ABSTRACT
Addressing the normative and empirical debate regarding the nature of patriotism, this paper examines the social contexts in which patriotism – defined here as an expression of national pride – thrives. Combining diverse theoretical explanations, it investigates whether expressions of patriotism are related to globalization, state function, social fractionalization and conflict. A multilevel regression analysis of data from 93 countries led to three principal findings. First, citizens of more developed and globalized countries are less likely to be proud of their country. Second, citizens are more likely to be patriotic in countries characterized by higher levels of income inequality and religiously homogeneity. Third, citizens of countries exposed to direct conflict – that is, suffering terror and causalities from external conflict – tend to exhibit higher levels of national pride. Patriotism frequently being identified as a mandatory political commodity, these results suggest that, overall, patriotism forms part of a less attractive matrix than its advocates tend to assume.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1. See Wagner et al. (2012) for a detailed exploration of this topic.
2. Data for replication are available at online Appendix A(see Table 1A in Appendix A).
3. 58% of the respondents choosing ‘very proud’, we replicated the analysis with the outcome variable as a dummy variable (see online Appendix D). The average missing answer for the question relating to national pride was 4.8%.
4. Based on the military balance documented by the International Institute for Strategic Studies.
5. Organized conflict is defined as at least 25 fatalities per year from armed conflict between two parties, of which at least one is the government of a state, the data being taken from the Uppsala Conflict Data Program. Terrorism is defined as the weighted average of the number of fatalities, injuries and property damage caused by terrorism over the past 5-years, taken from the Global Terrorism Database.
6. We used grand mean centring throughout the models, the estimation method adopted being RML (Restricted Maximum Likelihood).
7. See for correlations.
8. Model 5 in (see Appendix D) indicates that the results are sensitive to the way in which the outcome is defined. When the latter is defined in binary terms, reflecting the skewed to the ‘very prod’, the results are more significant.
9. ISSP: National Identity II, 2003. Distributor: GESIS Cologne Germany ZA3910.