INTERNET USE AND PROTEST POLITICS IN SOUTH KOREA AND TAIWAN

Abstract

This study examines the association between Internet use and protest politics in South Korea and Taiwan, using secondary data from the sixth wave (2010–14) of the World Values Survey. The data show that Internet use is positively associated with low-cost protest acts, like signing a petition. Internet use is also correlated with the willingness to consider higher-cost actions such as attending demonstrations and joining in boycotts. But it does not appear to consistently increase actual protests of this sort. Discussion is made to address comparable opportunity structures for protest politics in the two East Asian democracies, where the Internet is deeply integrated into the country's political landscape.

INTRODUCTION

In 2008, a new president of South Korea (hereafter Korea) agreed to relax beef imports from the US. But the decision ushered in a series of mass demonstrations against the trade policy and president Lee Myung-bak. The protesters framed the issue in a public-health discourse that appealed to mass interest through online discussion and community-building in cyberspace. In the eyes of the government, however, the motivation for the collective action was exaggerated anxiety about mad cow disease in imported meats. The gap in the framing of the risk was too wide to be bridged, and massive crowds protested for a prolonged period of time. As a result, the Korean government had to address growing public concern over US beef imports by delaying importation of meat, especially from cattle over 30 months of age.

The protest against US beef was unlike traditional protests, which had centered on civil-society groups and nongovernmental organizations in mobilization and coordination of large-scale collective action. Instead of such social-movement organizations working in perfect order, the Korean protesters came from heterogeneous political backgrounds and multiple social bases. In particular, the initiators of the mobilization were made up largely of middle- and high-school students as well as young mothers with strollers who had not been involved in previous protests but were informed and mobilized through the Internet. Although the most visible aspect was that a huge crowd gathered in the streets and squares of Seoul, people participated in multiple methods of protest, expressing their personal opinions in creative ways all of which were represented by the meme of the candlelight vigil. This vigil re-occurred in late 2016 when the largest rally in the history of Korean democracy was mobilized to demand the resignation of the then-President Park Geun-hye through voluntary chants and unstructured presentations among the protesters rather than a unified slogan and structured activity.

This new form of social movement was not unique to Korea. Notwithstanding a difference in the type, scale, and target of protest, the political landscape in Taiwan has also seen the rise of unprecedented forms of social movements in which youth groups mobilized large-scale collective action. On March 18, 2014, more than half a million Taiwan citizens held a record-breaking demonstration in front of the Presidential Hall to protest the Ma Ying Jeou administration's pro-China stance and policies. This so-called Sunflower Movement was initiated by hundreds of Taiwanese students who resented the Cross-Strait Service Trade Agreement with mainland China and occupied the parliament. When the police evicted protesters by force, and it was live-streamed to the public through the Internet, the occupation led to the biggest political movement in the past 30 years of Taiwan history. The student protesters were inexperienced activists who took advantage of social media, rather than mainstream media or sociopolitical organizations, for engaging disgruntled citizens, reinforcing concern about the One-China policy, and consolidating agitation against the president.

These cases of contemporary social movements in emerging democracies of East Asia are not irrelevant to the rise of a networked public relying on unprecedented modes of participation in politics without grand orchestration (Kim 2014). In the view of Western scholars, digitally enabled networks are likely to be a mobilization structure for the new social movements (Bennett and Segerberg 2013; Howard 2015). That is, the Internet equips activists and dissidents with personalized ways of grassroots organizing so that they instigate, coordinate, and expand the protest among ordinary citizens who have diverse interests as well as different degrees of commitment to a political cause. In this way, online communities emerge as the mobilizing agents for circulating contentious information, recruiting participants, forming shared identities, and organizing protest activities (Ayres 1999; Garrett 2006).

This new mechanism of digitally networked activism has important social implications in the context of political culture in East Asia. Although both countries have made the transition from an authoritarian regime to a multi-party democracy on top of rapid economic success in the last three decades, they are considered to be lagging behind Western democracies when it comes to the consolidation of civil-society organizations as a channel of unconventional political participation (Kim 2005; Marsh 2003). Nevertheless, Korea and Taiwan provide interesting cases for comparison in relation to widespread diffusion of Internet use that lowers citizens’ barriers to protest communications. In light of the Internet's mobilizing impact, we therefore examine 1) whether technology facilitates individual engagement in protest activity and 2) why the two tigers in Asia are experiencing similar approaches to the mobilization of unconventional political participation.

Our framework of analysis draws on resource-mobilization theory, in which the Internet is conceptualized as a communication network that affords the mobilization of resources for collective action. Based on this conceptualization, data from the sixth wave (2010–14) of the World Values Survey (WVS) is used to examine whether frequent use of the Internet for political information increases the likelihood of engaging in a protest. But, because protest activities have different thresholds for participation (Barnes and Kaase 1979), Internet use is also assumed to exert its mobilization effects at different levels between the committed and disengaged. Also, we examine our theoretical expectations by comparing Korea and Taiwan. Although the two countries have their own contextual constraints such as political institutions, the findings about the mobilization of protest politics are common to both Asian tigers. This comparative study adds to an understanding of the mechanism by which Internet use mobilizes unconventional forms of political participation in the democratic context of East Asia.

LITERATURE REVIEW

THE ROLE OF INTERNET USE IN PROTEST PARTICIPATION

To account for participation in protests, resource-mobilization theorists focus on structural conditions that provide external opportunities for individuals to be asked and organized to act on a common cause (Leighley 1995). Their view relates participation to the role that the mobilization agents play in the collection and strategic deployment of resources to resolve Olson's (1965) free-riding dilemma: since such collective efforts are mobilized to deliver a public good that is not selective in its utilities, rational individuals decide not to bear the costs of participation “on their own” as the way of maximizing the utility. For example, Bimber, Stohl, and Flanagin highlight civil-society groups as the agents of resource mobilization that “provide the mechanisms through which political issues are articulated; participants are recruited; targets, locations, and timing of collective actions are determined; complex tasks and strategies are coordinated; and methods and tactics are selected” (2009, 72–73).

In today's digital era, the Internet is increasingly viewed as a new agent of resource mobilization for protest participation. Bennett and Segerberg (2013) argue that the technology allows activists and their supporters to instigate shared grievances, form collective identities, and coordinate protest activities among individual citizens with different degrees of concern, knowledge, skills, and commitment. This suggests that the Internet provides a mobilization structure for resources in which the disaffected are organized and coordinated for protest activities even in the absence of conventional social movement organizations. Indeed, contemporary political activism often emerges from “loosely tied, opt-in/opt-out networks,” where fluid identities are connected for a common cause through personalized action frames across diverse sociopolitical groups (Bennett and Sergerberg 2013). Social media and customized optimization algorithms promote this trend in which interpersonal interaction enables activism to engage the disaffected and the like-minded (Howard 2015).

The mobilizing potential of technology is related to Granovetter's (1973) strength of weak ties, which is bridging not only information flows but also community organizations (Gil de Zúñiga and Valenzuela 2011). For example, Facebook use has been found to encourage heterogeneous interpersonal relations that help to access civically relevant social capital (Ellison, Steinfield, and Lampe 2011). Moreover, digital networks can facilitate incentives for joining collective action – such as recognition and emotional support from like-minded people or communities – which are easily formed and bridged in the virtual sphere (Polat 2005). And, in this mobilization process, digital networks afford a variety of repertories for engagement in protest that can be taken beyond previous spatial and temporal confines at little expense (Castells 2012; Earl and Kimport 2011). The Internet's political role thus takes the form of a mechanism by which activists and their groups connect with distant groups with similar grievances.

Certainly, information technology has been frequently studied in relation to its capacity for political communication. Digital media facilitate the process of citizen mobilization insofar as they are used for information-processing behavior based on opinion expression and activism (Valenzuela 2013). Ayres (1999, 137) highlights the changing dynamics of protest participation in cyberspace with a variety of options for the rapid diffusion of contention, “including posting messages on a discussion board, joining a listserv to receive up-to-date information on a campaign or new event through an email account, sending e-mail to politicians, government agencies, or other activists, and searching on related links for additional information related to various campaigns.” A recent study in Italy also found that, as people gain political information and, more importantly, express their political opinions on Facebook or Twitter, they participate in politics through “more demanding activities such as e-mailing politicians, campaigning for them on social media, and attending offline political events” (Vaccari et al. 2015, 222). This view treats technology as a communication network for activists and dissidents to easily and quickly connect with ordinary citizens who gain online information for a contentious campaign, recruitment for collective causes, and formation of group identity (Garrett 2006).

Of course, the reduced cost of conducting political acts in the virtual sphere is often denigrated as “slacktivism” – arguing that the increasing ease of gaining politically engaged feelings is bought at the expense of participation in consequential activities for social change (Morozov 2011). Theocharis et al. (2015) argue that, while social media afford important channels for information and conversation during the heydays of social movements in Spain, Greece, and the US, these are limited to communications, and they lack the agency of organizing and coordinating protest. This finding suggests that the Internet manifests its mobilizing potential in the process by which participation in protest comes mainly from a densely knit cluster of those who are predisposed to political action and affiliated with formal organization networks, rather than bridging diverse social groups for a common cause (Van Laer 2010).

Nevertheless, the Internet is conducive to democratization of the sense of political effectiveness, in that the less educated feel empowered to engage in the political process to a greater degree than the educated (Sasaki 2017). We agree with this view and, therefore, argue that Internet use encourages protest participation because the new information source bridges communication networks that connect protesters with like-minded individuals and strengthen collective identity for contention. That is, Internet use facilitates the mobilization process through which individual grievances are easily connected for collaborative endeavors to address shared problems, so that incentives for participation go beyond mere altruism. But the technology lowers the barriers to protest communications by providing easy access to low-cost, less-demanding activities as well. As a result, Internet use may increase the willingness to participate, termed “protest potential” by Barnes and Kaase (1979), without necessarily increasing actual participation. We thus propose two hypotheses:

H1

Internet use is positively associated with increased willingness to participate in protest activities.

H2

Internet use is positively associated with actual participation in protest activities.

SOCIAL CAPITAL, PSYCHOLOGICAL ORIENTATIONS, AND POLITICAL INTEREST

Regarding the political role of Internet use, there are skeptics who view it as conducive to social withdrawal or reinforcement of social inequalities. Among these scholars, Putnam (2000) attributes the increasing decline of civic engagement through traditional institutions to the widespread use of electronic media technology. That is, Internet use does not substitute for the loss of face-to-face social interaction but rather displaces real-life community activities (Nie 2001). For those with identification with formal groups rather than virtual ones, moreover, not acting for the cause of the organization is a cost, rather than a benefit, because of its moral sanctioning mechanism (Opp 2009). Also, the increased control that Internet users have over their exposure to media content can facilitate a desire to avoid public matters (Prior 2007). This pessimistic view supports the thesis of a “vicious circle” in which the pre-existing participation gap between the engaged and disengaged is reinforced (Norris 2001). Sociological inequality and endogenous orientations are relevant to this reinforcement trend: the information rich get richer (Price and Zaller 1993, 138).

From the reinforcement perspective, social capital is the underlying cause of structural variation in the opportunity to be mobilized for protest participation. For instance, closely knit interaction with fellow citizens broadens citizens’ range of interests and experiences that make community problems more relevant (Moy and Gastil 2006; Olsen 1972). Similarly, Pollock (1982) posited formal associations as an agent of mobilization in which people are informed about, oriented toward, and trained for engaging in the public domain as a byproduct of their group activity. Political mobilization also occurs in the context of informal conversation (Lake and Huckfeldt 1998; Kwak, Shah, and Holbert 2004). In a social-mediation process, exchanges of information and opinions foster political cognition, attitudes, and behaviors by increasing the opportunity to have access to out-of-bounds information about common grievances (Scheufele et al. 2004). The findings suggest that the exposure to non-like-minded ideas through frequent conversations motivate people and enhance attention to the acquisition and processing of information in support of political participation (Scheufele 2002). Interpersonal trust provides another dimension for the mobilization mechanism through which social interactions encourage protest participation. Putnam (1995) argues that trust, as a cultural component of social capital, facilitates acting on matters of common concern. This is because interpersonal trust strengthens the perception that cooperative actions work for mutual benefit (Benson and Rochon 2004).

While mobilization theory emphasizes individuals’ response to external opportunities for being invited to and organized for action, the reinforcement theory gives primary to intrapersonal resources for being participatory (Leighley 1995). In this view, socioeconomic status can matter for involvement in political life. A better education enables people to have cognitive skills and social networks that help elaborate on public grievances and have adequate ways to take part in protest action (Verba, Schlozman, and Brady 1995). Youth is an important factor for prioritizing alternative forms of activism over conventional paths to a public life in either online or offline settings (Bennett, Wells, and Rank 2009). Gender-related inequity is also conducive to the participation gap in this context (Hargittai and Hsieh 2013).

Moreover, attitudinal sources of unconventional political participation cannot be ignored. The psychological origins of protest action, specifically, include discontent about life and distrust in the government, as well a self-placement on the ideological left–right spectrum (Dalton, Van Sickle, and Weldon 2010; Pierce and Converse 1990). Protest participation is also incentivized by those who lack formal representation and political institutions for making claims on the state (Eisinger 1973). In that regard, a sense of identification with or support for the ruling party reduces the need for political action outside institutions.¹ Unconventional political participation is also related to the embrace of postmaterialist values that affect not only the level of expressing political opinion but also the type of activities performed (Inglehart and Welzel 2005). Inglehart (1997) furthermore argues that economic well-being and educational achievement allow for cultural orientations to self-expression values rather than survival ones, so that people are poised to engage in elite-challenging politics.

When it is viewed as an element in the aforementioned theory of reinforcement, political interest can be treated as a crucial resource that has been consistently confirmed to determine political participation. Bimber (2003) suggests that greater political interest reduces the costs of being informed and increases the perceived benefits of the participation. More importantly, Xenos and Moy (2007) found that even if equal Internet access occurs, a gap in political involvement is posited between the more and less politically interested. That is, political interest affects the process by which the Internet disproportionately benefits those who have intrapersonal resources that facilitate pro-civic orientations. Since individuals are given greater agency in accessing social capital and consuming information on the Internet, their civic skills and motivation to use the technology for a political purpose should matter (Prior 2007). The preexisting disparity between the engaged and disengaged is reinforced by the Internet even if both are involved in the mobilization structure (Norris 2001). However, in the context of Internet use to encourage protest participation and the potential for it, the moderating role of political interest has rarely been examined with respect to its presumable role in translating personal grievances into collective concerns. Therefore, we want to consider how political interest affects protest behavior.

COMPARING OPPORTUNITY STRUCTURES IN KOREA AND TAIWAN

According to McAdam, Tarrow, and Tilly (2009), the political opportunity structure is understood as the landscape of incentives for and constraints on protest participation in relation to social, economic, or institutional situations. That is to say, structural opportunities for protest participation are subject to how formal representation is constituted and whether activists perceive a favorable signal of political support and public opinion (Meyer and Minkoff 2004). In this view of political-opportunity theory, Korea and Taiwan provide an interesting opportunity for comparison with respect to the rise of contemporary political mobilization in which the participants are afforded low cost-of-access to protest communications through Internet use.

First, both countries are “third-wave” democracies that achieved a successful transition to democracy in late 1980s. The two tigers in East Asia are indeed distinct from regional neighbors, except for Japan, in which authoritarian regimes have endured. But they also witness “a haze of nostalgia for authoritarianism” (Chang, Zhu, and Park 2007), insofar as authoritarian rule had been in power under martial law established by the military in Korea and by the single-party Kuomintang (KMT) regime in Taiwan. And, during the period of dictatorship, both countries achieved effective state-sponsored industrialization and poverty reduction under a centralized bureaucracy constituted of highly educated officials (Evans 1995).

In that regard, the neighboring democracies in East Asia are often considered with respect to their “bounded” way of democratization. Slater (2012) points out that the presidential designates, Roh Tae Woo in Korea and Chiang Ching-kuo in Taiwan, compromised public demands for liberalization by initiating and implementing the reform program with their own hands. For example, Roh Tae Woo decided to hold a direct election of the Korean president and Chiang Ching-kuo lifted Taiwanese martial law so that opposition parties were legally formed to contest the elections. Ironically, their response to the pressure of the dissident movement, backed by middle-class support for liberalization, kept the ruling parties, Democratic Justice Party of Korea and KMT, as the mainstream in political circles. Thus, activists and dissidents have been constrained by such a political environment that increased the perceived cost of access to the political system so that extra-institutional action was encouraged.

Furthermore, since the early 2000s, Korea and Taiwan have gone through substantive development of online platforms and expansion of accessible technologies based on their advanced infrastructure. And the diffusion of Internet use has afforded new structures of popular mobilization, given the growth of an online public sphere dominated by educated youth and civil activists rather than vested information-providers and traditional journalists (George 2006; Lee 2015). For such actors who are marginalized in political circles, cyberspace was indeed a source of hope and opportunity that would connect with the mass public and empower their voice to mobilize protest communications. And both Asian tigers have the opportunity structure for widespread Internet use to help activist groups who organize and promote a large grass-roots campaign force (Shin 2005).

In Korea, for example, the media market is characterized by strong development of mass-circulation newspapers that are privately owned, as well as by television broadcasters at least partly under the control of the government. The top three national dailies—Chosun, JoongAng, and Dong-A—have dominated the newspaper market by serving a middle-class readership (Kwak 2012). As of 2008, the three privately owned newspapers accounted for nearly 58 percent of printed newspaper subscribers. Two major broadcasting networks—the Korean Broadcasting System (KBS) and the Munhwa Broadcasting Corporation (MBC)—predominated in the media-rich market. But they are public broadcasters regulated by the Korea Communications Commission in which the ruling party appoints a majority of commissioners (Haggard and You 2015).

Moreover, Korea has recently witnessed the soaring influence of privately owned enterprises in the media market because newspapers and conglomerates have allowed for investment in the broadcasting sector since 2011. The size and competition of the market have increased so that commercialization is prevalent among the news agencies seeking a wider readership and more advertising revenues. Thus, the Korean situation makes dissident voices hard to hear in the mediated formation of public discourse, so that dissidents and activist groups without institutional allies need to resort to unconventional, direct methods for engaging the public.

Compared with Korea, the Taiwanese media system has a relatively low-circulation press while the news market has not yet come to the peak of commercialization. An important reason is the market dominance of party-owned or party-affiliated newspapers, as well as a vigorous external pluralism among cable news channels (Hong 1999; Tiffen and Kwak 2005). The competition between party allies became so dynamic that news agencies became an important channel for dissent publicity. They were able to do so with support from the pro-independence Pan-Green Coalition's opposition to the pro-unification Pan-Blue Coalition, and, consequently, they encroached on the Kuomintang (KMT)-dominated media landscape (Lo 2012).² These partisan allies provide dissidents with mediated access to the formal political process, which disincentivizes mobilization of high-cost unconventional acts to some extent (Lee 2011).

However, the Taiwanese public recently suffered the diminished pluralism of the mainstream news market, which has motivated the opposition and its allied groups to seek the Internet as a means of networked advocacy and protest communications. Because of the deepening commercial relationship with Chinese corporations, media owners and practitioners have shown growing self-censorship on matters affecting mainland China for the sake of protesting their financial interests (Freedom House, 2013). At the same time, Taiwan faces the Chinese government's increasing attempts to influence media outlets via advertising, so civil activists express and spread anxiety about the erosion of press pluralism. The Internet, therefore, emerges as an important structure for protest communications opposing the increasing pro-China views in the Taiwanese media.

The Korea–Taiwan comparison in this study reveals a meaningful aspect of political opportunity structures for Internet use to encourage protest politics. This structural condition should not be ignored, since media systems are institutions that provide certain macro-level constraints on individual attitudes and behavior (Hallin and Mancini 2004). It is therefore important to examine whether Internet use encourages not only willingness to participate in protest activities but also actual participation, given the context of limited access to formal channels of political communication that give greater incentives for unconventional activity. To be sure, previous research suggests that Internet use in the emerging Asia democracies is positively associated with political participation in non-institutionalized manners (Fu et al. 2016; Lee 2017). But our study contributes to this scholarship by comparing opportunity structures for protest politics in East Asian “bounded” democracy with high Internet penetration.

METHOD

Our data are from the sixth wave (2010–14) of the WVS, which provided individual-level measures of the public's attitudes, beliefs, and behaviors in Korea and Taiwan. The WVS data were gathered by a regional network of research teams, which carried out face-to-face interviews with nationally representative samples of voting-age adults (17–19 years old and above). The sample sizes were 1,200 for Korea and 1,238 for Taiwan, based on stratified random sampling, and the fieldwork was conducted in 2010 and 2012, respectively. In accordance with American Association for Public Opinion Research standards, the calculated response rates were 55.4 percent and 28.4 percent, respectively.

DEPENDENT VARIABLE

Ranging from relatively low-risk legal acts to relatively high-risk illegal acts, protest activities involve different costs and benefits that come to into play systematically in decision-making calculations for participation (Barnes and Kaase 1979). In that sense, unconventional political participation was measured using each of the three WVS items asking respondents’ experience of protest activities, including signing a petition, joining in boycotts, and attending peaceful demonstrations (V85 through V87).³ Also, the protest potential was measured by the items that distinctively asked about willingness to participate. These items offered the following statement: “I'm going to read out some forms of political action that people can take, and I'd like you to tell me, for each one, whether you have done any of these things, whether you might do it or would never under any circumstances do it.” The scoring was 1 = “have done,” 2 = “might do,” and 3 = “would never do.” Table 1 shows descriptive summaries of each item.

Table 1 Descriptive Summary of Dependent Variables

Data From: World Values Survey Wave 6, Online Data Analysis

INDEPENDENT AND CONTROL VARIABLES

• • Internet use: This ordinal variable was measured by an item that tapped the frequency of Internet use to learn about what is going on in the country and the world (V223). The original scoring ranged from 1 (daily) to 5 (never), but this item was reverse coded for ease of interpretation (Korea: M = 3.89, SD = 1.58; Taiwan: M = 3.26, SD = 1.86).

• • Political interest: We used an item (V84) that assessed how interested respondents said they were in politics. This variable was coded: “not at all interested,” “not very interested,” “somewhat interested,” and “very interested” (Korea: 16.92%, 40.92%, 36.17%, 6.00%; Taiwan: 32.07%, 40.23%, 23.91%, 3.80%).

• • Left–right location: Self-positioning on the ideological scale (V95) was measured by an item that asking respondents to place their views on the left–right scale in general. The scoring ranged from 1 = “left” to 9 = “right” (Korea: M = 5.33, SD = 2.07; Taiwan: M = 4.63, SD = 1.97).

• • Disaffected voters: We measured the level of disapproval of or apathy toward the ruling party based on an item (V228) asking respondents which party they would vote for. A binary variable was created by differentiating respondents who did not support the then ruling party from those who did so (Korea: 64%; Taiwan: 63%).

• • Distrust in government: This ordinal variable was measured by an item (V115) that asked how much confidence respondents had in the government. The scoring ranged from 1 = “a great deal” to 4 = “none at all” (Korea: M = 2.44, SD = 0.77; Taiwan: M = 2.40, SD = 0.80).

• • Governance criticism: Critical evaluation of democratic governance (V141) came from an item asking respondents how democratically their country is being governed today. Reverse coding of responses was on a 10-point scale, ranging from 1 = “completely democratic” to 10 = “not at all democratic” (Korea: M = 5.04, SD = 1.94; Taiwan: M = 4.13, SD = 2.28).

• • Life dissatisfaction: This ordinal variable was counted by using an item (V23) asking respondents how satisfied they are with their life as a whole these days. Reverse coding of the scoring ranged from 1 = “completely satisfied” to 10 = “completely dissatisfied” (Korea: M = 4.39, SD = 1.86; Taiwan: M = 4.14, SD = 2.09).

• • Post-materialism: an item offered a materialist/post-materialist 12-item index (Y001) the scoring of which ranged from 0 = “materialist” to 5 = “postmaterialist” (Korea: M = 2.13, SD = 1.07; Taiwan: M = 1.63, SD = 1.18).

• • Newspaper use: This ordinal variable was measured by an item on the frequency with which the respondents learned what was going on in their country and the world and from a daily newspaper (V217), coded as a five-point ordinal scale from 1 = “never” to 5 = “daily” (Korea: M = 3.47, SD = 1.55; Taiwan: M = 3.62, SD = 1.55).

• • Interpersonal conversation: This ordinal variable tapped into an item measuring the frequency with which the respondents learned what was going on in their country and the world from talks with friends or colleagues (V224). The original scoring ranged from 1 (daily) to 5 (never), but the item was reverse coded for ease of interpretation (Korea: M = 4.16, SD = 1.14; Taiwan: M = 3.32, SD = 1.49).

• • Interpersonal trust: A binary variable was derived from the item (V24), distinguishing respondents who reported trust in other members of society from those who did not (Korea: 29.67%; Taiwan: 30.21%).

• • Group membership: Regarding social capital, group membership was measured by 11 items that asked respondents if they were an active member of each type of social organization.⁴ A binary variable was created to distinguish those who reported active membership in any organization from those who did not (Korea: 39.08%, α = 0.67; Taiwan: 51.13%, α = 0.73).

• • Socioeconomic factors: Three items were included to control for gender (V240; Female; Korea: 50.67%; Taiwan = 52.10%), age (V242; Korea: M = 43.17, SD = 14.94; Taiwan: M = 45.62, SD = 17.34), education level (V248; 9-point scale; Korea: M = 7.56, SD = 1.78; Taiwan: M = 6.65, SD = 2.44), and household income (V239: 10-point scale; Korea: M = 5.01, SD = 1.81; Taiwan: M = 4.72, SD = 1.69).⁵

MISSING DATA

Using the WVS data, 27 survey-based items were translated into 15 independent and control variables in our model of protest participation. For the Korea model, most variables had fewer than 3% missing values, with the exceptions of organizational membership (11.1%) and disapproval of the ruling party (3.3%). In the Taiwan model, four variables had more than 3% missing values: self-positioning in political ideology scale (8.8%), disapproval of the ruling party (18.7%), distrust in the government (5.5%), critical evaluation of democratic governance (5.3%), and the postmaterialist 12-item index (6.3%). Concerning the incomplete data, therefore, the mi package in the R environment was employed to impute missing values (Su et al. 2011), as outlined by Kropko et al.: multiple imputation for the data with missing values was performed by drawing imputed values iteratively from “the conditional distribution for each partially observed variable, given the observed and imputed values of the other variables in the data” (2014, 501).⁶

DATA ANALYSIS

Since the dependent variables were categorical, a multinomial logistic regression model was run on each of the datasets for Korea and Taiwan by means of the R environment.⁷ The baseline model included a chain of mobilization, socioeconomic, and psychological variables to predict whether an individual reported previous participation in protests or willingness to participate rather than answering “would never do.” In the regression, survey weights were included because response rates to items on political attitudes and behaviors might differ by socioeconomic status. Survey weighting could not be ignored in this context.

RESULTS

Before looking more closely at the results it is worth looking at the descriptive differences between Korea and Taiwan in the level of both willingness to participate and actual participation in protest politics. As can be seen in Table 1, in virtually all categories of action Taiwan citizens are somewhat less willing to engage in these activities than Koreans. This tendency cuts across different protest activities ranging from signing a petition, as a relatively accessible act, to actually attending peaceful demonstrations. While the share of respondents joining in boycotts are the same, many more Koreans are also willing to engage in this activity (47.1% to 21.6%) and in all categories of action than Taiwanese. In other words, the share of Taiwan respondents who would never participate in the activity is significantly higher than that of Koreans. This means there is a difference in the protest repertories between the two neighboring countries in East Asia.

To address the proposed hypotheses, we examined how Internet use was associated with unconventional political participation and the potential for it. But the interpretation of the parameters in the logistic regression model is difficult in that the dependent variable is assumed to be linearly related with the independent variable through the logit transformation. Thus, the multinomial logistic estimates are reported with odds ratios for ease of interpretation.

Above all, if we look at the willingness to protest, it is clear that the Internet has an effect on all types of action. That is to say, the mobilizing impact of Internet use was robust when it comes to the protest potential. Not only across action repertoires but also between the countries the willingness to participate was consistently greater among those who reported more frequent use of the technology. Particularly, a one-unit increase in the frequency of Internet use increased the odds of being inclined toward signing a petition, attending peaceful demonstrations, and joining in boycotts by 22%, 23%, and 11%, respectively, in Korea (see Table 2), and by 26%, 15%, and 20%, respectively, in Taiwan (see Table 3). Therefore, the WVS data support H1.

Table 2 Predictors of Unconventional Political Participation in South Korea

Note: Entries are odds ratios and standard errors in parentheses. Sig. * p < 0.05; ** p < 0.01; *** p < 0.001. The reference category of unconventional political participation is “would never do.”

Table 3 Predictors of Unconventional Political Participation in Taiwan

Notes: Entries are odds ratios and standard errors in parentheses. Sig. * p < 0.05; ** p < 0.01; *** p < 0.001. The reference category of unconventional political participation is “would never do.”

When we look at actual protest, however, the findings are more mixed. There is some evidence of Internet use driving low-cost political participation, but it is equivocal when it comes to the higher cost of action. Specifically, for a one-unit increase in Internet use for information, we expect to see about 29 percent and 34 percent increases in the odds of having signed a petition among people living in Korea (see Table 2) and Taiwan (see Table 3) respectively. Given attending peaceful demonstrations, a one-unit increase in Internet use predicted a 31 percent increase in the odds of having such an experience of protest participation among Koreans. But it was not related to joining boycotts. In Taiwan, on the other hand, a one-unit increase in Internet use predicted a 46 percent increase in the odds of having joined in boycotts. But the Taiwanese Internet was not associated with an actual experience of attending peaceful demonstrations. Thus, the findings lend only partial support for H2 insofar as the impacts of Internet use are mixed and different between the countries. This suggests that, even if the two Asian democracies differ in the protest repertories, Internet use exerts a bounded influence on protest participation that is subject to the cost of action.

In addition, the findings indicate that the mobilization effects of Internet use are important in opportunity structures for protest politics in East Asia. In the Korean model, for instance, there are the structural barriers to actual participation in protest, in that the only variables that had significantly positive effects, other than education and group membership, were Internet use and political interest (see Table 2). This “third-wave” democracy shows that the Internet provides ordinary citizens with easy access to protest communications that overcome the institutional constraints in an unprecedented manner.

Of course, the findings show important differences in protest politics between Korea and Taiwan. Basically, the Taiwanese model accounts for variance in protest behavior to a greater degree than the Korean model by 11 percent for signing a petition, 6 percent for attending peaceful demonstrations, and 10 percent for joining in boycotts. Also, the Taiwanese data demonstrate that group membership has overarching effects on actual participation in all types of protest acts. And interpersonal conversation and social trust, which are not significant in Korea at all, increase the odds of having signed a petition in Taiwan (See Table 3). This between-country gap suggests that mass mobilization for protest politics in Korea is more conditional on situational opportunities, which go beyond individual-level factors, compared with the relative robustness of structural or dispositional origins of participation in Taiwan.

DISCUSSION

Before discussing the results, one limitation should be addressed given our secondary data analysis. The WVS data do not provide nuanced measures of Internet use for social interaction and the resulting network characteristics such as network size and strength of social ties. These measures are nonetheless important for assessing how the Internet serves as a mobilization structure for political participation through its capacities for generating or accessing social-capital resources (Ellison, Steinfield, and Lampe 2011; Gil de Zúñiga and Valenzuela 2011). This is particularly the case given the decline of formal civic organizations and traditional media systems (Putnam 2000). Bakker and de Vreese (2011) also found that, above and beyond informational media use, the socializing activities of youth online were related to their unconventional forms of political participation. In the same vein, along with their rapid adoption of social media, younger people may display a narrowing gap in mobilization opportunities for protest participation. Future studies should address the increasing role of social-media use in information flows and political involvement.

Nevertheless, one advantage of this study lies in the comparative approach that situates the model within institutional and structural settings of “third-wave” democracies in East Asia. The two Asian tigers—Korea and Taiwan—have witnessed the widespread use of the Internet for protest politics in which alternative web-based media are playing a growing role in engaging and mobilizing the public. The diffusion of social media has also led citizen journalism to become prevalent among those who are dissatisfied with the institutional working of traditional media. In that regard, the results of this study contribute to our understanding of unconventional political participation in the context of increasing Internet use for collective organizing. Specifically, Internet use facilitates the participation in a relatively legal protest action such as signing a petition. This mobilization capacity of the new communication networks appears more dramatic as a facilitator of protest inclination than in other contexts. Regardless of forms of action or contexts of political opportunities, Internet use reduces the cost of considering protest participation. And this pattern of relationship transcends the two democracies in East Asia. The results suggest the effectiveness of “loosely tied, opt-in/opt-out networks” in mobilization of Internet users for low-cost protest activities in that a strong commitment to traditional agents of social movements is not necessary for the participation (Bennett and Segerberg 2013).

Of course, our interpretation of the results may invite criticism that the effects of Internet use would discriminate between those who are predisposed to the participation and those who are not on the basis of intrapersonal resources. Indeed, political interest has been found to not only determine political participation directly but also moderate the effects of Internet use on that participation (Bimber 2003; Xenos and Moy 2007). This view of a participation gap echoes the reinforcement thesis that highlights differentiated benefits of technology between the privileged and the marginalized (DiMaggio et al. 2004).

However, when a post-hoc analysis was conducted to examine whether the effects were conditional on political interest, we found that the Internet rarely adds to the existing inequality in protest politics. The WVS data show that, no matter how uninterested in politics respondents were, Internet use increased willingness to participate in all types of protest activities. Although the effects of Internet use on actual participation were more modest than the protest potential, the technology-mediated mobilization did not differentiate between those who “very interested” in politics and those who were “not at all interested.” This suggests that the Internet provides the mobilization structure for protest communication offers “networks of recruitment” for collective action (Verba, Schlozman, and Brady 1995), such that the propensity to consider engaging in protest transcends preexisting disparities in personal resources for political action. In the East Asian context of democracy, the new communication network encourages protest potential that is more a matter of mobilization than of personal psychology. This loosely organized but network-based protest mobilization explains why contemporary political activism seems to entail empowerment of the traditionally disengaged, given the above-mentioned cases in the region.

Our comparative study adds to the understanding of political opportunity structures for contemporary social movements in Korea and Taiwan, where ordinary citizens are getting easier access to protest communications more and more through the Internet. Previously, the well-established structure of the mass media was one of the main constraints on protest politics, as political institutions are organized to increase the structural barriers to the representation of dissident voices. But as seen in the recent large-scale social movement against President Park, the Internet facilitated mass mobilization of ordinary citizens and effective coordination of protest activities. Also, a small group of Taiwanese students who were digitally enabled enough to live-stream their occupation developed into the historical political movement against a Taiwan–China trade treaty. In the Taiwanese context, protest communication could reach a wider public through online social-mediation, which is relatively free from the Chinese-government-imposed censorship of political information. Thus, our comparative study contributes to the scholarship on social movements by showing the importance of opportunity structures that enable the Internet to encourage protest politics.

Will the greater diffusion of Internet use promote democracy in East Asia? East Asian politics has been characterized by social norms at odds with elite-challenging politics (Shi 2015), such that protest activities have been restricted to small civil-society groups and college-student associations (Castells 2008). But digital networks prioritize personalized communication over formal organization in the mobilizing process by which dissident voices are organized to address shared grievances. In doing so, Internet use has increasingly emerged as an important pathway to elite-challenging politics among those who are marginalized by institutions but are equipped with digital platforms for personalized politics (Bennett and Segerberg 2013). That is, the technology has changed the dynamics of protest communication and mobilization.

To be sure, the new technology for protest communication has limited effects on actual participation in relatively more extra-institutional activities. Thus, some may argue that the new mobilization structure by itself is not sufficient to overcome the free-rider problem in collective action. Given the “slacktivism” thesis, furthermore, the democratization of Internet-enabled access to low-cost acts can come at the expense of deepening citizenship that is manifested by more consequential activities to challenge the status quo (Morozov 2011). Also, others may be wary of the growing trend toward popular mobilization of those who lack civic skills or citizens’ democratic norms but reinforce networked activism by homophily—an inherent tendency to connect to and bond with the like-minded. In the digitally mediated process of mobilization, protest potential also arises among the disgruntled, who have a low-cost means of personalized communication through algorithmic recommender systems. This structure of political-information flow is also isolating the collective identities of the protesters from a plurality of citizen voices. In doing so, the Internet amplifies political polarization or fragmentation in which participatory behaviors are mobilized in the absence of shared discourses and democratic deliberation. Future studies should take into consideration this new mechanism of mobilization through which protest participation occurs without strong qualities of citizenship.

Today, when disgruntled citizens protest, the Internet is necessary for grassroots organizing and the expansion of an action repertoire. The present empirical analysis of WVS data also suggests that the emerging pathway to unconventional political participation does not mirror the traditional social bases for mobilization. And, in doing so, cheap mobile devices and digital social networks facilitate low-cost protest acts for public mobilization in such a way that personal grievances are connected for collective action. Activists and dissident groups may benefit from such a new mobilization structure by forming “filter bubbles” (Pariser 2011), so much so that they are keen to reach and organize their like-minded support and are adept at it (Howard 2015). Technological potential, observed in East Asian democracies, can be therefore related to the expansion of personalized, flexible networked activism that arises at the expense of declining traditional civic and political commitments.