Applying Sociological Theories of Emotions to the Study of Mass Politics: The Rally-Round-the-Flag Phenomenon in the United States as a Test Case

ABSTRACT

A remarkable progress has been made in integrating emotions into studies of various aspects of social life, but sociological theories of emotions, which center on group membership and meaning-making, have not been applied to the study of political attitudes and behavior. In order to demonstrate the utility of integrating sociological theories of emotions into the analysis of large-scale political phenomena, this study revisits the rally-round-the-flag effect (i.e., sudden increases in public support for national leaders during war or security crisis). The article claims that rallies are driven by emotional reactions to leaders’ rhetoric promising to restore the nation’s collective honor and status through military action. Analysis of survey data collected during and between two rally periods (2001–2003) in the United States supports this argument vis-à-vis competing theories of attitude formation that ignore the role of emotions or apply a non-sociological framework that detaches emotions from collective identities and meaning-making.

KEYWORDS:

• Sociology of emotions
• peace, war, and social conflict
• rally-round-the-flag
• nationalism
• September 11
• Iraq War
• public opinion
• identity theory
• collective status

View correction statement:

Correction

Acknowledgments

The author wishes to thank Rogers Brubaker, Michael Mann, Robert Mare, David Sears, and Peter Bentler who provided help and advice regarding theory and research methods; to Philippe Duhart, Jennifer Eggerling-Boeck, Wesley Hiers, and Zeynep Ozgen who helped to edit the manuscript; and to Leonie Huddy for generously sharing data. The author is especially grateful to Andreas Wimmer for his mentorship and for providing help and advice regarding theory, research design, and report.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. Including how socialization shapes emotional dispositions (Gordon 1989; Pollak and Thoits 1989; Power 1984), emotion management in social interactions (Hochschild 1979; Kemper 1981; Smith-Lovin and Heise 1988), and the relationship between social categories and emotions (Harlow 2003; Shields et al. 2006).

2. For a few exception to political sociology’s general lack of interest in emotions, see Barbalet (2006), Berezin (2002), Demertzis (2006), Flam (1990), Heaney (2011), and Ost (2004). Some students of social movements have also highlighted the role of emotions (Aminzade and McAdam 2001; Emirbayer and Goldberg 2005; Goodwin, Jasper, and Polletta 2000).

3. In addition to wars and security crises, nation-disrupting events include some national elections, major policy announcements, and even natural disasters.

4. This type of rhetoric draws temporal comparisons (Mummendey, Klink, and Brown 2001; Nigbur and Cinnirella 2007) between the present state of the nation and its actions and either the perceived national past or what the public perceives as the original or ideal moral essence of the nation based on the national mythology. Such discourse often becomes nostalgic, contrasting an inadequate present with an idealized past (Hutcheon and Valdés 2000:21), thus making the mythical memories about what the nation used to be or how it used to behave the “blueprints for the future” (Smith 1986:177).

5. This is only a briefly sketch of the role of the media in RRTF processes. For extended discussions, see Baum and Potter (2008), Bennett and Paletz (1994:several chapters), and Groeling and Baum (2008).

6. Two other examples illustrate this pattern: At the onset of Operation Desert Storm in January 1991, CNN’s prime time rating was ten times higher than normal and the ratings for the morning news shows increased by 15 percent (Hallin and Gitlin 1994); after the invasion of Iraq in March 2003, the average number of people watching prime time cable news per minute increased from about one million in February to about 3.5 million in April (ibid:173).

7. During rally processes, the media often undertake this type of meaning-making purposefully and explicitly. For example, following the September 11 attacks, the New York Times added a daily advertisement-free section titled “A Nation Challenged” that included reports about the challenges security forces faced in their efforts to protect the nation against terrorism as well as reports about the challenges facing ordinary Americans, such as how to maintain positive relations between Muslim and non-Muslim Americans (Carey 2003:75).

8. A transition from “cold cognition” to “hot cognition” (Cerulo 2010).

9. This reinforcement of national identity standards was especially important in the wake of the humiliating failure of the U.S.-sponsored coup in Cuba during the Bay of Pigs invasion – when the United States let others do its dirty work and thus did not act in a way that reflected identity standards (Feinstein 2016).

10. Borrowing from the terminology of the appraisal theory of emotions, the term “positive emotions” is used here in reference to emotions that stem from a positive orientation toward a situation (e.g., joy) or toward the actual or expected ability to cope with a negative situation (e.g., pride, hope, confidence); by contrast, “negative emotions” stem from a negative orientation toward a situation and its causes (e.g., grief, fear, anger) or the lack of ability to cope with the situation (e.g., shame, embracement) (see Folkman 2008 for review).

11. Recall, for example, how congresspersons and senators from both parties stood on the steps of the U.S. Capitol, holding hands and singing “God Bless America.”

12. Data on presidential approval rating were retrieved from the American Presidency Project [online]. Santa Barbara, CA. Available one the World Wide Web: https://www.presidency.ucsb.edu/statistics/data/presidential-job-approval.

13. Gallup. 2003. Gallup/CNN/USA Today Poll # 2003–19: US at War. March 22–23, 2003.

14. For technical information about TNSS see Appendix B.

15. Re-running the analysis with the omitted non-significant control variables did not change the results.

16. Gallup, CNN, and USA Today 2001 (September).

17. CBS News and New York Times 2001 (September).

18. CBS News, and New York Times 2001 (September); Gallup 2001 (September); Gallup, CNN, and USA Today 2001 (September).

19. Model estimation was done with the Arbitrary Distribution Generalized Least Squares (AGLS) method, using the Lee-Poon-Bentler technique of estimating correlations among ordinal variables. Each of the variables had only few missing observations, but because list-wise deletion would have resulted in the loss of about half of the cases, pair-wise deletion was used. All models were re-estimated using the Maximum-Likelihood method with Jamshidian-Bentler EM-type missing data procedure. The differences in the parameters estimates produced by the two methods are minimal.

20. This combination of rage toward terrorists and positive feelings about the war in Iraq was especially common among those who embraced the official White House rhetoric accusing Iraq of sponsoring terrorists; more on this rhetoric in the next section.

21. Only about one percent of respondents moved from opposing to supporting the president.

22. For the same reason, two control variables – income and gender – were also dropped from the model.

23. Jacobson (2010) discussed a cognitive bias associated with partisanship that emerged in this period, arguing that Republicans suppressed information from the media about the real premises of the war in Iraq, while Democrats were eager to absorb negative information about the war.

24. As an additional test, I estimated fixed-effects models (results not shown). All findings were similar to those reported in Table 3, except one: the coefficient for anger toward those who criticized the United States was not statistically significant.

25. House and Senate votes on a bill authorizing the president to use military power against Iraq (which was passed in 2002) highlighted the controversial character of the invasion of Iraq: While only 39 percent of Democrat congresspersons and 61 percent of Democrat senators supported the bill, support rates among Republicans were 97 percent in the House and 98 percent in the Senate (Jacobson 2010). By contrast, on September 18, 2001, a bill authorizing military action in Afghanistan or against any countries, groups, or individuals involved in terrorist activities directed at the United States passed with almost unanimous support among members of both parties (420–1 in the House and 98–0 in the Senate).

26. These accusations against Iraq were not grounded in definite intelligence assessments (Entman 2003; Kaufmann 2004; Rampton and Stauber 2003) and were made despite the CIA’s official opinion that Iraq was not cooperating with al-Qaeda, that its contribution to anti-American terrorism was minor, and that it was not coming close to developing nuclear weapons (Hersh 2003; Kaufmann 2004; Zulaika 2009:194–5).

27. Later, the criticism increased when American troops stationed in Iraq found no evidence of weapons of mass destruction.

28. For a discussion of this division within American nationalism, see Lieven (2004).

Additional information

Notes on contributors

Yuval Feinstein

Yuval Feinstein is a senior lecturer in the department of sociology at the University of Haifa, Israel. His main line of research examines the relationship between political attitudes and the ideas, beliefs, symbols, and affective tendencies that compose national identities. Several of his studies jointly seek to better understand the emergence of “rally-round-the-flag” effects during wars and security crises. He also studies interethnic relations in deeply divided societies, as well as the development and global spread of the nation-state in the past two centuries. His articles have been published in various peer-reviewed journals, including (but not limited to) the American Sociological Review, Social Science History, Sociological Science, Politics & Gender, Social Science Research, Sociology, and the Journal of Ethnic and Migration Studies.