This study takes the position that political ideology on a left-right scale cannot be appropriated to the East Asian case. Citing Kinder (1998), Feldman (2003) asserts:
“Although politicians, philosophers, and social scientists often discuss politics along a single left-right dimension, 50 years of research on public opinion show that a
unidimensional model of ideology is a poor description of political attitudes for the overwhelming proportion of people virtually everywhere” (p. 76).
Unfortunately, ABS data does not have a measure for the left-right scale, and thus, this position cannot be properly tested before conducting this study’s analysis. This is precisely because some proponents of the survey itself argue that there is no clear understanding of left and right among East Asians even if such self-proclaimed political parties do exist in some countries.
We also find support for not using a singular dimension of political ideology as a
predictor for immigration attitudes in Hlavek’s (2011) research, which uses the variables laissez-faire views, social traditionalism, and civic/national pride. He contends that for cross-national comparisons, the left-right index can be problematic because the understanding of what is left or right may vary from one country to another. In Wu (2017), one recommendation in dealing with the unidimensional model problem is to “directly measure the substance of left/right ideology rather than using it as it is” (p. 77).
Thus, in this study, instead of choosing political ideology per se as the explanatory variable, we use two political attitudes: defensive nationalism and social traditionalism. These are related to an existing theoretical perspective identified in literature – the significant influence of symbolic interests, such as values and personal identifications with a group, to the formation of opinions on immigrants and immigration (Ceobanu & Escandell, 2010; Hainmueller &
Hopkins, 2014). Additionally, we also include social trust – a belief that people in general are trustworthy – as a third explanatory variable, in light of its widely-accepted role as “the glue for more integrated and cooperative communities” (Herreros & Criado, 2009).
In political psychology, extensive research exists on the relationship between values and political attitudes (Feldman, 2003), and scholars have also proposed multidimensional models of political ideology by linking it with specific value systems (eg. Rokeach, 1973; Braithwaite, 1982). The most relevant would be the body of theories on the economic and social dimensions
‧
of ideology (Feldman & Huddy, 2014). Schwarz (1994), one of the earlier proponents of this model, labels the two dimensions ‘economic egalitarianism’ and ‘classical liberalism’. While economic egalitarianism “refers to whether the government should devote itself more to promoting equality by redistributing resources or to protecting citizens’ ability to retain the wealth they generate in order to foster economic growth and efficiency” (p. 40); classical liberalism – the social dimension – “refers to whether government should devote itself more to guarding and cultivating individual freedoms and civil rights or to protecting the societal status quo by controlling deviance from within or enemies from without” (p. 39).
Some studies find evidence that in East Asia, the concept of cultural threat is more significant than economic threat when it comes to views of immigrants (eg. Green & Kadoya, 2013, 2015; Ha & Jang, 2014). The two-dimensional model previously described allows us to decompose the left-right spectrum and focus instead on classical liberalism, or the social
dimension of political ideology. Two major theories can link nationalism – as part of this social dimension – with anti-immigrant attitudes.
4.1. Social Identity Theory
In 1979, Tajfel and Turner developed and introduced social identity theory (SIT), which later became one of the most prominent theoretical positions when it comes to the social
psychology of intergroup behavior. Moving from earlier theories of identity shaped by mostly individual interactions, social identity is based on group membership, which can be an important source of pride and self-esteem. However, social groups are associated with either positive or negative value connotations, and thus, social identity is also evaluated as either positive or negative.
SIT identifies three cognitive processes that individuals go through in defining their in-group and out-in-group classifications. First, social categorization enables people to simplify reality by organizing individuals based on the group they belong to, and generally involves emphasizing similarities and differences between groups. Although someone can be part of several categories, one becomes more salient depending on the circumstances, drawing attention to the performative aspect of identity. During social identification, individuals adopt the identity of their group, such that they start being emotionally invested in being a part of collective, as well as acting like how a group member is expected to act. This is the stage when individuals’ self-esteem becomes
‧
bound to group membership. Because people want to maintain positive feelings about
themselves, they will then resort to social comparison and measure their group against others.
Eventually, negative feelings about the status of one’s group can lead to social competition and prejudice.
Following Pedersen et al. (2005), we contend that national identity is a “specific form of social identity”, where “the more strongly a person identifies with a particular group, the more that group and its relations with other groups will affect that person and direct his or her beliefs and behaviours” (p. 150). By itself, national identity does not necessarily lead to out-group rejection (Jeong, 2013), but some studies have linked heightened nationalism with prejudice (Pettigrew & Meertens, 1995; Pedersen & Walker, 1997). There is also evidence that strongly identifying with one’s in-group is correlated with negative evaluations of out-groups
(Mummendy, Klink, & Brown, 2001). Thus, nationalism can conjure an image of who ‘others’
are, and serve as a natural divide between ‘us’ and ‘them’.
For example, if language can be considered a boundary marker between nations, then people who believe that speaking their national language is a very important part of their identity, culture, and tradition and will hold those who do not speak it as ‘others’ (Barber et al., 2013; Hainmueller & Hopkins, 2014; Lubbers & Coenders, 2016).
4.2. Symbolic Politics Theory
Usually contrasted with self-interest, which is an instrumental approach to an individuals’
goal attainments, symbolic politics theory (SPT) argues that “human beings are intensely
concerned about remote and abstract political symbols, even though the emotional costs they pay and benefits they receive from such involvement are modest” (Sears, 1993, p. 113). Through the process of classical conditioning in early life, strong and stable reactions to symbols (could be images, people, words, etc.) – ‘symbolic predispositions’ – are formed, leading people to acquire racial prejudice and identify with certain political parties and ideologies.
Contending that previous literature has overlooked the importance of national symbols as an integral part of SPT, Jeong (2013) proposes to extend the theory to include ‘national feelings’, which may encompass similar but different concepts relating to nationalism, such as national identity and national pride. He reasons that often times, as “an object of strong allegiances…the nation itself serves as a political symbol which rivets citizens’ attention and evoke strong
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
emotions” (Jeong, 2013, p. 1463). He adds that immigration is “expected to be strongly associated with what individuals feel about their nation”, especially since immigrants are
“outsiders in contexts where nationhood is the basis of self-categorisation and psychological attachment” (Jeong, 2013, p. 1463). He maintains that immigration attitudes are primarily conditioned by how an individual understands the meanings and symbols attached to his/her nation, such that he/she reacts negatively to those who seem to violate what a nation and/or people stand for. Thus, SPT can be theoretically linked to SIT, since the strength of the nation as a symbol is also anchored on the construction and adoption of an individual’s identity in relation to the national group he/she belongs to.
Testing his argument with the 2004 General Social Survey, Jeong (2013) finds that different kinds of national feelings evoke different reactions to immigration. National pride, which is self-referential and not conditioned upon comparison with others, can lead to positive attitudes. However, nationalism, defined as “a feeling of superiority and contempt for
foreigners”, is inherently related to out-group derogation and does indeed contribute to anti-immigrant views. National identity, on the other hand, seems to be a neutral feeling.
4.3. Right-Wing Authoritarianism
Nationalism can be viewed as an ideology in itself – grounded in authoritarianism, strongly endorsed by conservatives (the right-aligned), and often involving the support of political leaders (Huddy & Khatib, 2007; Schatz, Staub, & Lavine, 1999). As established in this study’s literature review, nationalism has long been associated with exclusivist worldviews and the conservative right (Fischer, 2009; Skenderovic, 2007).
Thus, nationalism can be related to the political-psychological concept of right-wing authoritarianism (RWA), a type of personality that exhibits high degrees of three interrelated traits: conventionalism, or agreement with tradition and societal norms; submission to
established authorities in society; and aggression in the name of authorities towards those who violate traditional norms (Altemeyer, 1981). Contemporary theoretical frameworks and expanding empirical evidence indicate that RWA can predict prejudice. In the context of immigration attitudes, natives with high RWA react unfavorably towards immigrants who deviate from cultural norms and whom they perceive to be threats to security and national
‧
identity (Cohrs & Stelzl, 2010). However, some have criticized this unidimensional approach, arguing that although the three traits related, they are also distinct (Duckitt et al., 2010).
In line with these criticisms and my earlier contention that ideology in East Asia cannot be labeled as simply left or right, we only use the first component of RWA – conventionalism – as an explanatory variable. Duckitt et al. (2010) calls this ‘traditionalism’, which “expresses the value and motivational goal of maintaining traditional lifestyles, norms, and morality, and resisting “modern” liberal, secular, open lifestyles…likely to stem from threats of disruptive social changes…” (p. 691).
In this study, it is termed ‘social traditionalism’, and it is important in upholding social order and harmony in society. The logic is that prior to their assimilation and integration, immigrants will be seen as outsiders to a national community, and citizens who adhere more to society’s conventions and traditions will tend to have less favorable views toward them as potential violators of the norm and bringers of different thoughts and beliefs.
4.4. Social Trust
Politicians and academics alike believe that national identity is a prerequisite for the kind of social trust and solidarity that building a democratic welfare state requires (Sandelind, 2012).
It increases understanding among people who already have reasons to collectively define themselves as one group. However, social trust can either be particularized or generalized.
According to Rothstein and Uslaner (2005), particularized trust is limited to people’s in-group and is indicative of social strains, where each individual or group only serves its own interests and have little faith in the good intentions of others. Generalized trust, on the other hand, “links us to people who are different from ourselves” and “reflects a bond that people share across a society and across economic and ethnic groups, religions, and races” (Rothstein & Uslaner, 2005, p. 45). Uslaner (2002) further argues that social trust has a moral foundation, wherein
‘trusters’ are compelled to treat strangers as if they are trustworthy, regardless if they share the same views on policy issues or ideology.
Herreros and Criado (2009) proposes that a ‘social truster’ is someone with ‘social intelligence’, “an individual who has learned to discern better which people are trustworthy” (p.
341). This can lead a person to trust strangers because he/she does not base his/her judgement anymore on racist or cultural stereotypes. Social trusters can then extend their trust to those who
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
are ethnically and/or culturally different from them, including immigrants. As generalized trust can reduce ethnocentrism, it can also serve as a foundation for multiculturalism (Kaltenthaler &
Miller, 2013). If particularized trust and a strong national identity lead to social cohesion within a national community, generalized trust can be one of the key ingredients in promoting social cohesion within the same community even when a foreign population is present.
However, trust has also been associated with happiness and well-being, where ‘winners’
with higher socio-economic status – the wealthier and the more educated – can afford to risk trusting their interests to others (Inglehart, 1999; Putnam, 2000). In contrast, societal ‘losers’
who are poorer and more pessimistic about life can be less trusting and more suspicious of others. Consequently, social trust is higher in richer societies that have more equality (Rothstein
& Uslaner, 2005).
4.5. The Hypotheses
Based on previous literature and the theoretical framework presented, three hypotheses are formulated:
H1: Individuals who display higher degrees of defensive nationalism are less likely to support immigration.
H2: Individuals who believe more in social traditionalism are less likely to support immigration.
H3: Individuals who exhibit higher levels of generalized social trust are more likely to support immigration.
‧
In this study, the relationship between the outcome of interest – support for immigration in three East Asian countries – and three key explanatory variables, including defensive
nationalism, social traditionalism, and social trust, is explored. Using data from the fourth wave of the Asian Barometer Survey (hereafter, ABS4), three kinds of logistic regressions are applied applied in the empirical analysis – binary, ordered, and multinomial.
5.1. The Data
The ABS4 is a cross-national, face-to-face representative survey under the supervision of the Center for East Asian Democratic Studies in National Taiwan University, with headquarters in Taipei, Taiwan. The fourth wave was conducted in 14 countries, including Japan, Korea, and Taiwan, between 2014 and 2016. A model ABS usually has a sample size of around 1,200 respondents, which will vary depending on a certain country’s population. The total sample size for the three East Asian countries being studied is 3,938 respondents. Table 3 summarizes the sample and population sizes (at survey time), coverage areas, and survey administration dates for Japan, Korea, and Taiwan.
Table 3. Background Information on ABS Wave 46
Country Population Size Sample Size Coverage Areas Survey Dates
Japan 126.9 million 1,081 January to
March 2016 South Korea 50.42 million 1,200 all, except Jeju
Island
October to December 2015
Taiwan 23.43 million 1,657
6 municipalities 3 cities
13 counties
June to
November 2014
Since Wave 1 was conducted in 2001, this was the first time that a survey item on immigrants was added to the questionnaire under the section ‘Globalization’, suggesting that
6 Information is sourced from the Asian Barometer Survey website and the actual dataset. Data on coverage areas for Japan was unavailable at the time of writing.