Introduction
The COVID-19 pandemic seems to have increased the appeal of national identity, which was mobilized to ensure solidarity (Berrocal et al. Reference Berrocal, Kranert, Attolino, Santos, Santamaria and Henaku2021; Gülseven Reference Gülseven2021; Mylonas and Whalley Reference Mylonas and Whalley2022). As Yael Tamir stated, “In times of crisis, when social solidarity is necessary, it is useful to turn to the flag, the hymn, the national symbols, in order to rally people around a common task. Nothing exemplifies this better than the present Corona crisis” (Reference Tamir2020, 451). Tamir’s claim reflects the broader national identity argument that was already developed in the 1990s and according to which national identity provides a type of cement or glue that hold societies together and ensures solidarity (Miller Reference Miller1995; Tamir Reference Tamir1993). The national identity argument was originally motivated primarily by the issue of cultural diversity, but the COVID-19 pandemic seems to have provided new evidence of the mobilization of national identity for invoking solidarity.
Given the policy implications of the national identity argument, several studies have explored whether national identity does, indeed, promote solidarity (Breidahl, Holtug and Kongshøj Reference Breidahl, Holtug and Kongshøj2018; Gustavsson Reference Gustavsson, Gustavsson and Miller2020; Gustavsson and Miller Reference Gustavsson and Miller2020; Gustavsson and Stendahl Reference Gustavsson and Stendahl2020; Johnston et al. Reference Johnston, Banting, Kymlicka and Soroka2010; Kongshøj Reference Kongshøj2019; Rapp Reference Rapp2022; Reeskens and Wright Reference Reeskens and Wright2013). However, these studies’ results are rather inconclusive and contradictory (Miller and Ali Reference Miller and Ali2014). Likewise, empirical investigations about the influence of the pandemic on national identity have, thus far, been indecisive, with some suggesting that national identity supports certain types of pro-social behaviour while others have pointed to the contrary (Chan et al. Reference Chan, Wang, Zuo, Chiu, Liu, Yiu and Hong2021; Lim and Prakash Reference Lim and Prakash2021; Rupar et al. Reference Rupar, Jamróz-Dolińska, Kołeczek and Sekerdej2021; Sibley et al. Reference Sibley, Greaves, Satherley, Wilson, Overall, Carol and Milojev2020; Vignoles et al. Reference Vignoles, Jaser, Taylor and Ntontis2021).
The current article set out to examine the national identity argument using the COVID-19 pandemic as a test case of solidarity. It inspects one specific aspect of the national identity argument during the pandemic, namely, the link between two dimensions of national identity – national attachment and national chauvinism – and one dimension of emotional solidarity – caring for those vulnerable groups who were affected by the pandemic. The article asks two specific questions: 1) Were national attachment and national chauvinism during the COVID-19 pandemic related to emotional solidarity? and 2) Did levels of national attachment and national chauvinism increase or decrease during the pandemic?
In order to answer these questions, this article adopted a longitudinal research design to analyze changes in the relationship between national identity and solidarity during the pandemic. Taking Israel as a test case, a three-wave panel survey was conducted including national attachment and national chauvinism measures from both before and during the pandemic. The survey is used to trace changes over time and to determine whether differences emerged in the relationship between national identity and emotional solidarity. Overall, there was a clear reduction in the levels of national attachment and national chauvinism over time. Moreover, the positive connection between national attachment and solidarity grew stronger over time, while the connection between national chauvinism and solidarity grew weaker and insignificant over time. In addition, at the time of the pandemic, Israel was mired in a deep political crisis of instability and repeated election cycles. The article therefore also discusses the extent to which this crisis might have influenced national identity regardless of the pandemic and the role of political elites in enabling the mobilization of national identity to ensure solidarity in time of crisis.
The National Identity Argument and the Pandemic as a Test Case for Solidarity
The national identity argument asserts that national identity guarantees cooperation, social cohesion, solidarity, and welfare (Miller Reference Miller1995; Tamir Reference Tamir2020). In the face of social diversification and other features of modern societies, national identity is seen as essential for supporting democracy and social justice (Gustavsson and Miller Reference Gustavsson and Miller2020). Identification with the nation and with strangers as compatriots creates a shared identity, which enables people to trust and cooperate with people they do not know personally. Such a trust promotes solidarity and support for the redistribution of resources between members of the nation, thus facilitating policies that benefit those worse off in terms of housing subsidies, unemployment benefits, and other such policies that require the public to overcome its self-interest. National affiliation serves as an essential framework for the flourishing of egalitarian politics. According to this perspective, without a sense of national “we,” redistributive taxation and welfare provision are difficult to justify. The sharing of a recognized culture by co-nationals reflects the significant fact that “communal solidarity creates a feeling, or an illusion, of closeness and shared fate, which is a precondition of distributive justice” (Tamir Reference Tamir1993, 121). Put differently, national identity is an important structural element of the liberal welfare state. Therefore, the state should encourage a specific national culture that enables national identity to embody and implement liberal practices.
Additional support for the national identity argument can be found in social psychology that acknowledges, in line with social identity theory, that a strong sense of identification with the group enhances the likelihood of behavior that will favor group members like co-nationals (Gustavsson Reference Gustavsson, Gustavsson and Miller2020; Reicher and Hopkins Reference Reicher and Hopkins2001). The mechanism of identifying with the in-group – the nation in this case – provides members of the nation with the motivation to help the in-group and forms a source of mobilization for national goals. Many studies following the social identity framework have demonstrated that people like to think positively about themselves and about the groups to which they belong (Brown Reference Brown2000). The social identity theory is crucial for inspecting the national identity argument as it points to the underline mechanism of caring for co-nationals (Gustavsson Reference Gustavsson, Gustavsson and Miller2020). Different outcomes, such as social cohesion, trust, and solidarity, have been described as outcomes of national identity. Following social identity theory, the precondition for solidarity is feelings of care for those who are in need. In the words of a leading advocate of the national identity argument, “When it comes to caring for others, national identity stands out as an important psychological motivation” (Tamir Reference Tamir2020, 541). The current study, therefore, takes advantage of the COVID-19 pandemic and focuses on emotional solidarity, namely, caring for those vulnerable groups who were affected by the pandemic.
The COVID-19 pandemic can also provide an up-to-date opportunity to examine the national identity argument. The pandemic posed an unprecedented challenge to social order across nations with a profound influence in terms of human loss, economic collapse, the unimaginable closure of borders, and others. Among these various effects are two issues that are central to the national identity argument: the possible rise of nationalism and the significance of national solidarity for addressing the consequences of the pandemic. The closure of national borders, the halt of international movement, and the difficulty of mobilizing support across borders have all been seen as evidence of a decrease in the process of globalization and an increase in nationalism (Bieber Reference Bieber2022; Woods et al. Reference Woods, Schertzer, Greenfeld, Hughes and Miller-Idriss2020). With national medical priorities dominating state policy, the pandemic reinforced the centrality of the state as a provider of security and health. The immediate reaction of governments was to reduce international cooperation and to concentrate on domestic needs. It seems that the pandemic brought the national to the front and ceased globalization (Woods et al. Reference Woods, Schertzer, Greenfeld, Hughes and Miller-Idriss2020). The evolving distinction between “us” and “them” is another sign of growing nationalism as states responded by turning inward (Mylonas and Whalley Reference Mylonas and Whalley2022 ; Woods et al. Reference Woods, Schertzer, Greenfeld, Hughes and Miller-Idriss2020). Analyses of crisis speeches by the premiers of several countries have shown how nationalism was mobilized using metaphors of the pandemic as war while calling for solidarity based on nationhood (Berrocal et al. Reference Berrocal, Kranert, Attolino, Santos, Santamaria and Henaku2021).
Assessment of the National Identity Argument
We know very little about the extent to which national identity leads to solidarity. Most social psychology studies on social identity theory were conducted on small groups, which are easy to analyze in lab experiments (Reicher and Hopkins Reference Reicher and Hopkins2001). National identity in day-to-day political and social processes has been much less studied. More significantly, empirical attempts to test the national identity argument have had indecisive conclusions (Gustavsson and Miller Reference Gustavsson and Miller2020; Miller and Ali Reference Miller and Ali2014). This section reviews these empirical assessments in order to explain how the current article differs in its approach.
The most systematic review of these empirical assessments of the national identity argument was conducted by Miller and Ali (Reference Miller and Ali2014) who showed that while some studies support the national identity argument, others do not. For example, a study in Canada (Johnston et al. Reference Johnston, Banting, Kymlicka and Soroka2010) found that people with high national identity scores are a little more likely to support redistribution policies. In the United States, higher levels of national attachment were found related to expressions of feelings of obligation to help fellow group members in many different ways but not to explicit support for welfare programs (Theiss-Morse Reference Theiss-Morse2009). In Britain, national identity was shown to be negatively related to support for the welfare state (Martinez-Herrera Reference Martinez-Herrera, Karolweski and Suszycki2010). Such differences between countries are well reflected in cross-national analyses, according to which feelings of national pride were, on the whole, found to be negatively connected to support for redistribution in dozens of countries (Shayo Reference Shayo2009).
Other studies have also lacked conclusive evidence regarding the influence of the national identity.Footnote 1 A cross-national study across 29 European countries found that there is higher support for the welfare state under ethnic conceptions of national identity but not civic and cultural conceptions (Reeskens and Wright Reference Reeskens and Wright2013). However, a comparative study of Belgium and Germany, which used a different measure of national attachment, found federal rather than regional identification to be related to increased support for redistributive policies but only in richer regions and among richer individuals (Holm and Gays Reference Holm and Geys2018). A study in Denmark offered a different take by inspecting the way in which public commitment to specific shared values is related to trust and solidarity from a belief that these values are shared with other community members (Breidahl, Holtug, and Kongshøj Reference Breidahl, Holtug and Kongshøj2018). It found that both conservative and liberal nationalism are negatively correlated to solidarity, whereas multiculturalism and liberal citizenship are positively correlated. Another study that extended the national identity argument to the case of ethnic minorities demonstrated that stronger attachment to Belgium among citizens of Turkish or Moroccan descent is positively related to support for redistribution (Galle and Fleischmann Reference Galle and Fleischmann2020).
The studies reviewed above were conducted prior to the pandemic and aimed to explicitly explore the national identity argument. In the last two years, several studies inspected the nexus between national identity and various issues that are related to the pandemic. While none of them set out to examine the national identity argument explicitly, their inconclusive findings can be used for exploring the complex links between national identity and solidarity. For example, there is evidence that nationalism is unrelated to pro-social behaviors. A study conducted in the United Kingdom found that national attachment, unlike other types of social identification, does not predict helping others (Vignoles et al. Reference Vignoles, Jaser, Taylor and Ntontis2021). In fact, higher levels of national satisfaction (i.e., being proud to be British) were found to be less helpful for others, perhaps reflecting national hubris. However, when it comes to another type of pro-social behavior, such as disease prevention, national attachment was found to be related beyond other factors in both the United States and China (Chan et al. Reference Chan, Wang, Zuo, Chiu, Liu, Yiu and Hong2021).
The findings regarding the influence of the COVID-19 on national identity over time have also been inconclusive. Higher levels of national attachment and patriotism following lockdown were found in New Zealand (Sibley et al. Reference Sibley, Greaves, Satherley, Wilson, Overall, Carol and Milojev2020). Similarly, in South Korea, there was an increase in general national pride following the pandemic but also a decrease in pride in the quality of domestic politics and democracy (Lim and Prakash Reference Lim and Prakash2021). In China there were no significant changes in national attachment during the pandemic, whereas in the United States there was a significant decrease in national attachment (Chan et al. Reference Chan, Wang, Zuo, Chiu, Liu, Yiu and Hong2021).
There are three reasons for the inconclusiveness of the empirical findings thus far. First is the issue of the conceptualization and operationalization of national identity (Gustavsson and Miller Reference Gustavsson and Miller2020; Miller and Ali Reference Miller and Ali2014), which is not surprising since it is recognized as a multidimensional concept (Huddy Reference Huddy, Grimm, Huddy, Schmidt and Seethaler2016). A recent study has, indeed, shown how different dimensions of national identity are dissimilarly related to different dimensions of solidarity (Rapp Reference Rapp2022). Second, the outcomes of national identity, namely, social cohesion, trust, solidarity, and others, have also been conceptualized and operationalized in different ways. Yet, these outcomes are also multidimensional and different operationalizations have led to conflicting conclusions (Rapp Reference Rapp2022). Of the various possible outcomes associated with the national identity argument, the current study focuses on one basic aspect of solidarity, caring for those affected by the pandemic, since caring is a precondition for other aspects of solidarity. Third, most of the existing studies are cross-sectional and therefore cannot determine whether national identity influences social cohesion or vice versa. They reflect the assumption that solidarity is a by-product of identity (Kaplan Reference Kaplan2021). That is, national identity is presumed to be a stable component of people’s worldview that can be mobilized to promote solidarity. Experimental studies exploring the impact of national identity on tax compliance and motivation to cooperate (Gangl, Torgler, and Kirchler Reference Gangl, Torgler and Kirchler2016), trust (Robinson Reference Robinson2016), and solidarity with minorities (Charnysh, Lucas, and Singh Reference Charnysh, Lucas and Singh2015) have indicated that solidarity might, indeed, be a by-product of identity. However, these studies did not explore the multidimensionality of national identity comprehensively in real-time events. The COVID-19 pandemic provides a real-life event that can be used to inspect solidarity.
The Current Study
This article uses the COVID-19 pandemic in Israel to examine the national identity argument in unsettled times. It examines whether national identity is related to solidarity. National identity is defined as a pervasive cognitive and affective orientation of lay people toward their nation (Bonikowski Reference Bonikowski2016). As illustrated above, previous studies of the national identity argument have also explored different dimensions of national identity, with the sole consensus to date being that national identity is multidimensional (Huddy Reference Huddy, Grimm, Huddy, Schmidt and Seethaler2016). This article makes no attempt to include all the dimensions of national identity examined so far; rather, it follows Gustavsson’s (Reference Gustavsson, Gustavsson and Miller2020) suggestion that the focus should be on national attachment. She argued, based on intra-group social behavior, that the strength of emotional attachment to national identity is the key to solidarity with members of the nation. To avoid limiting themselves by only concentrating on national attachment, scholars of national identity commonly use a dual analysis: they examine national attachment (or patriotism) through national chauvinism (sometime labeled nationalism) (see Blank and Schmidt Reference Blank and Schmidt2003; De Figueiredo and Elkins Reference De, Rui and Elkins2003).Footnote 2 While national attachment pertains to one’s level of attachment to the national community and reflects the level of commitment to the national group (Theiss-Morse Reference Theiss-Morse2009), national chauvinism is defined as the feeling that one’s nation is superior to others (Schatz, Staub and Levine Reference Schatz, Staub and Lavine1999; Wagner et al. Reference Wagner, Becker, Christ, Pettigrew and Schmidt2012).
Solidarity is usually used to describe the bonds between members of social groups and their levels of commitment to each other and to the group as reflected in support for redistribution (Banting and Kymlicka Reference Banting, Kymlicka, Banting and Kymlicka2017; Gustavsson Reference Gustavsson, Gustavsson and Miller2020). National solidarity implies solidarity with co-nationals (Banting and Kymlicka Reference Banting, Kymlicka, Banting and Kymlicka2017). However, the various meanings of solidarity and its scales are disputed as it is discursively constructed to legitimize conflicting political actions and public mobilization (Wallaschek Reference Wallaschek2020). Moreover, relations between national identity and solidarity are influenced by the consideration of who counts as a member of the national community from the outset (Theiss-Morse Reference Theiss–Morse, Gonzalez, O’Toole, Gustavsson and Miller2020).
Solidarity has been conceptualized and measured in different ways across studies (Rapp Reference Rapp2022). The current study focuses on solidarity during the COVID-19 pandemic, specifically emotional solidarity expressed by, as mentioned above, caring for those vulnerable groups who were affected by the pandemic. Focusing on emotional solidarity corresponds with Holtug’s (Reference Holtug2022) idea that there are several possible paths linking national identity and solidarity. This study highlights the most basic aspect of emotional care and not other aspects of solidarity, such as support for redistribution or people’s behaviors during the pandemic. Emotional care is the precondition for other manifestations of solidarity, such as allocating resources to those who suffered from the pandemic.
The consequences of the pandemic are a crucial test of national solidarity. While most people suffered from worries about their health and the effect of lockdown, sectors that were already worse off are likely to have suffered more costs. In addition, the pandemic created groups at risk that were not previously worse off: for example, the elderly population whose health is at greater risk or those newly unemployed due to sectors that were shut down. Solidarity toward these groups is critical due to the extensive state resources allocated to support them. This is not a solidarity that is required in settled times but one that is required at times of increasing economic anxiety (Fetzer at al. Reference Fetzer, Hensel, Hermle and Roth2021).
This article asks two specific questions: 1) Was national attachment and national chauvinism during the COVID-19 pandemic related to emotional solidarity?; and 2) Did levels of national attachment and national chauvinism increase or decrease during the pandemic? To answer these questions, a three-wave longitudinal survey was conducted in which participants answered the same set of questions about national attachment and national chauvinism both before (T1, December 2019) and during (T2, June 2020 and T3, January 2022) the pandemic. In the T2/T3 survey they were also asked specific questions about the pandemic and solidarity. The survey can thus be used to trace changes in the participants’ mean attitudes over time and to determine any differences in the relationship between national attachment and national chauvinism and emotional solidarity.
While this article examines, specifically, changes in national identity during the pandemic and how they relate to emotional solidarity, it cannot rule out that other factors besides the pandemic might have also influenced the respondents. As explained below, the intensive political crisis and numerous election cycles during this period might have impacted national identity, solidarity, and the relationship between them.
The Setting: Israel
The national identity argument was developed mainly in the context of established liberal democracies that were looking to increase solidarity in the face of growing diversity. While Israel (within the 1967 borders) is regularly considered and classified as a democracy (Dowty Reference Dowty, Dowty, Hazan, Hofnung and Rahat2018), it is quite far from being an established liberal democracy. The dominant role of Zionism as its national ideology and the status of its non-Jewish citizens, among other factors, have led to the proposal that Israel is a diminished type of democracy like an illiberal or ethnic democracy (Peleg Reference Peleg2007; Smooha Reference Smooha1990). Despite its deviation from the standards of the established liberal democracies, the case of Israel can offer essential insight into the national identity argument. The dominance of national identity (among Jews) and its historical role in justifying in-group solidarity in the face of external threat is well established (Kelman Reference Kelman1999). Jewish Israelis have high levels of identification with their country. For example, in the ISSP National Identity Module III (2013), of the 33 countries included in the survey, the level of “feeling very close to your country” was highest for Jewish Israelis.Footnote 3 Israel is, therefore, a sort of “extreme case” of national identity. The rationale for selecting an extreme rather than a representative case is that it allows for the accumulation of the greatest possible amount of information on a given phenomenon (Flyvbjerg Reference Flyvbjerg2006). In other words, if the assessment of the pandemic impact on national identity and solidarity is not supported by findings in the extreme case of Israel, it is less likely to be found in other countries.
The COVID-19 Pandemic and Solidarity in Israel
T1 data collection was conducted between December 1 and 2, 2019, nearly four months before the outbreak of the pandemic in Israel. T2 data collection started on May 31, 2020 and ended on June 8, 2020. Before T2, Israel government adopted harsh measures that included imposing lockdowns, closing schools, and closing most of the job market. During May 2020, the spread of the pandemic was declining, and restrictions were eased. The government was, by and large, successful in curbing the spread of the first wave (Maor Reference Maor2021). T3 data collection took place during January 2022. Between T2 and T3, there were additional waves of the pandemic following various COVID-19 variants and Israel experienced two additional lockdowns and the spread of vaccines. In the weeks prior to T3, the Omicron variant led to a wave with a much higher morbidity rate than in the previous waves.Footnote 4
The pandemic was not the only possible factor influencing national identity and solidarity during the period of the study. In September 2019, prior to T1, there were general elections whose results did not enable Benjamin Netanyahu to form a stable government with the incumbent right-wing and ultra-Orthodox religious parties. Prior to T2 data collection, on May 17, 2020, a new national unity government was launched under the influence of the pandemic. There is evidence that the forming of this government reduced polarization (Bassan-Nygate and Weiss Reference Bassan-Nygate and Weiss2022) and thus, perhaps, increased both national identification and solidarity. However, the national unity government lasted only a few months, leading to additional elections, which highlighted the fierce clashes between political camps. Netanyahu, the strongest political figure in Israel over the previous decades and the prime minister between 2009 and 2021, was at the center of this debate. Netanyahu was subject to criminal investigations on charges of fraud, bribery, and breach of trust, which further intensified the polarization. After the 2021 elections, Netanyahu lost his office, and the government was formed by other parties, but the political stability was short as additional elections were held in November 2022.
This political crisis, which was described and understood in different ways (e.g., Migdal et al. Reference Migdal, Ben-David, Abulof, Le Penne, Persico, Sebba-Elran and Rosenfeld2021), might have influenced feelings of solidarity and national identity regardless of the COVID-19 pandemic. Responses to the pandemic and judgment of government actions might, in addition, have been affected by political affiliation. While this article’s analysis included measures for voting in the elections as well as for overall estimation of the way the state handled the crisis, its design cannot determine the relative role of the pandemic and the political crisis in shaping national identity, solidarity, and their relationship during this period.
Sampling and Procedure
Participants were recruited via an opt-in online survey, following the practice of previous studies of national identity and other issues in Israel (Barak-Corren, Feldman, and Gidron Reference Barak-Corren, Feldman and Gidron2018; Bassan-Nygate and Weiss Reference Bassan-Nygate and Weiss2022; Feinstein Reference Feinstein2018; Feinstein and Bonikowski, Reference Feinstein and Bonikowski2021).Footnote 5 Population estimates with matching respondents demographic characteristics was used. The participants were invited to take part in a “study about social and political issues” and to complete an online questionnaire in exchange for a sum approximating $2. Overall, 1,197 respondents completed the three surveys and quota sampling was used along demographic characteristics.Footnote 6 While the sample was heterogeneous (54.2% women; age M=47, SD=15; 10.8% born outside Israel; 61.7% secular; 44.3% higher education), opt-in online sampling is not based on probability sample, so it is not a representative sample of the target population. Nonetheless, such opt-in online sampling has been used in numerous studies of national identity in Israel (e.g., Feinstein and Bonikowski Reference Feinstein and Bonikowski2021) and is seen as an approximation of the target population. Using opt-in online sampling enables consistency with the common practice of survey methodology (Evans and Mathur Reference Evans and Mathur2018). The overall attrition from T1 to T3 was 50% and was higher among younger people and males, who were controlled in the analysis.
The target population of the study was limited to Jewish citizens of Israel. While it is common for studies of national identity in Israel to focus only on Jews (see Canetti, Hirschberger, and Rapaport Reference Canetti, Hirschberger, Rapaport, Elad-Strenger, Ein-Dor, Rosenzveig, Pyszczynski and Hobfoll2018; Feinstein and Bonikowski, Reference Feinstein and Bonikowski2021; Sorek and Ceobanu Reference Sorek and Ceobanu2009), such a decision might itself be seen as a form of banal Israeli nationalism. Palestinian citizens of Israel were not included in the current study. The study’s theoretical framework is not suitable for an analysis of Palestinians citizens of Israel because the national narrative that dictates the character of Israel’s national identity excludes them (Jamal Reference Jamal2004). In addition, the repertoire of Palestinian identity in Israel (Amara and Schnell Reference Amara and Schnell2004) requires a totally different theoretical framework, and the identity of Palestinians citizens cannot be fitted into the framework of national identity adopted here. A comprehensive integration of Palestinian identity demands the use of alternative theories and measures to account for the dissimilar interpretation of national identity and thus requires a different study.
Measures
Studies of redistributive solidarity tend to use general questions about support for welfare payments – for those worse off, or for reducing income inequality (Gustavsson Reference Gustavsson, Gustavsson and Miller2020; Shayo Reference Shayo2009). In the current article, the specific situation of the COVID-19 pandemic was used to examine emotional solidarity with vulnerable groups. Relations between national identity and solidarity are influenced by considerations of who is counted as a member of the national community from the outset. Put differently, when assessing the scope of solidarity, the underline assumption is that solidarity is with prototypical members of nation (Theiss-Morse Reference Theiss–Morse, Gonzalez, O’Toole, Gustavsson and Miller2020). When solidarity is examined beyond the scope of prototypical members of nation, it is common to identify specific groups like minorities (Barak-Corren, Feldman, and Gidron Reference Barak-Corren, Feldman and Gidron2018).
However, the current study followed the former approach of considering solidarity with prototypical members of nation, focusing here on solidarity with those affected by the pandemic. The independent variables, national attachment and national chauvinism, were measured based on previous studies (Davidov Reference Davidov2009; Roccas, Klar and Liviatan Reference Roccas, Klar and Liviatan2006). Descriptions of the measures are detailed in Table 1, and a discussion of the operationalization of the measures can be found in Appendix 1.
Table 1. Measures
In addition to these variables, the analysis also included age, gender, education, income, political affiliation, level of religiosity, and voting in the elections. The level of the pandemic’s effects on the participants and estimation of the ways Israel handled the crisis were also measured (see Table 1).
Results
Changes Over Time
The first part of the analysis examined whether national identity measures changed over time. Because the same participants answered the same questions in the three survey waves, change in the means of the measures would have indicated the impact of the pandemic. National attachment and chauvinism were measured across the three time points, so a repeated measures ANOVA was used. Table 2 shows the means across the three time points. The decrease in national attachment, albeit modest, was found to be statistically significant. A repeated measures ANOVA showed that, overall, time significantly affected national attachment [F(1.98, 2376)=7.79 p=.000], but the reduction between T1 and T2 was insignificant.Footnote 7 The decrease in national chauvinism was higher than the decrease in national attachment, as a repeated measures ANOVA showed that, overall, time significantly affected chauvinism, [F(1.46, 1756)=142 p=.000]. The decrease in national chauvinism, unlike in national attachment, was also significant between T1 and T2.Footnote 8
Table 2. National attachment and chauvinism across time
While there is an overall decrease in national attachment and chauvinism over time, this might also be misleading. During this period, Israel was subject to an intense political crisis and increasing polarization, which might have impacted levels of national identity alongside the influence of the pandemic. The focal point of the polarization during this period was the divide between the supporters and opponents of Benjamin Netanyahu. After the fourth election in 2021, even though his party, Likud, remained the largest party with 25% seats of the parliament, Netanyahu failed to form a government. Instead, an alternative government was formed whose coalition included left, center, and right-wing parties as well as, for the first time in the history of the State of Israel, a Palestinian citizens party.
Since T3 was after the forming of the coalition, it enabled an observation of whether the decrease in national identity was only evident among supporters/opponents of Netanyahu based on the 2021 voting. The results of the analysis indicated that a decrease in national attachment and chauvinism could be seen among both those who voted for the Netanyahu government and those who voted against it in the 2021 elections. Levels of national attachment (T1=5.81; T2=5.80; T3=5.67) and chauvinism (T1=5.81; T2=5.75 T3=4.91) were higher among those who voted for the Netanyahu government, which comprised right-wing and religious parties, than among those who voted for the left (national attachment, T1=5.65; T2=5.57; T3=5.54) and chauvinism (T1=4.81; T2=4.65 T3=4.31); however, a reduction in national attachment and chauvinism was found among both groups.Footnote 9
In addition to political polarization, the analysis also explored whether the changes were related to the overall estimation of the ways in which Israel handled the COVID-19 crisis, which was measured only at T3.Footnote 10 Unlike the similar reduction in national attachment and chauvinism among both supporters/opponents of the Netanyahu government, here the pattern is different.Footnote 11 Among respondents who expressed satisfaction with the state’s handling of the pandemic, there was no decrease in national attachment (F(1.98, 1182.14)=.314 p=.729), but there was a decrease in national chauvinism [F(1.46, 871.14)=62.13 p=.000)]. Among respondents who expressed dissatisfaction with the state’s handling of the pandemic, there was a decrease in both national attachment [F(1.96, 1172.92)=.14.25 p=.000] and chauvinism [F(1.47, 879.41)=81.71 p=.000].
Overall, the findings indicate a reduction in national identity during the pandemic that crosses the two Israeli political camps. Nevertheless, the design cannot be used to determine whether these changes in the levels of national identity are a consequence of the pandemic. The political crisis might have also had some influence, and, therefore, the possibility that the reduction in national identity is a consequence of the political crisis or, perhaps, other factors and not even the pandemic cannot be ruled out.
The Connections Between National Identity and Emotional Solidarity Over Time
The second part of the analysis explored whether national attachment and national chauvinism are related to emotional solidarity and how this relationship changes over time. For robust measurement and estimation, structural equation modeling (SEM) was used. The explained variable is emotional solidarity, and three separate models were constructed in order to evaluate the measures of national identity in T1, T2, and T3 distinct relations with emotional solidarity in T3. The models controlled for age, gender, religiosity, political attitudes, voting, satisfaction with the state’s handling of the pandemic, and the influence of the pandemic on personal well-being.Footnote 12 Table 3 presents the results of the models.
Table 3. Emotional solidarity across time
Notes: N=1197 Standardized coefficients.
Prior to discussing the key variables of national identity, a clear pattern also emerged from the control variables. Political affiliation and voting for the Netanyahu government, despite being central explanations shaping Israeli public attitudes across many issues, were found to be unrelated to emotional solidarity. In fact, even when simple correlations were analyzed, there was a very weak correlation between emotional solidarity and voting for Netanyahu government (r=.55, p=.069) or political affiliation (r=.063*). It should be emphasized, however, that this does not entirely rule out the potential effect of the political crisis on national identity, emotional solidarity, and their relations.
Across the three models, those who expressed greater satisfaction with the state’s handling of the pandemic were found less likely to express emotional solidarity. This finding might imply that when people believe that the state is taking care of the situation, they are less worried about those who are worse off. Gender and age were found to be positive predictors of emotional solidarity, with women showing more solidarity than men and older people more solidarity than younger. The age factor is not surprising due to the pandemic’s greater impact on older people. Gender is also not surprising given that higher levels of solidarity during the pandemic were found among women than among men (Kaup et al. Reference Kaup, Schweda, Krakowczyk, Dinse, Skoda, Teufel and Bäuerle2022).
The strongest predictor of emotional solidarity was the influence of the pandemic on personal well-being. People who testified to the pandemic influencing their economic situation or quality of life were more concerned for those who were adversely affected by the pandemic. This finding might imply that people who are affected by a crisis have more solidarity with others who are also affected by the crisis and do not focus on their own condition. This, in turn, might suggest that national identity in times of hardship reinforces care among those who are affected. This interpretation, however, require additional research.
Of most interest in the analysis are the measures of national identity. The results of the models reflect changes in the relations between the variables within the respondents. Differences in the coefficients can thus be understood as differences in the magnitude of the relations between the variables. National attachment became a stronger explanation of solidarity from its measurement in T1 (.126***) to T2 (.176***) to T3 (.218***). National chauvinism, on the other hand, was found related to emotional solidarity in its measurement in T1 (.123***) but not in T2 (.048) or T3 (.029). This shows a clear pattern: national attachment became a stronger predictor of emotional solidarity across time, but chauvinism became less important. This pattern might be caused by the pandemic and the mobilization of national attachment to ensure solidarity. Yet, given other events parallel to the pandemic such as the political crisis, other factors causing these changes cannot be ruled out. The study design cannot be used to determine the relative weight of these factors. This emphasizes the need to consider the distinction of political trust from national identity, as is further discussed in the conclusions.
Conclusions
In a recent paper entitled “Why Nationalism Works,” Andreas Wimmer (Reference Wimmer2019, 4) explained the establishment of the welfare state as one of the benefits of nationalism since: “A sense of mutual obligation and shared political destiny popularized the idea that members of the nation – even perfect strangers – should support one another in times of hardship.” The COVID-19 pandemic has been a time of hardship; in fact, scholars have identified it as a critical juncture (Twigg Reference Twigg2020). It is critical because it was a situation of extreme challenges and uncertainty that threatens both self and society. It is a juncture because it disturbed the narratives and institutionalized routines of societies and states and engendered practices that may result in lasting, fundamental change. The influence of the pandemic on nationalism, however, is hard to predict (Mylonas and Whalley Reference Mylonas and Whalley2022).
The influence of the COVID-19 pandemic on nationalism has, thus far, been estimated at the levels of political discourse (Berrocal et al. Reference Berrocal, Kranert, Attolino, Santos, Santamaria and Henaku2021; Gülseven Reference Gülseven2021), attitudes, and behaviors (Chan et al. Reference Chan, Wang, Zuo, Chiu, Liu, Yiu and Hong2021; Lim and Prakash Reference Lim and Prakash2021; Sibley et al. Reference Sibley, Greaves, Satherley, Wilson, Overall, Carol and Milojev2020; Vignoles et al. Reference Vignoles, Jaser, Taylor and Ntontis2021). In line with earlier studies exploring the national identity argument (Breidahl, Holtug and Kongshøj Reference Breidahl, Holtug and Kongshøj2018; Gustavsson Reference Gustavsson, Gustavsson and Miller2020; Gustavsson and Miller Reference Gustavsson and Miller2020; Gustavsson and Stendahl Reference Gustavsson and Stendahl2020; Kongshøj Reference Kongshøj2019; Johnston et al. Reference Johnston, Banting, Kymlicka and Soroka2010; Miller and Ali Reference Miller and Ali2014; Reeskens and Wright Reference Reeskens and Wright2013), results from studies about the influence of the pandemic are inconclusive and contradictory. This article used the pandemic as an opportunity to focus on the national identity argument. Unlike previous assessments of this argument, which demonstrated only the correlation between national identity and solidarity, this article used longitudinal survey designs. Taking Israel as a test case, there are two key findings. First, there is a clear reduction in the levels of national attachment and national chauvinism over time. Second, national attachment’s positive connection with solidarity grew stronger over time, while national chauvinism’s connection with solidarity grew weaker and insignificant over time.
The article’s findings provide some insights on the national identity argument. First, national attachment is found to be positively associated with solidarity. This link holds firm even when national chauvinism and other explanations are controlled. This finding by itself is not new as it was found in previous studies (Gustavsson Reference Gustavsson, Gustavsson and Miller2020); what is new, however, is that this article’s longitudinal design indicates that this connection became stronger in a time of hardship. This finding supports Tamir’s (Reference Tamir2020) account of liberal nationalism building on the hardship experienced during the pandemic and her claim that national identity contributes to solidarity. This article shows that national attachment is indeed a source of mobilization for solidarity.
Second, the decrease in national attachment and national chauvinism over time does not, however, provide support for the national identity argument. It seems that national identity does not intensify in the case of crises not caused by an external enemy. This also points to the question of the stability of national identity over time and in the face of large-scale events. Despite changes in a society’s demographic composition and the intensity of globalization, national attachment (Gustavsson and Miller Reference Gustavsson and Miller2020) and national chauvinism (Coenders, Lubbers, and Scheepers Reference Coenders, Lubbers and Scheepers2020) have not changed much among the public over the last two decades. Collins (Reference Collins2012) pointed out that large-scale events that mobilize national identity are characterized by an intense initial three months and a subsequent gradual return to normal internal divisions after about six months.
Third, it is important to emphasize that the reduction in national attachment and national chauvinism, as well as changes in the connection of national attachment and chauvinism with emotional solidarity, might not be the consequence of the pandemic. The political crisis in Israel might, indeed, have caused these changes. This possibility calls for additional attention to the nexus between national identity and political trust. The idea that national identity intensifies social and political trust as a precondition for solidarity is an important component of the national identity argument (Lenard, Miller, and Uslaner Reference Lenard, Miller, Uslaner and Uslaner2018; Miller and Ali Reference Miller and Ali2014). National identity increases trust in state authorities which, in turn, supports their ability to allocate resources. Indeed, findings from the United States and the Netherlands have shown that national attachment and patriotism are positively related to political trust (Gustavsson and Stendahl Reference Gustavsson and Stendahl2020). However, it should be remembered that political trust has long been considered an aspect of system support, of which national identity is also regarded as a key component (Easton Reference Easton1965; Norris Reference Norris2011). Given that both are considered aspects of system support, further distinction between them is required, especially given the implications of the pandemic for system support.
The COVID-19 pandemic might well be a critical juncture that has changed national identity more than any previous large-scale events. There is indeed evidence that it has changed levels of national identity. An increase in national attachment and patriotism was found in both New Zealand (Sibley et al. Reference Sibley, Greaves, Satherley, Wilson, Overall, Carol and Milojev2020) and South Korea (Lim and Prakash Reference Lim and Prakash2021), while in the United States there seems to have been a significant decrease in national attachment over time (Chan et al. Reference Chan, Wang, Zuo, Chiu, Liu, Yiu and Hong2021). This implies that in times of hardship there is not always a flag to rally around: in other words, nationalism cannot always be mobilized. The reduction of national identity in the United States was possibly the consequence of both the American government’s poor handling of the pandemic and ongoing political polarization (Chan et al. Reference Chan, Wang, Zuo, Chiu, Liu, Yiu and Hong2021). In Israel too, the political crisis might have exerted a similar influence. Israel, unlike the United States, used aggressive measures to address the pandemic and served as a pioneer in vaccinating its public; nonetheless, the combination of the pandemic and the political crisis might have influenced the public’s overall judgment of the political elite. The finding that there was no decrease in national attachment among respondents who expressed satisfaction with the state’s handling of the pandemic might suggest this possibility. More importantly, the findings from both Israel and the United States imply the centrality of the political elite’s ability to mobilize nationalism to promote solidarity. The question of whether national identity can promote solidarity when political elites are deeply polarized and perceived as malfunctioning should be examined in future studies on the national identity argument.
Despite the unique opportunity provided by the pandemic, it is important to acknowledge the study’s limitations. Most notable is the use of surveys, which limit the concept of national identity to measurable indexes. While such indexes are essential for observing changes and causal relations with solidarity, they may overlook many nuances. Future studies might attempt to follow a qualitative approach that could provide rich understanding of national identity and solidarity. In addition, the indexes themselves are subject to debate. As Miller and Ali (Reference Miller and Ali2014) showed, the different ways of measuring national identity might account for the contradictory nature of previous results. This article followed the social psychology framework and adopted Gustavsson’s (Reference Gustavsson, Gustavsson and Miller2020) suggestion to examine national attachment while controlling for national chauvinism. When the psychological framework is used, one should remember that other psychological factors must be considered: for example, authoritarianism, which is linked to national chauvinism (Huddy and del Ponte Reference Huddy, del Ponte, Gustavsson and Miller2020).
A second limitation is the fact that only one country was examined here. Israel is an extreme case of nationalism, and national attachment would therefore be expected to have a profound impact on solidarity. A third and related limitation is that the study only focused on Jewish Israelis, in line with most studies of national identity that focused on the majority group. Yet, as Galle and Fleischmann (Reference Galle and Fleischmann2020) claimed, a sampling of both majorities and minorities is required in order to test the national identity argument. Finally, during the time of the study, Israel was embroiled in a deep and intense political crisis that might have influenced national identity regardless of the pandemic. While the analysis does address the question of polarization, it is reasonable to assume that part of the reduction in national identity was related to this political crisis.
A generalization of this study must consider that the values attributed to national identity differ between countries (Pehrson, Vignoles, and Brown Reference Pehrson, Vignoles and Brown2009). As context is one of the reasons for the contradictory results from previous studies (Miller and Ali Reference Miller and Ali2014), a comparative longitudinal or experimental design might provide insights that can be generalized beyond one specific case.
This article does not provide simple evidence that either supports or rejects the national identity argument. However, it shows that, in contrast to the assumptions made in political theory or the insights derived from social psychology, the ways in which national identity is linked to solidarity are multifaceted and conditional. The COVID-19 crisis provided an excellent opportunity to explore these insights. Even though “it is hard to predict the weather from inside the storm” (Mylonas and Whalley Reference Mylonas and Whalley2022), it nonetheless enabled a critical investigation of national identity and solidarity. It also emphasized the need to examine whether a functioning political establishment is required in order to forge links between national identity and solidarity.
Acknowledgements
I would like to thank to Danny Kaplan, Hizky Shoham, and the participants of the “Theorizing National Solidarity: Mechanisms, Models and Paradoxes” 2022 workshop for their comments and helpful suggestions.
Financial Support
This research was supported by the Israel Science Foundation (Grant No. 379/17).
Disclosures
None.
Appendix 1. Scales Construction
The emotional solidarity scale was constructed for this study, and its items refer to groups that were vulnerable during the COVID-19 pandemic: elderly people; people with disabilities; the unemployed; people at risk that cannot return to their regular life. While the wording of the question and the items does not explicitly state that the vulnerable groups are co-nationals, in the context of the survey and from other similar studies in Israel, it can be understood that the scope of the respondents’ awareness refers to other Jewish Israelis and not to Palestinians with Israeli citizenship. This is based on implicit prototypes, as can also be found in other countries such as the United States (e.g., Yogeeswaran and Dasgupta Reference Yogeeswaran and Dasgupta2010). The scale was constructed using the simple means of the items and assuming that the loadings of the different items and the internal reliability supported this indexing (factor loadings .74 to .87, α=.84). In the second part of the analysis, the scale was estimated as part of the measurement model. The fit of the measurement model was very good (e.g. T3: SRMR=.031; RMSEA=.044; Pclose=.844; CFI=.986).
The variables for national attachment (“Israel is an important part of my identity”; “Israeli identity is more important to me than other types of identity”; “It is not important for me to see myself as Israeli [R]”) and national chauvinism (“The world would be a better place if people from other countries were more like Israelis”; “Generally speaking, Israel is a better country than most other countries”; “In comparison with other nations, Israel is very moral nation”; “Other countries can learn a lot from Israel”) were measured based on items adopted from Roccas et al. (Reference Roccas, Klar and Liviatan2006) and Davidov (Reference Davidov2009). An exploratory factor analysis found the scales to be clearly distinct across the three time points. To further validate the measures, the divergent validity between the variables was examined by comparing models in which all of the items were loaded on one or two factors for T1. While the single factor model was far from acceptable (CFI < .8), the two-factor model was acceptable (CFI=.970; SRMR=.383; REMSA=.079). Despite their distinctiveness, they are positively correlated (T1: r=.47; T2: r=.52; T3: r=.24). In the first part of the analysis, the scale was constructed by using the simple means of the items and assuming that the loadings of different items and the internal reliability supported this indexing (national attachment T1: factor loadings .65 to .90, α=.75; national attachment T2: factor loadings .75 to .90, α=.80; national attachment T3: factor loadings .72 to .90, α=.76; national chauvinism T1: factor loadings .84 to .88, α=.88; national chauvinism T2: factor loadings .86 to .90, α=.90; national chauvinism T3: factor loadings .83 to .91, α=.91). In the second part of the analysis, the scales were estimated as part of the measurement model.
Open access
