ads. Clinton addressed four potential ideas (actually, he addressed more than four ideas; we focus on these particular options to make this point) in his TV spots: Clinton acclaimed his both past deeds (accomplishments) and his future plans (specific campaign promises), and his ads attacked both Bob Dole’s past deeds (failures) and his future plans (specific campaign promises). In sharp contrast, Dole’s ads acclaimed his own future plans (but rarely praised his past deeds), and Dole’s spots attacked Bill Clinton’s past deeds (but rarely criticized Clinton’s future plans). An analysis that combined past deeds and future plans together could not have detected the differences in their emphases (some research, such as issue ownership theory [see, e.g., Petrocik, Benoit, & Hansen, 2003/2004] examines particular
issues such as crime, education, jobs, or national defense).
Fourth, past research adopting the Functional approach (Benoit, Blaney, & Pier, 1998; Benoit et al., 2003) found that during the primaries, campaign messages which attack may have several different targets. In the 2004 primary, for example, Democratic primary TV spots attacked President Bush (and the Republican party), other Democratic candidates, the Washington establishment (status quo). Clearly it is important to realize whether Howard Dean was attacking Joe Lieberman, John Kerry, Dick Gephardt (all Democrats) or Republican George W. Bush. However, past research on presidential television advertising employing other approaches ignores the target of attacks in primary campaign messages.
Furthermore, the Functional approach combines analysis of function (acclaim, attack, defend) and topic (policy, character). Some studies of political TV spots only investigate one topic or the other, but not both. Furthermore, studies of other message forms, such as debates or speeches, frequently do not to examine the topic of campaign messages. Thus, the Functional Approach to political campaign research provides a more comprehensive understand of political campaign messages than other research approaches.
Finally, Functional approach has been applied to several different kinds of political campaign messages, such as televised political spots (primary and general), debates (primary and general), talk radio appearances (primary and general), web pages (primary and general), nominating convention speeches (acceptances, keynotes) and news coverage of political campaigns (see Benoit, in press). We have also begun to apply the Functional approach to non-presidential campaign messages in the U.S. (e.g., Airne & Benoit, 2005a, 2005b; Brazeal & Benoit, 2001) as well as to political campaign messages in other countries (Lee & Benoit, 2004, 2005; Wen, Benoit, & Yu, 2004). Most political campaign research focuses on television spots and debates - and predominantly on general election campaign spots and debates - or on Keynote Speeches and Acceptance Addresses. Thus, research from the Functional approach offers insight into a variety of political campaign message forms. It is surprising that the concepts which have been used to study television spots (positive, negative; issue, image) rarely are used in research on other message forms, such as debates or Acceptance Addresses. Our research is systematic and has applied the same method to each message form we analyze, allowing for comparisons across media.
PREDICTIONS AND FINDINGS
Relative frequency of the three functions. As discussed earlier, Functional Theory argues that acclaims have no drawbacks, attacks have one, and defenses have three disadvantages. Thus, it predicts that:
F1. Candidates will use acclaims more frequently than attacks and attacks more than defenses.
Research on a variety of candidate message forms over multiple American presidential campaigns has supported this prediction. The data displayed in Table 1 shows that overall, 68% of utterances in these messages are acclaims, 30% attacks, and 3% are defenses. Acclaims ranged from 49% to 85%, attacks varied from 15%-51%, and defenses comprised 0-8% of candidate statements. These three functions were significantly different, and in the predicted direction, in all but one message form (radio ads, in which there was no significant difference between acclaims and attacks). This may have occurred because most radio stations have audiences that are more homogeneous than other media, which would allow candidates to target attacks to those who might be most receptive to them.
This relationship holds true in U.S. non-presidential television spots (ads for local offices, state governor, U.S. House, and U.S. Senate). 69% of theses were acclaims (the range was 61%-72%), 31% were attacks (27%-38%), and fewer than 1% were defenses (0.5%-2%; see Table 2). Non-presidential debates in the U.S. conform to this relative emphasis for the three functions: 65% acclaims, 29% attacks, and 5% defenses (Table 3). Research on television spots and debates in other countries (much of it not using Functional Theory, so not including defenses) also reveals this basic relationship: 75% acclaims and 25% attacks (Table 4). Finally, studies of political debates in other countries (Table 5), conform to the same basic pattern: acclaims were 54%, attacks comprised 48%, and defenses constituted 8% of the utterances in these debates. There are a few exceptions (there were a few more attacks than acclaims in the 2004 Ukraine debates); however, this relationship is quite robust across years, level of elective office, message form, and country. Therefore, the evidence suggests that the relative frequency of the three basic functions of political campaign discourse is inherent in the campaign situation.
Functional Theory also makes a prediction about the relative frequency of the discussion of the two topics of political campaign discourse, policy and character. In the U.S., at least, most voters report
that policy is a more important factor in their presidential vote than character (Benoit, 2003). Benoit (2003) also reports that as a group, U.S. presidential candidates who emphasize policy more, and character less, than they opponents are significantly more likely to win elections. Because political candidates monitor voter attitudes, and the majority of U.S. voters consider policy more important in presidential votes than character, Functional Theory predicts:
F2. Presidential candidates will discuss policy more than character.
Table 6 shows that in every U.S. presidential campaign message form studied, policy (54%-78%) is discussed more often than character (22%-46%), 67% to 33% overall. This pattern also occurs in American non-presidential television spots (policy 57%, character 43%) as shown in Table 7. In American non-presidential debates, we also find policy more common than character, 72%-28% (Table 8). Overall, non-U.S. political advertising emphasizes policy over character, 59% o 41%, but in two countries the emphasis was equal (Israel 1992, Korea 2002) and in three character was more common than policy (Greece 1996, Taiwan 2000, and Turkey 1995) as Table 9 indicates. Finally, policy was stressed more than character (78% to 22%) in the non-U.S. political debates studied (Table 10). It is possible these exceptions occurred for one of two reasons: those voters considered character more important than policy and the candidates conformed to their attitudes, or the candidates may have misread public sentiment. Whatever the reason, we must keep in mind that the relative emphasis of policy and character is not as consistent outside the U.S.
Having shown that the basic ideas of Functional Theory can be employed beyond U.S. presidential campaign messages (both to other levels of American campaigns and in other countries), for reasons of space I want to cover other topics in the remainder of this essay more briefly (and I will not discuss some findings, such as the fact that the candidates who are ahead in the preference polls during the primary campaign receive more attacks than less popular candidates; Benoit et al., 2002). Communication can be conceptualized as a process with sources, a context, and media. Functional Theory has investigated each of these communication variables in the context of political campaigns and I will discuss some of that work briefly here.
Source of Political Campaign Messages
Two major political parties, Democratic and Republican, dominate American politics. Although each party, and the candidates who
affiliate with these organizations, do respond to voters and historical events, the parties do have distinct ideologies and this has been found to influence the campaign messages produced in presidential campaigns. Benoit (2004) found that although virtually all candidates discuss policy more than character, Democrats discuss policy even more, and character less, than Republicans. Democrats have a tendency to advocate governmental solutions to problems, which means they are naturally inclined to emphasize policy. Republicans often argue for smaller government and, in fact, discuss policy less than Democrats in campaign discourse. We need research from the functional perspective on the nature of discourse from political parties in other countries.
Another aspect of source is the question of whether a candidate or another person (an anonymous announcer, an “ordinary citizen,” or a public figure who endorses the candidate) is talking. Benoit (in press) shows that at the American party nominating conventions, Keynote Speeches (given by surrogates) have more attacks than Acceptance Addresses (given by the nominee). Benoit (1999) found that attacks are more common in presidential TV spots when someone other than the candidate is speaking in the ad. Benoit (in press) also notes that campaign messages sponsored by political parties include more attacks than messages sponsored by the candidates themselves. So, in general, candidates tend to make fewer attacks than other sources in political campaign messages. The theory presumably is that if there is a backlash from the attack (recall the one potential drawback of attacks is that many voters do not like mud-slinging), it may damage the surrogate instead of the candidate. Nevertheless, this work has identified two ways in which source variables influence the messages produced by political candidates.
Finally, research from the Functional perspective has begun to contrast messages from candidates with messages from a different kind of source: journalists. Research on presidential primary (Benoit, Hansen, & Stein, 2004) and general (Benoit, Stein, & Hansen, 2004) debates shows that news coverage of these events is more negative than the messages themselves. And, of course, the news obsesses over horse race coverage (who is ahead in the polls, campaign strategy, and campaign events). For treatments of campaign news coverage, see Robinson and Sheehan (1983), Farnsworth and Lichter
(2003), and Benoit, Stein, and Hansen (2005).
Context of Political Campaigns
Research from the Functional perspective has examined two
contextual variables in political campaigns. In the U.S., candidates for president must first win the nomination of one of the two major political parties before they have a realistic chance of winning the presidency. This part of the election process is termed the primary campaign. Benoit (in press) reports that American presidential primary campaign messages have fewer attacks, and more acclaims, than messages in the general election campaign. Examination of Table 1 shows, for example, that primary TV spots have more acclaims and fewer attacks than general TV spots (and the same pattern holds true for debates and direct mail brochures in these two phases of the campaign). This probably occurs for two reasons. First, members of the same political party (who are contesting the primary) are likely to have more policy agreement than candidates of different political parties (who contest the general election). So, primary candidates, in general, have fewer policy differences to discuss. Second, the candidate who wins the primary battle wants the support of other candidates of his or her political party - and the voters who preferred those other candidates. This concern was an incentive for John Kerry not to attack Dennis Kucinich, Joe Lieberman, or other Democratic primary candidates, too much. Campaign phase influences the amount of attacking in a political campaign.
Benoit (in press) also reports that primary campaigns emphasize policy less, and character more, than general campaigns (keep in mind most candidates, even in primary campaigns, stress policy; the point here is that policy is emphasized even more in general campaigns than in primary campaigns). See Table 6 to for data on this point. Again, candidates from the same party, who contest the primary, have fewer policy differences to discuss. Furthermore, candidates in the general election (e.g., President Bush and Senator Kerry in 2004; Vice President Gore and Governor Bush in 2000; President Clinton and Senator Dole in 1996) tend to be better known than many primary candidates (e.g., Dennis Kucinich, Carol Moseley-Braun, or Al Sharpton in 2004). This means candidates have a reason to spend more time introducing themselves to the voters (emphasizing character more in the primary than the general campaign). So, campaign phase also influences the topic of political discourse.
A second contextual variable is incumbency. In many countries (although not South Korea, for example) a president, chancellor, or prime minister is allowed to seek a second office. In the U.S., presidents are limited to two terms in office. However, the Vice President usually seeks the presidency after the president has served two
terms in office, and the Vice President usually runs as an incumbent with experience in the presidency (interestingly, Vice President Dick Cheney has said he would not run for office in 2008 when President George W. Bush’s second term is complete). Benoit (in press) reported that incumbents acclaim more, and attack less, than challengers. The key difference between these two groups of candidates is that although both usually have experience in elective office, only the incumbent has experience in the office being sought in the campaign. Unlike senators, for example, presidents have administrative experience. Presidents have national defense and foreign policy experience that governors do not have. So, the incumbent’s experience is more relevant than the challenger’s experience. Both candidates discuss the incumbent’s record more than the challenger’s record, but they of course emphasize different aspects. Incumbents acclaim their own past deeds (successes) more than they attack the challengers’ past deeds (failures); challengers attack the incumbents’ past deeds more than they acclaim their own past deeds. So, whether the candidate is an incumbent, has a record in the office being sought, can influence the nature of political campaign discourse.
Medium
I also want to briefly consider potential influence from campaign medium. Table 1 shows that American presidential debates (both primary and general) have more defenses than any other medium studied. This probably occurs for several reasons. Functional theory suggests that debates have three potential drawbacks and candidates generally use relatively few defenses. However, one drawback does not apply to debates: everyone watching the debate just heard the attack which provoked a defense. This means that, unlike in other media, the candidate using defense does not have to worry that he or she will inform or remind voters of a potential weakness: the audience will have just seen the attack. Furthermore, although presidential candidates do prepare for debates, it may be more difficult to resist the urge to respond to an attack. Other message forms, such as TV spots and Acceptance Addresses, do not have an opponent who attacks before the candidate speaks in the ad or speech. Those messages are carefully scripted - with few defenses. But in the heat of debate candidates may be unable to stop from refuting attacks, even though defenses have several drawbacks. This finding suggests that there can be differences between messages given in different media.
CONCLUSION
The Functional Theory of political campaign discourse was developed and initially applied to American presidential campaign discourse. However, it has since been applied to discourse in campaigns for other levels of office in the United States (e.g., governor, senator) and in other countries. We must keep in mind that there can be important differences between elections in different countries. As mentioned earlier, there cannot be a true incumbent candidate in South Korea because the president is limited to a single term and there is no vice president in that country. Research on Israeli political debates (Benoit & Sheafer, 2006) found differences between incumbents and challengers except in 1988. However, in 1984 the vote was very close and the leader of each party served as Prime Minister for two years between the 1984 and 1988 elections. This meant that both candidates were, essentially, incumbents in 1988.
Clearly, political campaigns, including speeches, TV spots, debates, and, in recent years, other media such as candidate webpages, are being used around the world for elections. Functional theory can help understand the nature of these campaign messages. Clearly, more work needs to be done to extend this theory and improve our understanding of the messages which help voters select the leaders of their countries.
Table 1. Functions of U.S. Presidential Campaign Discourse
Acclaims Attacks Defenses
Primary TV Spots 4123 (72%) 1544 (27%) 56 (1%)
(1952-2004) Primary Debates 12525 (64%) 6084 (31%) 818 (4%)
(1948-2004) Primary Brochures 7776 (85%) 1361 (15%) 7 (0.08%)
(1948-2004) Acceptance Addresses 2193 (77%) 652 (23%) 20 (1%)
(1952-2004) Keynote Speeches 474 (50%) 463 (49%) 12 (1%)
(1960-1996) General TV Spots 3454 (59%) 2339 (40%) 71 (1%)
(1952-2004) General Debates 4050 (57%) 2501 (35%) 604 (8%)
(1960, 1976-2004) General Brochures 8036 (70%) 3393 (39%) 48 (0.4%)
(1952-2000) General Radio Spots 429 (49%) 454 (51%) 0
(1972-1992, 2000) Total 43060 (68%) 18791 (30%) 1636 (3%)
Spots Acclaims Attacks___________________Defenses
US Senate 541 2355 (66%) 1145 (34%) 22 (0.6%)
1980-2000
US House 438 1905 (68%) 867 (31%) 15 (0.5%)
1980-2000
US Congress 29 114 (61%) 68 (37%) 4 (2%)
1998
Governor 488 1444 (72%) 544 (27%) 16 (0.8%)
1974-1998
Local 70 279 (71%) 111 (28%) 5 (1%)
1998-2000
Total 1566 6097 (69%) 2735 (31%) 62 (0.7%)
X2(df=2) = 6169.24, p < .0001
Benoit (2000), Brazeal & Benoit (2001), Pier (2002)
Table 3. Functions of U.S. Non-Presidential Debates
Acclaims Attacks Defenses
U. S. Senate Debates 1803 (61%) 848 (29%) 297 (10%)
1998-2004
U. S. Governor Debates 3007 (68%) 1309 (30%) 94 (2%)
2000-2004
Total 4810 (65%) 2157 (29%) 391 (5%)
X2(df=1) = 4034.35, p < .0001
Senate debates: Benoit, Brazeal, & Airne (2006) Governor debates: Airne, Benoit, & Brazeal (2006)
Table 4. Functions of Non-U.S. Political TV Spots
Acclaims Attacks Defenses
France 1988 75% 25% -
Germany 1992 68% 32% -
Italy 1992 85% 15% -
Britain 1992, 1997 69% 31% -
Israel 1992 58% 42% -
Korea 1963-1992 67% 33% -
Poland 1995 93% 7% -
Turkey 1995 89% 11% -
Greece 1996 71% 29% -
Russia 1996 72% 28% -
Taiwan 1996 81% 19% -
Subtotal! 75% 25% -
Taiwan 2000 63% 35% 3%
Korea 2002 72% 27% 1%
*These samples included newspaper advertisements as well as TV spots.
Chang (2000), Kaid (1999), Kaid and Holtz-Bacha (1995), Lee and Benoit (2004); Tak, Kaid, and Lee (1997), Wen, Benoit, and Wu (2004)
IPercentages of spots reported by authors converted into number of spots for statistical analysis: x2(df=1)=146.84, p<.0001.
Table 5. Functions of Non-U.S. Political Debates
Acclaims Attacks Defenses
France, 1988, 1995 716 (61%) 386 (33%) 66 (6%)
Israel, 1984, 1988, 165 (50%) 124 (38%) 38 (12%)
1992,1996,1999
South Korea, 1997, 1044 (55%) 668 (35%) 180 (10%)
2002
Taiwan, 2004 320 (49%) 303 (46%) 35 (5%)
Ukraine, 2004 256 (43%) 290 (48%) 52 (9%)
Total 2501 (54%) 1771 (38%) 371 (8%)
X2(df=2) = 1514.06, p < .0001
Benoit & Klyukovski (2006), Benoit & Sheafer (2006), Benoit,
Wen, & Yu (2006), Choi & Benoit (2006).
Table 6. Functions of U.S. Presidential Campaign Discourse
Policy Character
Primary TV Spots 3066 (54%) 2601 (46%)
(1952-2004)
Primary Debates 12734 (78%) 5875 (22%)
(1948-2004)
Primary Brochures 5660 (62%) 3424 (38%)
(1948-2004)
Acceptance Addresses 1558 (55%) 1287 (45%)
(1952-2004)
Keynote Speeches 509 (54%) 410 (46%)
(1960-1996)
General TV Spots 3581 (59%) 2212 (41%)
(1952-2004)
General Debates 4885 (75%) 1666 (25%)
(1960, 1976-2004)
General Brochures 8742 (76%) 2687 (24%)
(1952-2000)
General Radio Spots 474 (57%) 409 (43%)
(1972-1992, 2000)
Total 41209 (67%) 20571 (33%)
Spots Policy________________Character
US Senate 1980-2000 541 1930 (55%) 1570 (45%)
US House 1980-2000 438 1403 (51%) 1370 (49%)
US Congress 1998 29 120 (66%) 62 (34%)
Governor 1974-1998 488 1308 (66%) 666 (34%)
Local 1998-2000 70 229 (59%) 161 (41%)
Total 1566 4990 (57%) 3829 (43%)
X2(df=1) = 152.84, p < .0001
Benoit (2000), Brazeal & Benoit (2001), Pier (2002)
Table 8. Topics of U.S. Non-Presidential Political Debates
Policy_____________Character
U. S. Senate Debates 1813 (70%) 794 (30%)
1998-2004
U. S. Governor Debates 3166 (73%) 1150 (27%)
1994-2004
Total 4979 (72%) 1944 (28%)
x2(df=1) = 1328.64, p < .0001
Senate debates: Benoit, Brazeal, & Airne (2006)
Governor debates: Airne, Benoit, & Brazeal (2006).
Table 9. Topics of Non-U.S. TV Spots
Policy________________Character
France 1988 100% 0
Germany 1992* 31% 69%
Germany 1994 69% 31%
Italy 1992 71% 29%
Britain 1992, 1997 88% 12%
Israel 1992 50% 50%
Korea 1992*1 65% 35%
Poland 1995 66% 34%
Turkey 1995 33% 67%
Greece 1996 42% 58%
Russia 1996 58% 42%
Subtotal^ 59% 41%
Taiwan 2000 32% 68%
Korea 2002 50% 50%
*These samples included newspaper ads as well as TV spots. IThis study included a third category, combination, which is ex-
cluded here.
Kaid (1999), Kaid and Holtz-Bacha (1995), Lee and Benoit
(2004); Tak, Kaid, and Lee (1997), Wen, Benoit, and Yu (2004)
зо
^Percentages of spots reported by authors converted into number of spots: X (df=1)=47.42, p<.0001.
_____________Table 10. Topics of Non-U.S. Political Debates___________________
Policy Character
France, 1988, 1995 981 (89%) 121 (11%)
Israel, 1984, 1988, 1992, 222 (77%) 67 (23%)
1996, 1999
South Korea, 1997, 2002 1443 (84%) 269 (16%)
Taiwan, 2004 372 (60%) 251 (40%)
Ukraine, 2004 333 (61%) 213 (39%)
Total ^ -—- - 3351 (78%) 921 (22%)
X2(df=1) = 1381.1, p < .0001
Benoit & Klyukovski (2006), Benoit & Sheafer (2006), Benoit, Wen, & Yu (2006), Choi & Benoit (2006).
References
1. Airne D., & Benoit W.L. (2005a). 2004 Illinois U.S. Senate debates: Keyes versus Obama. American Behavioral Scientist, 49, 343-352.
2. Airne D., & Benoit W.L. (2005b). Political television advertising in campaign 2000. Communication Quarterly, 53, 473-492.
3. Airne D., Benoit W.L., & Brazeal L.M. (2006). A functional analysis of gubernatorial political debates. Unpublished manuscript.
4. Allen M., & Burrell N. (2002). The negativity effect in political advertising: A meta-analysis. In J.P. Dillard & M. Pfau (Eds.), The persuasion handbook: Developments in theory and practice (pp. 8396). Thousand Oaks, CA: Sage Publications.
5. Benoit P.J. (1997). Telling the success story: Acclaiming and disclaiming discourse. Albany: State University of New York Press.
6. Benoit W.L. (1995). Accounts, excuses, apologies: A theory of image restoration strategies. Albany: State University of New York Press.
7. Benoit W.L. (1999). Seeing spots: A functional analysis of presidential television advertisements from 1952-1996. New York: Praeger.
8. Benoit W.L. (2000). A functional analysis of political advertising across media, 1998. Communication Studies, 51, 274-295.
9. Benoit W.L. (2003). Presidential campaign discourse as a causal factor in election outcome. Western Journal of Communication, 67, 97-112.
10. Benoit W.L. (2004). Political party affiliation and presidential campaign discourse. Communication Quarterly, 52, 81-97.
11. Benoit W.L. (in press). Communication in political campaigns. New York: Peter Lang.
12. Benoit W.L., Brazeal L.M., & Airne D. (2006). A functional analysis of U.S. senate debates, 1998-2004. Unpublished manuscript.
13. Benoit W.L., Blaney J.R., & Pier P.M. (1998). Campaign ‘96: A functional analysis of acclaiming, attacking, and defending. New York: Praeger.
14. Benoit W.L., Hansen G.J., & Stein K.A. (2004). Newspaper coverage of presidential primary debates. Argumentation and Advocacy, 40, 246-258.
15. Benoit W.L., & Klyukovski A. (2006). A functional analysis of 2004 Ukrainian political debates. Unpublished manuscript.
16. Benoit W.L., McHale J.P, Hansen G.J., Pier P.M., & McGuire J.P. (2003). Campaign 2000: A functional analysis of presidential campaign discourse. Lanham, MD: Rowman & Littlefield.
17. Benoit W.L., Pier P.M., Brazeal L.M., McHale J.P., Klyuk-ovksi A., & Airne D. (2002). The primary decision: A functional analysis of debates in presidential primaries. Westport, CT: Praeger.
18. Benoit W.L., & Sheafer T. (2006). A functional analysis of Israeli Prime Ministerial debates. Unpublished manuscript.
19. Benoit W.L., Stein K.A., & Hansen G.J. (2004). Newspaper coverage of presidential debates. Argumentation and Advocacy, 41, 16-26.
20. Benoit W.L., Stein K.A., & Hansen G.J. (2005). New York Times coverage of presidential campaigns, 1952-2000. Journalism & Mass Communication Quarterly, 82, 356-376.
21. Benoit W.L., Wen W-C., & Yu H-T. (2006). A functional analysis of 2004 Taiwanese political debates. Unpublished manuscript.
22. Brazeal L.M., & Benoit W.L. (2001). A functional analysis of Congressional television spots, 1986-2000. Communication Quarterly, 49, 436-454.
23. Chang C. (2000). Political advertising in Taiwan and the U.S.: A cross-cultural comparison of the 1996 presidential election campaigns. Asian Journal of Communication, 10, 1-17.
24. Choi Y.S., & Benoit W.L. (2006). A functional analysis of French and South Korean political debates. Unpublished manuscript.
25. Duchneskie J., & Seplow, S. (2000, December 15). Gore’s vote lead totals 540,435. Philadelphia Inquirer, p. A1.
26. Farnsworth, S.J., & Lichter S.R. (2003). The nightly news
nightmare: Network television’s coverage of U.S. presidential elections, 1988-2000. Lanham, MD: Rowman & Littlefield.
27. Hallin D.C., & Mancini P. (2004). Comparing media systems: Three models of media and politics. New York: Cambridge University Press.
28. Hofstetter C.R. (1976). Bias in the news: Network television coverage of the 1972 election campaign. Columbus, OH: Ohio State University Press.
29. Johnson-Cartee K.S., & Copeland G. (1989). Southern voters’ reactions to negative political ads in the 1986 election. Journalism Quarterly, 66, 888-893, 986.
30. Kaid L.L. (1999). Comparing and contrasting the styles and effects of political advertising in European democracies. In L.L. Kaid (Ed.), Television and politics in evolving European democracies (pp. 219-236). Commack, NY: Nova Science.
31. Kaid L.L., & Holtz-Bacha, C. (1995). Political advertising across cultures. In L.L. Kaid & C. Holtz-Bacha (Eds.), Political advertising in Western democracies (pp. 206-227). Thousand Oaks, CA: Sage.
32. Lau R.R., Sigelman L., Heldman C., & Babbitt P.R. (1999). The effectiveness of negative political advertising: A meta-analytic assessment. American Political Science Review, 93, 851-876.
33. Lazarsfeld P.F., Berelson B., & Gaudet H. (1948). The people’s choice: How the voter makes up his mind in a presidential campaign (2nd ed.). New York: Columbia University Press.
34. Lee C., & Benoit W. L. (2004). A functional analysis of presidential television spots: A comparison of Korean and American ads. Communication Quarterly, 52, 68-79.
35. Lee C., & Benoit W.L. (2005). A functional analysis of the 2002 Korean presidential debates. Asian Journal of Communication, 15, 115-132.
36. Lenart S. (1994). Shaping political attitudes: The impact of interpersonal communication and mass media. Thousand Oaks, CA: Sage.
37. Merritt S. (1984). Negative political advertising: Some empirical findings. Journal of Advertising, 13, 27-38.
38. Patterson T.E. (1980). The mass media election: How Americans choose their president. New York: Praeger.
39. Patterson T. E., & McClure R. D. (1976). The unseeing eye: The myth of television power in national politics. New York: Putnam.
40. Petrocik J.R., Benoit W.L., & Hansen G.L. (2003-2004). Is-
sue ownership and presidential campaigning, 1952-2000. Political Science Quarterly, 118, 599-626.
41. Pier P.M. (2002). He said she said: A functional analysis of differences between male and female political campaign messages. Unpublished dissertation, University of Missouri.
42. Popkin S.L. (1994). The reasoning voter: Communication and persuasion in presidential campaigns (2nd ed.). Chicago: University of Chicago Press.
43. Popkin S.L., Gorman J., Smith J., & Phillips C. (1976). Comment: toward an investment theory of voting behavior: What have you done for me lately? American Political Science Review, 70, 779805.
44. Robinson M.J., & Sheehan M.A. (1983). Over the wire and on tv: CBS and UPI in campaign ‘80. New York: Russell Sage.
45. Stewart C.J. (1975). Voter perception of mud-slinging in political communication. Central States Speech Journal, 26, 279-286.
46. Tak J., Kaid L.L., & Lee S. (1997). A cross-cultural study of political advertising in the United States and Korea. Communication Research, 24, 423-430.
47. Wen W.-C., Benoit W.L., & Yu T.-H. (2004). A functional analysis of the 2000 Taiwanese and U. S. presidential spots. Asian Journal of Communication, 14, 140-155.
© Бенуа Л. Уильям,2006
P.S. Перевод статьи размещен на сайте www.cognitive.narod.ru
