On the Automaticity of Nationalist Ideology: The Case of the USA

How do people typically form opinions and make choices about national matters? What factors infl uence how people behave in the political realm? The majority of work on these questions in psychology and political science has emphasized the role of consciously accessible infl uences on political behavior, such as political party affi liation and an assortment of principles and values. Although there is no doubt that people consciously consult their nationalist ideologies in order to make decisions in the political arena, we focus in the present chapter on the possibility that nationalist ideological knowledge also operates automatically to guide and infl uence opinions and actions. Much work in social, cognitive, and political psychology would suggest that stored representations of nationalist-relevant knowledge, attitudes, goals, and behaviors can be activated by nationalist symbols and cues outside of people’s awareness and intention, and then shape and infl uence their behavior. We consider this possibility in detail, and also summarize recent empirical fi ndings in support of it. ON THE AUTOMATICITY OF AMERICAN NATIONALIST IDEOLOGY In so many little ways, the citizenry are daily reminded of their national place in a world of nations. However, this reminding is so familiar, so continual, that it is not consciously registered as reminding. The metonymic image of banal nationalism is not a fl ag which is being consciously waved with fervent passion; it is the fl ag hanging unnoticed on the public building. (Billig, 1995, p. 8) The results of a massive Zogby poll on the views of American voters in the months before the 2004 presidential election were summarized in a report entitled “E Pluribus Duo: Red State versus Blue State America” (Zogby, 2004). The play on the American national motto of “E Pluribus Unum” ( From Many, One ) refl ects the theme evident throughout the polling results, which is that Americans showed remarkable divergence in their opinions on issues related to war, religion, the economy, and many others, according to whether they 3070-161-1pass-003-r02.indd 53 9/15/2008 1:32:10 PM 54 IDEOLOGY AND AUTOMATICITY identifi ed as Republican or Democrat. This political divide continues today, in the lead-up to the 2008 presidential election, with Democrats and Republicans disagreeing strongly on the same basic issues. For example, whereas 77% of Republicans said they thought the Iraq Surge was making the situation better, only 28% of Democrats thought so according to a recent Gallup Poll (Saad, 2008). Another recent Gallup poll found that whereas the majority of Republicans (60%) believe that humans were created in their present form by God 10,000 years ago, only 38% of Democrats agreed (Newport, 2008). And, whereas nearly three-quarters of Republicans (72%) said they approve of the job President Bush is doing, only 9% of democrats said so, according to a Gallup poll ( Jones, 2008). These polling numbers, and others like them, would seem to suggest that political expression and behavior in the United States is determined largely by party-based ideologies, an impression that is supported by decades of research on the predictive validity of party identifi cation for voting behavior (e.g., Bartels, 2000; Bassili, 1995; Brady & Sniderman, 1985; Campbell, Converse, Miller, & Stokes, 1960; Cohen, 2003; Conover & Feldman, 1984; Green, 1988; Green, Palmquist, & Schickler, 2002; Miller & Schanks, 1996; Petrocik, 1989). If party affi liation is the ideological engine that drives much political behavior in the United States, it is notable that it is often understood as a consciously accessible and intentional infl uence. People can report it easily (e.g., Green et al., 2002; Miller & Shanks, 1996), and intentionally base their political decisions on it (e.g., Bassili, 1995; Cohen, 2003). At fi rst glance, this would seem to imply that much political expression and behavior proceeds in a conscious and intentional manner. And yet, there is also a rapidly growing literature on how choice and behavior can also proceed automatically (e.g., Bargh, 2007; Dijksterhuis, 2004; Ferguson & Hassin, 2007; Greenwald & Banaji, 1995; Hassin, Uleman, & Bargh, 2005; Quattrone & Tversky, 1988; Todorov, Mandisodza, Goren, & Hall, 2005; Zaller, 1992). This literature would predict that, whereas people might consciously behave in line with their explicitly avowed political ideologies and values, they may also be unintentionally and nonconsciously moved by subtle political and ideological cues in their environment. This possibility would indicate that people might answer polling questions, or make voting decisions, in line with ideological cues of which they are unaware, and by which they would prefer to remain uninfl uenced (Wilson & Brekke, 1994). In the current chapter, we fi rst consider research and theory that suggests how ideological knowledge might infl uence people in an automatic fashion and then discuss how nationalist ideologies in particular might operate implicitly. We then review recent fi ndings concerning American nationalist ideology and consider directions for future research. 3070-161-1pass-003-r02.indd 54 9/15/2008 1:32:11 PM On the Automaticity of Nationalist Ideology 55 THE AUTOMATICITY OF IDEOLOGY The word “ideology” (idéologie) was created by the French philosopher Count Destutt de Tracy in the late 18th century to denote a “science of ideas” (de Tracy, 1817–1818), and has since been one of the most widely studied topics across the social sciences (e.g., see Augoustinos, 1999). Although its defi nition varies somewhat according to the specifi c discipline, an ideology can generally be understood as “a set of consensually shared beliefs and doctrines that provide the moral and intellectual basis for a political, economic, or social system” ( Jost, Fitzsimons, & Kay, 2004, p. 265). One important element of such a defi nition concerns the prescriptive nature of the knowledge. This defi nition signifi es that a set of political opinions, values, principles, attitudes, and behaviors that together point toward a particular political system or perspective can be considered ideological in nature (see also Althusser, 1994; Hawkes, 2003; Minar, 1971; Mullins, 1972; Zizek, 1994). How are people infl uenced by their ideologies? There is no doubt that people are sometimes knowingly and purposefully infl uenced by their explicitly avowed ideologies and values. When considering national policies, people often deliberately consult the platform of the political party with which they identify (e.g., see Bassili, 1995; Cohen, 2003). However, there is a long history in sociology, social theory, political science, and social psychology of the idea that people might also be infl uenced by ideological knowledge in a more automatic manner. In seeking to explain the infl uence of ideology on thought and action, sociologists, social psychologists, and political scientists alike have converged on the metaphor that ideologies are like habits or practices that people perform spontaneously and unintentionally (e.g., Bem & Bem, 1970; Berger & Luckmann, 1966; Freeden, 2000). Like all habits, ideologies effectively narrow down choices and “free the individual from the burden of ‘all those decisions’ ” (Berger & Luckmann, 1966, p. 53). More directly, Althusser (1994) argued that “Ideology has very little to do with ‘consciousness’—it is profoundly unconscious” (see also Altemeyer, 1998; Bem & Bem, 1970; Brewer, 1979; Sidanius & Pratto, 1999; Sidanius, Pratto, & Bobo, 1994, p. 999; Tajfel, 1982). From this widely shared perspective, an ideology might be likened to an invisible guide, which not only effi ciently shapes a person’s thought and behavior, but also ultimately prescribes and helps to maintain a certain economic or cultural order. Interestingly, although numerous scholars across the social sciences have speculated on the automaticity of ideology, little empirical support exists for such claims. One notable exception is recent work by Jost and colleagues on system justifi cation (e.g., Jost, Banaji, & Nosek, 2004; Jost, Fitzsimons, & Kay, 2004; Jost & Kay, 2005; Jost, Pelham, & Carvallo, 2002; Kay & Jost, 2003; 3070-161-1pass-003-r02.indd 55 9/15/2008 1:32:11 PM 56 IDEOLOGY AND AUTOMATICITY Kay & Jost et al., 2007; see also Devos & Banaji, 2005). These researchers have argued that people possess ideologies that (implicitly) justify the economic, cultural, and social systems of society, even at the expense of personal or group interests. Such an ideology, for example, leads to disadvantaged members of society holding the very stereotypes and beliefs about their group that would seem to perpetuate their marginality. Those classifi ed in the lower strata of personal income may be especially likely to react to an instance of injustice (and thus, a perceived threat to the legitimacy of the system) by implicitly activating a stereotype that mollifi es or invalidates that injustice. For instance, when such people hear about the fact that 1% of people in the United States possess more than 38% of the wealth in this country (Wolff, 2000), they may implicitly generate the stereotype of “poor but happy, rich but miserable” and thereby feel less threatened by their “econom ically poor” but “psychologically rich” situation (see Kay & Jost, 2003). This work suggests that the perception of an injustice might automatically activate a principle or axiom that mitigates that injustice. We seek to expand on this work by investigating other kinds of implicit ideological effects, and we describe this approach below. IDEOLOGY FROM A SOCIAL COGNITIVE STANDPOINT How might ideological knowledge exert its infl uence in an implicit manner? Research in social and cognitive psychology over the last two decades demonstrates that a range of information is typically associated in memory with a given stimulus and can become activated on the mere perception of that stimulus (e.g., Bargh, 2007; Ferguson & Bargh, 2004; Greenwald & Banaji, 1995; Hassin et al., 2005). For instance, people possess associations in memory between group members and attitudes, stereotypes, behaviors, and goals related to that group. This body of work implies that a range of cues in our environment, such as political icons and symbols, might be associated with various types of ideological knowledge and information. These types of information would likely be diverse, and could include values, principles, exemplars, beliefs, expectations, behaviors, emotions, and motives (e.g., Carlston & Smith, 1996; Smith & Queller, 2001). Furthermore, the perception of one element of that array of knowledge should serve to activate the other elements, even without the person’s awareness or intention (e.g., Devine, 1989). This then suggests that the (conscious or nonconscious) perception of any of the cues interconnected with such ideological information should inevitably activate that information. What happens once such a potentially rich array of information has been unknowingly and unintentionally activated in memory? Research suggests 3070-161-1pass-003-r02.indd 56 9/15/2008 1:32:11 PM On the Automaticity of Nationalist Ideology 57 that this activated information can then infl uence not only the specifi c stimuli to which the perceiver then attends, but also the ways in which the perceiver will interpret and react to subsequently encountered stimuli more generally (e.g., Bargh, 2007; Dijksterhuis, Chartrand, & Aarts, 2007). For example, the subliminal perception of a member of a stereotyped group can activate in memory specifi c trait information, which can then infl uence how the perceiver interprets the ambiguous behaviors of an interaction partner (Bargh, Chen, & Burrows, 1996; Higgins, 1996). In a similar way, the perception of nationalistic symbols or icons, such as the American fl ag, would be expected to activate associated attitudes, beliefs, goals, and behaviors, which would then have the potential to implicitly infl uence one’s subsequent thought and behavior. It is important to note that the kinds of information that can become associated with political icons do not necessarily have to be endorsed by the perceiver. A long history of psychological research suggests that people spontaneously, easily, and sometimes nonconsciously learn about the covariation among stimuli. For example, research on implicit learning suggests that people can encode sophisticated and complex relations among nonsense stimuli according to an artifi cial grammar, even while remaining consciously unaware of such learning (e.g., Anderson, 1983, 1996; Howard & Howard, 1992; Kelly, 2003; Reber, 1993; Seger, 1994). The notion of unintentional learning also manifests in social psychological work on stereotypes and prejudice. Through a variety of channels, people unknowingly learn about various beliefs, attitudes, and behaviors in relation to certain groups. And, importantly, even though people can actively disagree with such implicit associations, the associations can nevertheless infl uence both subtle and overt behavior toward group members (e.g., for reviews see Bargh, 2007; Greenwald & Banaji, 1995; Hassin et al., 2005). The fact that implicit attitudes, beliefs, goals, and behaviors can be dissociated from intentional ones more generally suggests that people may also possess implicit ideological knowledge with which they would explicitly disagree. One guiding objective of our research on this topic is to examine the existence and scope of this kind of dissociation. We now turn to more specifi c questions about the characterization, content, and operation of nationalist ideology. We also discuss whether and why knowledge that becomes associated with nationalist symbols would tend to be ideological in nature—that is, supportive of the respective systems. AMERICAN NATIONALIST IDEOLOGY Given our discussion of how ideologies in general might operate implicitly, what are the potential characteristics of nationalistic ideologies in particular? 3070-161-1pass-003-r02.indd 57 9/15/2008 1:32:11 PM 58 IDEOLOGY AND AUTOMATICITY Although nationalist ideologies have been defi ned in myriad ways across disciplines and scholars (e.g., Billig, 1995; Breuilly, 1992; Gellner, 1983; Kosterman & Feshbach, 1989; Skitka, 2005; van Dijk, 2005), we view a nationalistic ideology as consisting of beliefs, attitudes, goals, and behaviors that together prescribe certain economic, political, and social systems for a particular nation. That is, a nationalistic ideology consists of the constructs that characterize a particular nation, distinguish it from others, and support its continued existence. For example, nationalistic ideology might consist of information that supports a given nation’s existing form of government (e.g., democracy, dictatorship), economic system (e.g., capitalism, socialism), and various social institutions (e.g., marriage, education, religion, family). It should be noted that we are therefore not limiting our examination of nationalist ideology to citizens’ expressed positive attitudes toward, and commitment to, their nation, as has been the case with previous research on “nationalism” in particular (e.g., Kosterman & Feshbach, 1989; Skitka, 2005). Instead, we consider a range of constructs that are implicitly associated with one’s nation, which can include both positive attitudes toward the nation as well as the variety of constructs mentioned previously. It is for this reason that we refer throughout this chapter to “nationalist ideology” rather than “nationalism.” It is perhaps worth reiterating that we do not characterize as implicit nationalist ideology anything that is implicitly associated in memory with the nation. Instead, in line with the defi nition of ideology we described earlier, we only consider as nationalist ideology those memories associated with the nation that are supportive of or consistent with current, national systems. This would include memories that could conceivably endorse and preserve the institutions, programs, policies, and perspectives of the respective nation. This distinction means, for example, that we would not consider the memories underlying the colors red, white, and blue to be by themselves ideological in nature, even though they are surely closely associated in memory with the American fl ag. However, any memories that are consistent with key American foreign or domestic policy (e.g., aggression) or values would qualify as ideological. Moreover, any information that increases regard for such policies or values would also be considered ideological. For instance, we discuss in the section on recent empirical work the prediction that the perception of U.S. cues implicitly evokes greater positive regard and endorsement for the materialistic values and behaviors theorized to be an inherent component of capitalistic/market-based systems (e.g., see Kasser, Ryan, Coucheman, & Sheldon, 2003). Although we are therefore making a distinction between ideological and nonideological nation-associated knowledge (i.e., prescriptive versus 3070-161-1pass-003-r02.indd 58 9/15/2008 1:32:11 PM On the Automaticity of Nationalist Ideology 59 nonprescriptive), we should emphasize that the information that people tend to learn about the nation is not randomly generated. The sources of much of the information that gets disseminated about the nation within the respective populace should by and large consist of national systems (e.g., media, government, corporations). As such, the nation-related information that is widely dispersed within a citizenry can, on some level, be expected to be largely supportive of the systems of that respective nation (e.g., see Herman & Chomsky, 1988). We should also note that while there presumably is a great deal of unanimity across citizens (at least within stable nation-states) with regard to their attitudes toward broad governmental and economic systems (e.g., capitalism, materialism), people should nevertheless vary in the degree of their support for both broad principles, as well as more specifi c political stances, positions, and current events. What would predict or explain such differences in nationalistic ideological knowledge of citizens of the United States? Although we have tested for multiple moderators, as we discuss throughout the empirical section below, the moderator that has so far proven important in explaining the variance in the results is the participants’ reported exposure to American political news. Why would news exposure matter? We argue that the primary source of information about the nation (in the case of the United States) is the national news media. People acquire information about the country through news programs on television, magazines, radio programs, and the Internet. Whereas some people follow political news about the nation closely, such as news concerning the war in Iraq and the war on terror, for example, others are less politically engaged and watch the news only rarely, if at all. In this way, the information that becomes associated with the United States, and in turn with national cues, should be affected to some extent by whether people pay attention to current events and national politics. We discuss this moderator in more detail in the empirical section below, as well as in the discussion section. Empirical Evidence for Implicit American Nationalist Ideology How do nationalistic ideologies become activated? Ideological knowledge concerning one’s nation is typically explicitly measured. For example, researchers might ask respondents to indicate their support of various governmental or economic systems, or the degree to which they value the nation, on Likert-type scales (e.g., Kosterman & Feshbach, 1989). Instead of asking people to knowingly and intentionally provide their opinions and attitudes concerning the nation, however, we are interested in those types of knowledge and information that are associated with the nation in memory and that can become implicitly activated in memory on the perception of nationalistic 3070-161-1pass-003-r02.indd 59 9/15/2008 1:32:12 PM 60 IDEOLOGY AND AUTOMATICITY stimuli. Thus, for our main experimental paradigm, we use symbols of the nation (e.g., the national fl ag) as primes, and then test the types of knowledge and information that become activated and infl uential on subsequent attitudes, beliefs, and behavior. In most of the research reported here, we used one of the most ubiquitous national cues in the United States—the national fl ag. Although the fl ag has been a highly visible national symbol since the founding of the country, its presence has greatly increased since the terrorist attacks on the United States in September of 2001 (Gerstenfeld, 2002; Kellner, 2003). The pervasive presence of the fl ag in the lives of Americans enhances the ecological validity of any fi ndings that emerge. In the research we summarize here, we examined three constructs—power, materialism, and aggression. More specifi cally, in each of the experiments we describe here, we subtly primed participants with either the American fl ag or a control stimulus, then measured their responses in terms of one of these three constructs. In each section, we review our predictions and then the fi ndings. First, however, we briefl y discuss political news exposure as a moderator.