900

Research Report

Campaign Advertising and Democracy in Taiwan Jonathan Sullivan

ABSTRACT This article sets out a framework for conceptualizing and researching an increasingly important element of electoral competition in Taiwan. Analysing the tone and content of television and newspaper advertising across three presidential campaigns, it provides empirical estimates that challenge some common stereotypes about advertising in Taiwan.

The tone and content of campaign advertising in Taiwan is clearly of concern to Taiwan scholars. Candidates reportedly ‘‘subject each other to the most vicious form of negative campaigning imaginable,’’1 whilst their advertisements are frequently criticized for ‘‘lacking beef.’’2 The problem of negative advertising is apparently so severe that ‘‘after the 2000 presidential election, there was agreement among media and law experts that negative advertising had reached new and worrying dimensions; and yet the 2004 race broke all records.’’3 Such sentiments are understandable, but they are seldom based on the systematic analysis of a large number of advertisements. Furthermore, though many commentaries imply some kind of harmful connection between advertising and democracy, there is no explication of what it is that advertising should do to in order to be beneficial to democratic competition. This is largely a normative question, and though normative choices are inevitably partial, I argue in this article that provision of information is one way of conceiving and evaluating the role of campaign advertising in Taiwan. The purpose of the article is twofold. First, it aims to provide empirical estimates of the level of negativity and extent of image advertising based on the analysis of a comparatively large number of advertisements over time. Second, it provides a comparative assessment of the extent to which positive and negative appeals provide ‘‘substantive’’ information to voters. Specifically, this involves comparing the degree of issue focus, the way the issues are dealt with and the provision of supporting evidence. Breaking with prior studies that focus on a 1 Gary Rawnsley, ‘‘Treading a fine line: democratization and the media in Taiwan,’’ Parliamentary Affairs, Vol. 57, No. 1 (2004), p. 222. 2 Gary Rawnsley, ‘‘Where’s the beef? The 1998 mayoral election in Taiwan,’’ American Journal of Chinese Studies, Vol. 7, No. 2 (2000), pp. 147–69. 3 Christian Schafferer, Election Campaigning in East and South-East Asia (Aldershot: Ashgate, 2006), p. 49. # The China Quarterly, 2008

doi:10.1017/S0305741008001185

Campaign Advertising and Democracy in Taiwan

single campaign or medium, the article analyses the content of television and newspaper advertisements for six candidates across three presidential elections.

Research on Campaign Advertising in Taiwan Research on campaign advertising, particularly from the perspective of its role in Taiwan’s democracy, is limited, despite recognition that advertising has become increasingly prevalent as a campaign tool and widespread concerns about the preponderance of personality over substance and rampant negativity. Campaigns in Taiwan have received coverage as part of more general democratization and election studies,4 which detail the structural developments in Taiwanese politics that have had a fundamental effect on the political and media contexts in which campaigns take place.5 As a result of these well documented changes,6 campaigning in Taiwan has evolved from a clientelistic mobilization battle supplemented by small-scale traditional practices to an enterprise that, at the presidential level, shares many elements with the ‘‘modern’’ or even ‘‘post-modern’’ campaigns witnessed in established democracies like the United States.7 These features include a pre-eminent television campaign, an expanded role for professional consultants, enormous campaign expenditure, a prolonged if not ‘‘permanent’’ campaign, and of course, abundant campaign advertising.8 Structural developments in the political and media environments and actual campaigning are generally thought to be connected through the adaptive behaviour of parties and candidates.9 Accordingly, the majority of scholars studying advertising in Taiwan have turned to political marketing theories for explanations of candidate behaviour as it has evolved in response to the changing campaign environment. The approach draws on theories of consumer marketing and business management to conceptualize the ‘‘political market,’’ and analyses candidate responses to prevailing ‘‘market conditions.’’ A key concern for researchers is to ascertain whether or not a chosen advertisement or advertising strategy has been ‘‘effective.’’ Zheng Zilong (鄭自隆), for instance, 4 One important exception is Dafydd Fell’s Party Politics in Taiwan (London: Routledge, 2005). His content analysis of candidate advertisements across a range of elections from 1991 to 2001 is primarily used as a way of measuring issue salience. In keeping with the book’s refreshingly sanguine tenor is the finding that candidates do emphasize the issues in their advertising. 5 For example Sylvia Chan-Olmstead and Peilin Chiu, ‘‘The impact of cable television on political campaigns in Taiwan,’’ International Communication Gazette, Vol. 61, No. 6 (1999), pp. 491–509; Gary Rawnsley, ‘‘An institutional approach to election campaigning in Taiwan,’’ Journal of Contemporary China, Vol. 12, No. 37 (2003), pp. 765–79. 6 For example the opening up of electoral offices, the legalization of cable television etc. 7 Terminology from Pippa Norris, A Virtuous Circle; Political Communications in Post-Industrial Societies (Cambridge: Cambridge University Press, 2000). For accounts of recent presidential elections in Taiwan see Shelley Rigger, From Opposition to Power: Taiwan’s Democratic Progressive Party (Boulder CO: Lynne Rienner, 2001); Christian Schafferer, ‘‘The 2004 presidential election in Taiwan,’’ East Asia, Vol. 3, No. 1 (2004), pp. 1–9. 8 Norris, A Virtuous Circle. 9 Kathleen Hall Jamieson, Packaging the Presidency (New York: Oxford University Press, 1996).

901

902

The China Quarterly, 196, December 2008, pp. 900–911

demonstrates how Chen Shui-bian’s (陳水扁) advertising campaign in 2000 was highly effective in creating an image of Chen as a reformer embodying ‘‘change,’’ and using homely, peaceful images to deflect attacks on his earlier proindependence stance and associated threats to Taiwan’s security.10 Whilst the majority of political marketing analyses on Taiwan are perceptive and convincing, they view campaign advertising too narrowly through the lens of candidate strategies for winning elections. However, if ‘‘campaigns are a main point – perhaps the main point – of contact between officials and the populace over matters of public policy,’’11 then scholars’ interest in campaign advertising should extend beyond micro-level electioneering and focus on its place in democratic competition.

An Alternative Framework for Analysing Campaign Advertising In contrast to the political marketing approach, research in political communication is primarily motivated by a dual concern for the role of campaign advertising and the broader implications for citizen participation in the democratic process. A key assumption is that the two-way flow of information between governors and citizens is a central element in representative democracy. Accordingly, much work on campaign advertising is rooted in the conception of the campaign as an ‘‘institutionalized forum that bridges civil society and the government.’’12 Through the mechanism of the campaign, citizens should be enabled and encouraged to communicate their preferences (for example, in favour of a particular policy, or to reward or sanction incumbents), which they can only achieve if ‘‘plentiful reliable political information’’ is available.13 Although the propensity of voters to pay attention and their ability to process campaign information has long been debated,14 researchers have nevertheless found that they are at least minimally able to assimilate and make choices based on the information they receive.15 Moreover, the way in which they do so suggests that, though often maligned, generally short and unambiguous candidate advertising is ideally suited to the task of transmitting information.16 10 Zheng Zilong, ‘‘2000 nian zongtong daxuan jingxuan wenxun guancha’’ (‘‘Investigation of election propaganda in the 2000 presidential election’’), paper presented at eighth ROC advertising and public policy symposium (National Zhengzhi University, April 2000). Professor Zheng has published analyses of campaign advertising for most elections since the early 1990s. 11 William Riker, The Strategy of Rhetoric (New Haven CT: Yale University Press, 1996), p. 2. 12 Adam Simon, The Winning Message: Candidate Behaviour, Campaign Discourse and Democracy (Cambridge: Cambridge University Press, 2002), p. 11. 13 Richard Gunther and Anthony Mugham, Democracy and the Media: A Comparative Perspective (Cambridge: Cambridge University Press, 2000), p. 425. 14 Angus Campbell, Philip Converse, Warren Miller and Donald Stokes, The American Voter (Chicago: Chicago University Press, 1960). 15 V.O. Key, The Responsible Electorate: Rationality in Presidential Voting 1936–1960 (Cambridge, MA: Belknap Press, 1966); John Zaller, The Nature and Origins of Mass Opinion (New York: Cambridge University Press, 1992). 16 Paul Freedman, Michael Franz and Kenneth Goldstein, ‘‘Campaign advertising and democratic citizenship,’’ American Journal of Political Science, Vol. 48, No. 4 (2004), pp. 723–41; Michael Alvarez, Information and Elections (Ann Arbor: University of Michigan Press, 1997).

Campaign Advertising and Democracy in Taiwan

The conceptual role of specifically negative advertising has been complicated by potential ‘‘unintended consequences,’’ that is, harmful exposure effects at the voter level. Exploring these effects, predominantly in the United States, has generated a massive empirical literature. The ‘‘demobilization debate’’ over negative advertising and voter turnout has dominated the research agenda for more than a decade.17 Whilst several early studies appeared to establish a connection between exposure to negative advertising and a reduction in levels of political participation,18 an overwhelming majority of subsequent findings suggest that negative advertisements have a ‘‘significant and substantial mobilizing effect.’’19 One explanation for this result combines the well-established connection between levels of political knowledge and participation, with the observation that negative advertisements are actually highly informative.20 Researchers have argued that some forms of necessary information are only available through negative appeals.21 For example, attacks are often the only way in which voters can learn of ‘‘flaws, faults and flights of fancy,’’ since candidates are understandably reluctant to provide ‘‘candid appraisals of their own histories and records.’’22 The expectation is that in their positive advertisements, candidates should outline their policy plans and qualifications for the job, whilst negative advertisements should question and challenge an opponent’s achievements, credentials and pledges. Since a voter’s decisionmaking calculus ideally requires this mix of informational inputs, ‘‘it is hard to imagine an election that would be informative without some degree of both positive and negative campaigning.’’23 Identifying this informational ‘‘division of labour’’ is a necessary but insufficient condition for justifying the presence of negative advertising, if, as common stereotypes hold, it is merely uninformative mudslinging. Clearly, additional standards are required to measure the level of information contained within an advertisement. Recent empirical work in the United States does just this, comparing the content of positive and negative appeals on several indicators of ‘‘informativeness.’’24 The first indicator is the degree to which 17 For an up-to-date review of this literature see Deborah Brooks, ‘‘The resilient voter: moving toward closure in the debate over negative campaigning and turnout,’’ Journal of Politics, Vol. 68, No. 3 (2006), pp. 684–96. 18 Stephen Ansolabehere and Shanto Iyengar, Going Negative (London: Simon Schuster, 1995). 19 Kenneth Goldstein and Paul Freedman, ‘‘Campaign advertising and voter turnout: new evidence for a stimulation effect,’’ Journal of Politics, Vol. 64, No. 3 (2002), p. 733. 20 Freedman et al., ‘‘Campaign advertising and democratic citizenship,’’ pp. 723–41. 21 Mathias Polborn and David Yi, ‘‘Informative positive and negative campaigning,’’ Quarterly Journal of Political Science, No. 1 (2006), pp. 351–71; William Mayer, ‘‘In defense of negative campaigning,’’ Political Science Quarterly, Vol. 111, No. 3 (1996), pp. 437–55. 22 Kathleen Hall Jamieson, Paul Weldman and Susan Sherr, ‘‘Eliminate the negative? Categories of analysis for political advertisements,’’ in James Thurber, Candice Nelson and David Dulio (eds.), Crowded Airwaves: Campaign Advertising in Elections (Washington DC; Brookings Institution, 2000), p. 45. 23 Deborah Brooks and John Geer, ‘‘Beyond negativity: the effects of incivility on the electorate,’’American Journal of Political Science, Vol. 51, No. 1 (2007), p. 1. 24 John Geer, In Defense of Negativity: Attack Ads in Presidential Campaigns (Chicago: University of Chicago Press, 2006).

903

904

The China Quarterly, 196, December 2008, pp. 900–911

appeals focus on substantive policy issues, since it is generally accepted in political science that ‘‘voting choices based on policy concerns are superior to decisions based on party loyalty or candidate image.’’25 The second is the degree of specificity and the way in which advertisements address the issues, based on the assumption that vague appeals to the general good do not help voters to identify a candidate’s policy position or record. The third indicator is the extent to which appeals are provided with supporting evidence, such as sourced statistics. Evidence does not make an appeal ‘‘true,’’ but it does render claims falsifiable. Theoretically this increases the utility of information by reducing the risk of deception, which as prior research makes clear, is just as likely, if not more so, in positive advertisements where candidates promote themselves.26 Comparing appeals in this way allows us to comment on the comparative merits of positive and negative appeals without having to make subjective judgements about absolute levels of informativeness. Using this approach and these indicators, Geer demonstrates that in American presidential elections, negative appeals are comparatively informative.27 The preceding discussion provides an alternative framework for conceptualizing and analysing campaign advertising in Taiwan. According to the position outlined above, campaign advertising can, and should, provide voters with relevant information that promotes and questions candidates’ policy positions, records of achievement and qualifications for office. This necessarily requires a mix of positive and negative appeals. In addition, the more issue-focused, specific and frequently supported with evidence that information is, the better it serves to inform voters. The following analysis evaluates the content of Taiwanese presidential candidates’ advertising by these measures.

Methods and Data As in most prior research, the distinction between positive and negative employed here is directional, and is based on the assumption that when candidates talk about themselves they do so in a self-promotional way. Conversely when candidates talk about their opponents it is assumed that they do so in order to criticize. The data strongly support the validity of this assumption.28 The units of analysis in this study are the textual or verbal appeals contained within an advertisement.29 An appeal is operationalized as a word or phrase representing a reason to vote for the sponsor (a positive appeal) or to vote against an opponent (a negative appeal). In order to facilitate comparison 25 Edward Carmines and James Stimson, Issue Evolution (Princeton NJ: Princeton University Press, 1989), p. 79. 26 Jamieson et al., ‘‘Eliminate the negative?’’ 27 Geer, In Defense of Negativity. 28 Candidates say something critical about themselves or supportive about their opponent in 0.5% of 6,868 appeals. 29 For a discussion of the appeal as the appropriate unit of analysis, see Jamieson et al., ‘‘Eliminate the negative?’’ p. 51.

Campaign Advertising and Democracy in Taiwan

on a single standard of informativeness, I assume that all appeals are equally weighted. This simplifying assumption, though frequently exploited in the literature, unavoidably introduces an element of omitted variable bias (such as ignoring the size or positioning of characters in print advertisements).30 Official televised and print advertisements were collected for each of the KMT and DPP candidacies across three presidential elections: a total of 342 advertisements. This includes all the unique television spots that aired during the official campaign period and, following Fell,31 the population of newspaper advertisements published in the main editions of the Liberty Times (Ziyou shibao 自由時報), China Times (Zhongguo shibao 中國時報) and United Daily News (Lianhebao 聯合報). The data collection is shown in Table 1. I carried out a manual content analysis using a code scheme constructed with the help of existing dictionaries and inductively through the analysis of a subsample of television and newspaper advertisements. In addition to the typical distinction between issues and traits, the code scheme records information on two dimensions that are salient features of Taiwanese campaign advertising. The first category, labelled values, measures narrow ideological appeals, such as those pertaining to ‘‘Taiwanese identity’’ (‘‘love Taiwan’’ and the like), and more universal values such as progress, prosperity and the future. Value appeals contain no references to specific policy sectors. The second category measures strategic appeals related to the business of the election and the campaign itself, such as appealing for votes, advocacy of strategic voting and so on. In addition to recording the policy area and tone, issue appeals are further separated into general statements, specific proposals and party or candidate performance. Thus an appeal on the issue of employment could be a general comment (‘‘unemployment is a problem in this society’’), a specific proposal (‘‘if elected I will reduce unemployment by investing in public works’’), or a comment on one’s own or one’s opponent’s record on the issue (‘‘since I became president, unemployment has fallen one per cent’’). The traits category records mentions of the personal characteristics of the candidates, such as leadership,

Table 1: Data Collection Year

Candidates (party)

1996

Li-Lian (KMT) Peng-Xie (DPP) Lian-Xiao (KMT) Chen-Lu¨ (DPP) Lian-Song (KMT/PFP) Chen-Lu¨ (DPP)

2000 2004

Newspaper

Television spots

27 35 43 20 65 30

17 8 35 15 31 16

30 Additionally, the ‘‘possibility that each mass medium invokes differential cognitive involvement’’ must also be accounted for in comparative analyses of the effectiveness of information transmission: Craig Brians and Martin Wattenburg, ‘‘Campaign issue knowledge and salience: comparing reception from TV commercials, TV news and newspapers,’’ American Journal of Political Science, Vol. 40, No. 1 (1996), p. 175. 31 Fell, Party Politics in Taiwan.

905

906

The China Quarterly, 196, December 2008, pp. 900–911 competence, integrity, compassion and their antonyms.32 An important distinction is made between personality and policy performance. An appeal such as ‘‘my opponent is corrupt’’ would be coded as a negative trait, whereas ‘‘my opponent has let corruption flourish during his time as mayor’’ refers to candidate performance on the issue of corruption. Citations of evidence, operationalized as documented statistics, direct quotations and sourced reports, were recorded in the relevant appeal categories.

Results Recent accounts of campaign advertising in Taiwan refer to high levels of negativity, and the data suggest some justification of this view. The mean level of negativity across all three elections is 32 per cent for newspaper advertisements and 26 per cent for television spots. There has also been a marked increase in negativity over the most recent two campaigns, where the average level of negativity is 42 per cent for newspaper advertisements and 37 per cent for television spots. To give these figures perspective, a similarly designed study of television spots in American presidential elections from 1960 to 2000 reports an average of 32 per cent negative appeals.33 Given that campaign advertising in the United States is often conceived as being archetypically negative, it is clear that the situation in Taiwan is worthy of investigation. A second concern is that campaign advertising in Taiwan is excessively centred on personality. Given the discussion in this article, this implied lack of substance may be a more serious concern than the level of negativity. Figure 1 shows the aggregate (not separating appeals by tone) distribution of appeals in the four categories, for both media types across all three elections. The data do not support the view that Taiwanese presidential candidates are systematically sacrificing substance for personality. Trait appeals constitute 15 per cent of newspaper appeals and 23 per cent of television spot appeals. Furthermore, the majority of these trait appeals (75 per cent for newspaper advertisements, 64 per cent for television spots) target characteristics that are legitimately related to the business of governing: leadership, competence and integrity. More striking, and in line with reports on Taiwanese elections,34 is the heavily ideological content of candidate advertising, particularly television spots (41 per cent of all appeals). Though newspaper advertisements focus more on the issues (39 per cent of appeals), ideological content still accounts for nearly onethird of newspaper appeals (32 per cent). Overall DPP candidates tend to be 32 For a discussion of ‘‘relevant’’ traits see Carolyn Funk, ‘‘Bringing the candidates into models of candidate evaluation,’’ Journal of Politics, No. 61 (1999), pp. 700–20. Extraneous traits (such as religiosity, sense of humour etc.) were also recorded, facilitating an estimate of the extent to which trait appeals were relevant to the business of governing. 33 Geer, In Defense of Negativity, p. 35. 34 For example Cal Clark, ‘‘The paradox of the national identity issue in Chen Shui-Bian’s 2004 presidential campaign: base constituencies vs. the moderate middle,’’ Issues and Studies, Vol. 41, No. 1 (2004), pp. 53–86.

Campaign Advertising and Democracy in Taiwan

Figure 1: Aggregate Distribution of Appeals by Category

more ideological (43 per cent newspaper appeals, 48 per cent television appeals) than KMT candidates (30 per cent newspaper, 41 per cent television). Substantively, these appeals reflect well-known partisan themes. For instance, on average over three campaigns, ‘‘Taiwanese identity’’ and the value (rather than any policy dimension) ‘‘democracy and reform’’ constitute 73 per cent of DPP value appeals.35 In line with Fell’s earlier findings,36 social issues, governance and the economy, rather than the set of issues pertaining to crossStrait relations, are the most salient policy sectors over time. Figure 2 shows the variation in emphasis on each category across the three elections. Not surprisingly, campaigns are not uniformly ideological or personal, but vary over time and across media types. For instance, whilst the newspaper (42 per cent) and television (54 per cent) campaigns in 1996 were both comparatively ideological, candidates in 2000 de-emphasized ideology. Instead, the campaign in 2000 was relatively personal, with 23 per cent of newspaper appeals and 36 per cent of television appeals targeting personal characteristics.37 Whilst newspaper advertisements in 2004 were heavily issue-focused (48 per cent), television spots in the same campaign were predominantly ideological (42 per cent), particularly those of Chen-Lu¨ (48 per cent).

35 Though ideological positions have helped parties develop distinct identities and allow voters to discriminate candidates, excessive emphasis of emotion-laden ideological themes, particularly national identity, in recent elections has justifiably provoked worries about the dangers of ideological polarization for Taiwan’s democracy. For example, Ste´phane Corcuff, ‘‘The supporters of unification and the Taiwanisation movement,’’ China Perspectives, No. 53 (2005), pp. 49–66; John F. Copper, ‘‘Taiwan; democracy’s gone awry?’’ Journal of Contemporary China, Vol. 12, No. 34 (2003), pp. 145–62. 36 Fell, Party Politics in Taiwan. 37 This reflects well-known contextual features of the 2000 campaign, e.g. the participation of the competitive but scandal-tainted independent candidate Song Chu-yu.

907

908

The China Quarterly, 196, December 2008, pp. 900–911

Figure 2: Appeals by Category in Each Election

Figure 3: Appeals by Category and Tone

I argue above that the presence of negative appeals may be justified by the level of information they provide vis-a`-vis positive appeals. The first question to address is the degree to which positive and negative appeals focus on the issues. Figure 3 shows the distribution of appeals in the four categories divided by tone. As expected, negative appeals for both media types are comparatively issue focused, with 48 per cent of negative newspaper appeals and 40 per cent of negative television appeals focusing on policy. By contrast 34 per cent of positive newspaper appeals and just 24 per cent of positive TV appeals deal with the issues. Positive appeals are much more ideological (39 per cent newspaper, 53

Campaign Advertising and Democracy in Taiwan

per cent television) than negative appeals (19 per cent newspaper, 12 per cent television). Negative appeals are comparatively personal, particularly television spots (44 per cent trait appeals), but the characteristics emphasized in negative appeals, especially integrity and competence, are more consistently relevant to the business of governing than positive trait appeals. The second question concerns the way in which issue appeals work. Figure 4 presents the breakdown of issue appeals in terms of general statements, policy proposals and policy performance. Taking newspaper advertisements first, as expected, negative appeals predominantly focus on performance (68 per cent). Candidates also dedicate the majority (51 per cent) of positive appeals to lauding their own policy achievements; thus overall the majority of information related to the issues is retrospective. Some 30 per cent of positive appeals are specific policy proposals, although only 8 per cent of negative appeals engage opponents’ plans. Somewhat surprisingly, negative newspaper appeals are comparatively vague, with a slightly higher proportion of general statements (24 per cent) than positive appeals (19 per cent). The differences between positive and negative issue appeals in television advertising are more pronounced and more in line with expectations. For instance, negative appeals are dominated by claims that target opponents’ performance (85 per cent). Positive television appeals contain more vague general statements (33 per cent) than negative appeals (6 per cent). Some 21 per cent of positive television appeals are used to make policy proposals, but again just 9 per cent of television appeals criticize opponents’ proposals. Clearly, and regardless of media type, campaign advertising is not the candidates’ chosen venue for engaging their opponents’ policy plans.38

Figure 4: Composition of Positive and Negative Issue Appeals

38 See Simon, The Winning Message on the dangers of ‘‘dialogue.’’

909

910

The China Quarterly, 196, December 2008, pp. 900–911

Figure 5: Provision of Evidence by Category and Tone

The final question is the extent to which candidates support their claims with evidence. Providing supporting evidence for negative claims has become a more important consideration for Taiwanese candidates. For instance, in 1996 less than 1 per cent of negative appeals in television spots were supported with evidence, compared with 12 per cent in 2000 and 10 per cent in 2004. Overall across all three elections, a mere 1 per cent of positive newspaper appeals and 0.5 per cent of positive television appeals were supported with evidence. For negative appeals this rises to 10 per cent for newspapers and 11 per cent for television. Whether these are acceptable figures or not is a matter of interpretation, but what we can say is that negative appeals are much more likely than positive appeals to supply evidence. The picture is even clearer when we look at variation across appeal categories. Figure 5 shows the provision of evidence for positive and negative appeals by category. In each category, regardless of media type, negative appeals are more likely to supply evidence in support of a claim. For instance, 17 per cent of negative issue appeals in newspaper advertisements are supported compared with just 3 per cent of positive issue appeals. Positive television appeals are almost entirely unsupported, with issues (2 per cent) the only appeals backed with evidence. By contrast 12 per cent of negative issues and negative trait appeals in television spots are supported with evidence. Candidates are also much more careful to support their negative strategic appeals, particularly in television spots.

Conclusion This article began by referencing concerns that advertising in Taiwan is highly personal and negative. These concerns deserve to be addressed given the

Campaign Advertising and Democracy in Taiwan

importance of campaign activities to democratic competition. The empirical findings do not support the view that substance is being systematically sacrificed for personality; instead the most striking feature of campaign advertising in Taiwan is the amount of ideological content. In line with prior research, the data do indicate high levels of negativity, with a mean level similar to that reported in the United States. However, in presenting a new framework for conceiving the role of campaign advertising that privileges the supply of information, the article challenges common stereotypes about negative advertising. Taiwanese presidential candidates’ negative appeals are comparatively issue-focused, more frequently supported with evidence and, though seldom questioning opponents’ policy plans, they frequently challenge opponents’ policy achievements. Although every campaign throws up egregious examples of scurrilous and irrelevant attacks, the broader picture suggests that negative advertising makes a potentially useful contribution to the information environment. However, whether voters are in fact benefiting from this information is an important research question that needs to be addressed. If negative campaigning is comparatively informative then, ceteris paribus, voters who are exposed to it should be comparatively knowledgeable. By extension we can hypothesize that they should also be comparatively engaged in the political process. On the other hand, there may be voter level effects that are orthogonal to information, so that affective responses, particularly to ideological appeals, may dominate cognitive ones. To date, little attention has been paid to these potential effects. With campaign advertising increasingly prevalent at all levels of office, it is time that the analysis of campaign activities at the party level, and their potential effects at the voter level, are brought in to the broader conversation on Taiwan’s democratic consolidation.

911