ABSTRACT
Why are people in Central and Eastern Europe more hesitant towards COVID-19 vaccination and more prone to believe in COVID-19 related conspiracy theories than other Europeans? The article claims that the spread of COVID-19 conspiracy beliefs in the post-communist region might be fostered by communist nostalgia. Drawing on the survey data from Lithuania, I show that communist nostalgia is one of the best predictors of COVID-19 conspiracy beliefs, controlling for other related factors such as populist attitudes, trust in political institutions, confidence in media and scientists and pro-Western attitudes. The paper claims that communist nostalgia in Central and Eastern Europe is conducive to conspiracy beliefs in a similar vein as nostalgic narratives employed by populist radical right in Western countries.
Introduction
Global coronavirus pandemic was followed by the global spread of COVID-19 related conspiracy theories which are widely believed in different countries (Freeman et al. 2022; Uscinski et al. 2020; Miller 2020). Previous works demonstrate that conspiracy beliefs have consequences on COVID-19 related behaviour: those who believe in COVID-19 conspiracies are less likely to adopt recommendations of health specialists and are more reluctant to get vaccinated (Allington et al. 2021; Stecula and Pickup 2021; Roozenbeek et al. 2020; Douglas 2021). Therefore, researchers conclude that the spread of COVID-19 conspiracy theories poses a serious challenge in fighting the pandemic (Romer and Jamieson 2020, Stecula and Pickup 2021).
While the evidence on the determinants explaining COVID-19 conspiracy beliefs is quickly growing, most research is primarily focused on universal psychological factors such as personality features, conspiratorial thinking, open-mindedness, the role of insecurity and political trust (Larsen et al. 2021, Freeman et al. 2022, Šrol et al. 2021, Alper et al. 2020, Sallam et al. 2020, Stoica and Umbres 2021). The factors related to specific cultural and political environment in different regions of the world are much less studied. Most research is focused on the United Sates, where COVID-19 conspiracy beliefs are proved to be related to conservativism, support for Donald Trump and the Republican party, and the use of conservative media (Stecula and Pickup 2021, Sturm and Albrecht 2021, Romer and Jamieson 2020, Calvillo et al. 2020). Moreover, findings from several different countries demonstrate that COVID-19 conspiracy beliefs are fostered by populist attitudes and far-right ideology (Eberl et al. 2021, Stecula and Pickup 2021).
Not much previous work is available on political and cultural determinants of COVID-19 conspiracy beliefs in Europe, particularly in Central and Eastern Europe, where the prevalence of COVID-19 related theories is high and willingness to take a COVID-19 vaccine is lower than in Western countries (EB 2021, Hajdu et al. 2021). At the beginning of 2022, the vaccination level in Central and Eastern Europe was still well below the EU average. For example, in January 2022, the percentage of fully vaccinated adults reached only 34.1 in Bulgaria and 49 in Romania, while the EU average was 77.6 percent (Statista 2022). What factors account for the vaccination hesitancy and the spread of COVID-19 conspiracy theories in this region, apart from universal psychological explanations?
Achimescu et al. (2021) demonstrate that in Romania the belief in COVID-19 conspiracy narratives and noncompliance with public health guidance is positively related to pro-Russian attitudes and correlates negatively to trust in NATO and the EU. These findings suggest that prevalence of COVID-19 conspiracies among Eastern Europeans might be related to higher vulnerability to Russian propaganda narratives related to COVID-19 such as spreading distrust in Western vaccines, implying that coronavirus originated in an American laboratory, etc. (Moy and Gradon 2020, Sukhankin 2020, Dubov et al. 2021). Nevertheless, this argument alone cannot explain the susceptibility of Eastern Europeans to COVID-19 conspiracies, as some of these countries are particularly anti-Russian and pro-Western.
This article focuses on the case of Lithuania, where pro-Western attitudes are dominating in the society and trust in Russia is very low, in contrast to Romania, Bulgaria or Slovakia where trust in Russia is quite widespread (see e.g. Hajdu et al. 2021). In Lithuania, the support for the EU is constantly one of the highest in Europe (see e.g. EB 2021) and friendliness to Russia is very low; nevertheless, COVID-19 related conspiracy narratives are quite pervasive (Hajdu et al. 2021).
The article aims to explore the determinants of the COVID-19 conspiracy beliefs in Lithuania, drawing on the representative survey carried out in January–February 2021. The paper argues that susceptibility to COVID-19 conspiracy theories is fostered by communist nostalgia which is prevalent in Lithuania as well as in other post-communist countries. The paper demonstrates that communist nostalgia is a strong predictor of the beliefs in COVID-19 conspiracy theories, even when controlling for such factors as partisanship, anti-Western orientations, political trust and populist attitudes.
Explaining conspiracy beliefs: mistrusting mind-set, populism and nostalgia
Conspiracy theories are commonly defined as explanations of events or processes based on beliefs in existence of a small secret group of people acting against the common good for selfish purposes (Uscinski et al. 2016). Some authors define conspiracy theories as false beliefs or ‘mistruths’, while others claim that they should not be false by definition; however, researchers are usually interested in the spread of ‘false conspiracy theories’, i.e. conspiratorial explanations that are unsupported by evidence (Freeman and Bentall 2017).
Conspiracy theories are regarded as a type of generalised political attitude (Imhoff and Bruder 2014) or a specific type of belief systems (Freeman and Bentall 2017). It has been observed that people who believe in some conspiracy theories tend to believe in others, even if they contradict each other (Goertzel 1994; Miller 2020; Swami et al. 2011). Therefore, it is claimed that that people differ in their ‘conspiracy mentality’ or ‘conspiratorial mind-set’, i.e. in their general propensity to explain events through conspiracy theories (Van Prooijen 2018). Conspiratorial thinking is found to be more common among people with a wide spectrum of negative life circumstances, e.g. disruptive parental experiences during childhood, social isolation and marginalisation; they usually have lower self-esteem, higher level of unhappiness, perception of lower social status, powerlessness, and distrust of authority (Freeman et al. 2022, Freeman and Bentall 2017).
Previous works demonstrate that conspiracy theories are most prevalent during crises, in times of great pressure to satisfy epistemic, existential or social needs (Achimescu et al. 2021, Larsen et al. 2021, Douglas et al.2019). In such circumstances, beliefs in conspiracies have multiple short-term benefits, e.g. reduction of uncertainty, increase of self-esteem, confirmation of distrust beliefs and access to the networks of like-minded people (Freeman and Bentall 2017). While psychologists find association between conspiracy beliefs and paranoia (Freeman and Bentall 2017), other findings suggest that conspiracy theories are endorsed by large numbers of normal, non-pathological people (Van Proijeen 2018). Therefore, beliefs in conspiracy theories cannot be attributed exclusively to psychological determinants. These beliefs are also determined by cultural, social and political factors, such as political predispositions, partisanship structure or patterns of media consumption (Uscinski et al. 2016). Therefore, it is vital to know what political predispositions are conducive to the spread of conspiracy beliefs in a given country or region.
A growing body of literature demonstrates a clear link between conspiratorial thinking and populist attitudes (Stecula and Pickup 2021, Castanho Silva et al. 2020, Eberl et al. 2021). Some authors even claim that populism might be regarded as political manifestation of conspiratorial thinking or that populism is a key political mentality underlying the belief in conspiracy theories (Stecula and Pickup 2021, Van Prooijen 2018). Arguments that relate populism and conspiratorial mind-set are useful for revealing the link between communist nostalgia and conspiration theories, therefore, they will be discussed here in more detail.
Populism is widely regarded as a set of ideas depicting society as divided between ‘corrupt elite’ and ‘pure ordinary people’ (Mudde and Kaltwasser 2017, 2018, Canovan 1999, Stanley 2008). The people are seen as honest and moral, while elite is regarded as acting against the will of people. Van Prooijen (2018) describes two main mechanisms connecting populism and conspiracy theories. First, as populism is based on a deep-rooted distrust of political, economic and cultural establishment, it is a small step to assume that these elites pursue malevolent aims by forming conspiracies. In fact, both populist attitudes and conspiratorial thinking are correlated with distrust not only of political institutions, but also of experts and mainstream media. Second, conspiracy theories are more prevalent among people who feel powerless and voiceless, hence, they classify themselves as being part of ordinary powerless ‘people’. These two mechanisms are closely interlinked, as mistrustful mind-set is a reaction to the perception of vulnerability and real or subjective marginalisation (Freeman et al. 2022). Vulnerability, marginalisation and distrust of elites are also the attributes of the feelings of communist nostalgia, therefore, the arguments of van Prooijen (2018) can be easily be expanded to the analysis of the link between conspiratorial thinking and ‘red’ nostalgia in post-communist countries.
Communist nostalgia is a well-known and widely studied cultural phenomenon (Ghodsee 2004, Ekman and Linde 2005, Volčič 2007, Velikonja 2009, Todorova and Gille 2010, Koleva 2011). While much research is focused on ‘Ostalgia’ in Eastern Germany and ‘Yugonostalgia’ in the countries of former Yugoslavia, the phenomenon of socialist or ‘red’ nostalgia is common to all countries of the socialist block, including Lithuania (Čepaitienė 2007, Klumbytė 2008, 2010, Šutinienė 2013, Ramonaitė 2013). Researchers define nostalgia as ‘a mourning for the irreversible loss of the past’ (Velikonja 2009). It is a collective attitude or a state of mind that is associated with a feeling of loss in a period of radical changes of post-communist transformation (Koleva 2011). While nostalgia refers to the past, it is related to discontent with the present situation. Researchers agree that nostalgia refers not so much to past realities, but mainly to past dreams, past visions and expectations that had not been fulfilled (Velikonja 2009). It is a ‘backward looking utopia, i.e. a longing for an idealised past pointing to the deficits of the present’ (Koleva 2011).
Communist nostalgia is an outcome of painful process of post-communist transformation experienced by virtually all countries of Central and Eastern Europe. Ghodsee and Orenstein (2021) reveal the devastating social outcomes of the transformation that are still underestimated and understudied: GDP, consumption levels and life expectancy fell drastically for more than a decade in many post-communist countries, unprecedented mortality and emigration caused massive loss of population in these countries. Anthropologists reveal various survival strategies taken by the citizens of post-communist countries to survive the conditions of chaos and recession, such as the use of informal networks, the return to subsistence farming, earnings in the West (Humphrey 2002, Morris and Poelse 2014), as well as their emotions: desperation, powerlessness, pessimism, anger, anxiety about the future, feelings of alienation and injustice (Burawoy and Verdery 1999, Ramonaitė 2007, Klumbytė 2004). Reports of ethnographic fieldwork in different post-communist countries are full of the sore metaphors such as ‘ex-people’, ‘thrown away people’, ‘sacrificed generation’, expressing the feelings of being unneeded and worthless (Velikonja 2009, Ghodsee and Orenstein 2021). The political outcome of the cultural trauma of the transformation was a dramatic raise of mistrust in political institutions and political apathy (Sztompka 2004). All these symptoms of the consequences of transformation are precisely those that were identified above as particularly favourable to the spread of conspiracy theories: the feeling of powerlessness, marginalisation, distrust and anger towards authorities.
An important feature of the post-communist transformation in Central and Eastern Europe is a notable gap between the popular views about the transition process and the mainstream elite narrative presenting the transformation as a success story. As Ghodsee and Orenstein (2021, p. 191) notice, the social drama of the transition is not publicly acknowledged in most of the countries of Central and Eastern Europe because damaging impacts of transition occurred at the same time when smaller but the more visible – educated, and urban – part of the societies experienced new opportunities and growing prosperity. As long as public discourse is dominated by anti-nostalgic strategy of dealing with the past (Velikonja 2009), red nostalgia becomes a tool for resistance to the dominant narrative (see e.g. Pan 2013). This is another mechanism linking communist nostalgia with conspiratorial thinking.
Gabriel (2017, p. 220) notices that nostalgic stories and conspiracy theories have a common feature – they both usually work as counter-narratives, i.e. an attempt of marginalised and disempowered groups to challenge the dominant narrative: ‘Like nostalgic narratives, conspiracy theories inhabit a narrative ecology in which they generally emerge as counter-narratives in opposition to the ecology's master narratives.’ Therefore, nostalgic narratives and conspiracy theories often combine and reinforce each other.
Freeman et al. (2022) also notice that conspiratorial thinking is a worldview marked by ‘antipathy to official or mainstream accounts or to those in higher status positions’. This feature might be regarded as a third mechanism (besides two other mechanisms offered by van Prooijen 2018 described above) linking not only conspiratorial thinking and communist nostalgia but also conspiracy theories and populism, since right-wing populism currently rife in Western countries also exhibits some features of nostalgia (Steenvoorden and Harteveld 2018, Gest et al. 2018, Browning 2019, Elgenius and Rydgren 2019). Gest et al. (2018) claim that support for the radical right is the outcome of nostalgic deprivation – ‘the discrepancy between individuals’ understandings of their current status and their perceptions about their past.’
The arguments presented above suggest that communist nostalgia might be a region-specific factor, explaining high prevalence of COVID-19 related conspiration theories in post-communist countries, independently from or in addition to pro-Russian and anti-Western attitudes. In this article, I test the relationship between communist nostalgia and COVID-19 conspiracy beliefs on an individual level. Using the data from Lithuania, I analyse whether Lithuanians who feel communist nostalgia are more inclined to believe in COVID-19-related conspiracy theories and, if so, whether communist nostalgia has an autonomous effect on COVID-19-related conspiracy beliefs when other related factors such as populism, distrust of authorities, experts, media, anti-Western attitudes and exposure to Russian media are controlled.
Data and methods
The data comes from a representative post-election survey1 carried out in 21 January–21 February 2021 by Norstat LT, one of leading data-collection companies in Lithuania. The survey is a part of the Lithuanian National Election Study 2020. The probability sample of 1830 adult Lithuanians included the residents of Lithuania aged 18 and older with the Lithuanian citizenship and the right to vote. As the survey was carried out during a lockdown2, a mixed-mode (internet and telephone) research design was used. Respondents were recruited via telephone, applying random digital dialling technique, in compliance with the ICC/ESOMAR code of data-collection and EU General Data Protection Regulation. Respondents were asked to fill in an online survey; if a respondent was not able or did not want to fill in an online survey but agreed to participate in the survey, he or she was asked to answer the shortened version of the survey during the same phone call. Telephone interviews, however, comprised only 2.7 percent of the responses. Overall response rate was 26.8 percent. The data was weighted by gender, age, education, residence type and political variables (specifically, turning out to vote and electoral choice in the 2020 parliamentary elections). The data and replication package used in the analysis are provided in Ramonaitė (2022).
Dependent variable
The COVID-19 conspiracy scale was adopted from Allington et al. (2021). The scale measures an agreement on a 5-point scale (from strongly agree to strongly disagree) with the following five statements: ‘Coronavirus was probably created in a laboratory’, ‘The symptoms that most people blame on coronavirus appear to be linked to 5G network radiation’, ‘There is no hard evidence that coronavirus really exists’, ‘The number of people reported as dying from coronavirus is being deliberately exaggerated by the authorities’, ‘The current pandemic is part of a global effort to force everyone to be vaccinated whether they want to or not’.
The index of disbelief in COVID-19 conspiracy theories was created using an arithmetic mean of the five items3 using pairwise deletion of missing data (Cronbach alpha = 0,82). The greater values of an index show less support for the COVID-19 conspiracy theories. Descriptive statistics of all variables used in the analysis are provided in Table A1 in the supplemental material.
Measuring communist nostalgia
While there is much academic literature on ‘red nostalgia’, most research on the topic is carried out in the fields of anthropology and culture studies using qualitative research methods. In the field of quantitative research of political behaviour and political attitudes, the topic is somewhat under-researched (see, however, Mishler and Rose 1996, 2002, Ekman and Linde (2005), Gherghina 2010, Duvold and Ekman 2016 and other works based on the New Europe Barometer data), and there is no single established measurement of the phenomenon. It is quite common to refer to any positive reference to the past communist regime as communist nostalgia. For example, Ekman and Linde (2005) and Gherghina (2010) measure communist nostalgia as an approval of the return to the communist rule; White (2010) defines soviet nostalgia as a regret of the demise of the USSR, Duvold and Ekman (2016) uses the ranking of the communist regime on a 200-point scale. While these works provide useful insights into the attitudes of post-communist citizens towards the communist past, these measurements are not fully in line with theoretical works on the communist nostalgia. As was described in the theoretical part of the paper, communist nostalgia is not a real wish to return back to the socialist past; it is rather a retrospective utopia, serving as a critique to the present. In fact, evidence from the countries of Central and Eastern Europe, including Lithuania, reveal that post-communist citizens tend to evaluate their social past favourably, but would not like those times to return (Ekman and Linde 2005, Velikonja 2009). Therefore, the measure of communist nostalgia should capture the positive evaluation of the communist past in relation to the present, rather than the support for communist ideas or a dream of the communist regime to return.
In the analysis of this article, I employ a measure of communist nostalgia that performed well in numerous works on political attitudes and political behaviour in Lithuania (Ramonaitė 2014, 2018, 2020, Jastramskis 2020). It measures an agreement on a 5-point scale with the statement that ‘Life in the Soviet Era was better than it is now’. The advantage of the measure is that it taps the relation of a respondent to the soviet past in contrast to the present.4 In the regression models, I first tested this item as a categorical variable. In the final models it is included as a numerical variable for ease of interpretation, as distance between the coefficients of different categories showed a roughly linear relationship (see the original model in the supplemental material).
Control variables
As control variables I include, first of all, those socio-demographic factors that were identified as significant antecedents of COVID-19 conspiracy beliefs in previous studies (for a comprehensive review see van Mulukom et al. 2020): age, gender, ethnicity, education, place of residence (rural vs. urban settlement) and religiosity (measured as a frequency of church attendance). I also control for the political party preference because soviet nostalgia is one of the most significant determinants of electoral choice in Lithuania (Ramonaitė 2014, 2018, Jastramskis 2020). It is measured as a party choice in the 2020 parliamentary elections in Lithuania.
Next, I include the measure of trust in the European Union (EU), since the work of Achimescu et al. (2021) demonstrates that trusting Russia and distrusting Western actors increases the propensity to believe in COVID-19 conspiracy narratives in post-communist Romania. Trust in the EU is measured on a 0–10 scale where 0 means ‘do not trust at all’ and 10 means ‘complete trust’. While we do not have a direct measure of trust in Russia, the potential impact of exposure to Russian narratives is taped by two other measures in our analysis: the use of Russian media and the rate of (dis)like of Vladimir Putin, the leader and the symbol of the authoritarian regime in Russia. The use of Russian media was captured by asking how often a respondent used Russian media to follow news in the last 3 months. Like – dislike of Putin was measured on a 0–10 scale where ‘0’ means ‘strongly dislike’ and 10 means ‘strongly like’.
Finally, I include those attitudinal factors that determine the belief in COVID-19 conspiracies in other countries and that might be intercorrelated with communist nostalgia: political trust, trust in media, trust in scientists and populist attitudes. Political trust index is composed of three items (Cronbach alpha 0,9): trust in parliament, trust in government and trust in political parties. It is measured in a 0–10 scale, where higher values show higher trust. Trust in media and trust in scientists are measured on a 0–10 scale where 0 means ‘do not trust at all’ and 10 means ‘complete trust’. To measure the populist attitudes, I used the items of the populist scale included in the CSES Module 5 (see Hobolt et al. 2016, Castanho Silva et al.2020).
Results
Table 1 shows the results of four nested models on the index of disbelief in COVID-19 conspiracy theories (CTs).5 Model 1 includes only socio-demographic variables and party preference. In the Model 2, I add exposure to Russian media, trust in the EU and like-dislike of Putin to account for the hypothesis of Achimescu et al. (2021) on the role of openness to anti-Western narratives. In Model 3, I introduce the variable of soviet nostalgia. Finally, in Model 4, I add other attitudinal variables to test if the relationship between the (dis)belief in COVID-19 conspiracies and soviet nostalgia holds when controlling for other attitudes that correlate with soviet nostalgia (for correlations among the attitudinal variables see the supplemental material), including political trust and populism.
. | Model 1 . | Model 2 . | Model 3 . | Model 4 . |
---|---|---|---|---|
Age (in years) | −0.040*** | −0.017* | −0.007 | −0.011 |
(0.008) | (0.008) | (0.007) | (0.007) | |
Age squared | 0.0004*** | 0.0002* | 0.0001 | 0.0001 |
(0.0001) | (0.0001) | (0.0001) | (0.0001) | |
Gender (1=female) | −0.280*** | −0.239*** | −0.157*** | −0.191*** |
(0.046) | (0.044) | (0.042) | (0.039) | |
Ethnicity (1=non-Lithuanian) | −0.088 | 0.096 | 0.080 | 0.077 |
(0.093) | (0.094) | (0.089) | (0.083) | |
Religiosity: sometimes (vs. often) | 0.160 | 0.130 | 0.109 | 0.067 |
(0.098) | (0.093) | (0.087) | (0.082) | |
Religiosity: never (vs. often) | 0.143 | 0.177 | 0.137 | 0.155 |
(0.109) | (0.103) | (0.097) | (0.091) | |
Religiosity: NA (vs. often) | −0.089 | −0.092 | −0.066 | −0.114 |
(0.143) | (0.134) | (0.127) | (0.119) | |
Education (1=University) | 0.517*** | 0.459*** | 0.354*** | 0.294*** |
(0.050) | (0.047) | (0.045) | (0.043) | |
Settlement type (1=urban) | 0.225*** | 0.218*** | 0.189*** | 0.131** |
(0.052) | (0.049) | (0.046) | (0.043) | |
Party: TS-LKD (vs. FP) | 0.124 | 0.070 | 0.048 | 0.056 |
(0.113) | (0.106) | (0.100) | (0.094) | |
Party: LRLS (vs. FP) | −0.021 | 0.026 | 0.056 | 0.117 |
(0.149) | (0.140) | (0.133) | (0.124) | |
Party: LVŽS (vs. FP) | −0.505*** | −0.365** | −0.268* | −0.076 |
(0.127) | (0.120) | (0.113) | (0.106) | |
Party: DP (vs. FP) | −0.919*** | −0.697*** | −0.393** | −0.207 |
(0.149) | (0.141) | (0.135) | (0.127) | |
Party: LSDP (vs. FP) | −0.204 | −0.082 | 0.032 | 0.045 |
(0.143) | (0.134) | (0.127) | (0.119) | |
Party: other (vs. FP) | −0.485*** | −0.332** | −0.224* | −0.085 |
(0.119) | (0.113) | (0.107) | (0.100) | |
Party: non-voters (vs. FP) | −0.681*** | −0.483*** | −0.340*** | −0.142 |
(0.102) | (0.097) | (0.092) | (0.087) | |
Trust in the EU (0-10) | 0.103*** | 0.080*** | 0.023* | |
(0.008) | (0.008) | (0.009) | ||
The use of Russian media: sometimes (vs. often) | −0.0001 | −0.084 | −0.136 | |
(0.082) | (0.078) | (0.073) | ||
The use of Russian media: never (vs. often) | 0.235*** | 0.165** | 0.090 | |
(0.063) | (0.060) | (0.056) | ||
The use of Russian media: NA (vs. often) | 0.265 | 0.265 | 0.365 | |
(0.212) | (0.200) | (0.187) | ||
Like – dislike Vladimir Putin | (0-10) | −0.021 | 0.013 | 0.011 |
(0.011) | (0.011) | (0.010) | ||
Communist nostalgia (1-5) | −0.287*** | −0.199*** | ||
(0.021) | (0.021) | |||
Political trust (0-10) | −0.007 | |||
(0.013) | ||||
Trust in media (0-10) | 0.025** | |||
(0.009) | ||||
Trust in scientists (0-10) | 0.049*** | |||
(0.010) | ||||
Populist attitudes (1-5) | −0.422*** | |||
(0.037) | ||||
Constant | 4.708*** | 3.315*** | 3.808*** | 5.000*** |
(0.218) | (0.235) | (0.225) | (0.269) | |
Observations | 1,529 | 1,529 | 1,529 | 1,529 |
Adjusted R2 | 0.246 | 0.336 | 0.407 | 0.485 |
Residual Std. Error | 0.847 (df = 1512) | 0.795 (df = 1507) | 0.751 (df = 1506) | 0.700 (df = 1502) |
F Statistic | 32.201*** (df = 16; 1512) | 37.793*** (df = 21; 1507) | 48.581*** (df = 22; 1506) | 56.243*** (df = 26; 1502) |
. | Model 1 . | Model 2 . | Model 3 . | Model 4 . |
---|---|---|---|---|
Age (in years) | −0.040*** | −0.017* | −0.007 | −0.011 |
(0.008) | (0.008) | (0.007) | (0.007) | |
Age squared | 0.0004*** | 0.0002* | 0.0001 | 0.0001 |
(0.0001) | (0.0001) | (0.0001) | (0.0001) | |
Gender (1=female) | −0.280*** | −0.239*** | −0.157*** | −0.191*** |
(0.046) | (0.044) | (0.042) | (0.039) | |
Ethnicity (1=non-Lithuanian) | −0.088 | 0.096 | 0.080 | 0.077 |
(0.093) | (0.094) | (0.089) | (0.083) | |
Religiosity: sometimes (vs. often) | 0.160 | 0.130 | 0.109 | 0.067 |
(0.098) | (0.093) | (0.087) | (0.082) | |
Religiosity: never (vs. often) | 0.143 | 0.177 | 0.137 | 0.155 |
(0.109) | (0.103) | (0.097) | (0.091) | |
Religiosity: NA (vs. often) | −0.089 | −0.092 | −0.066 | −0.114 |
(0.143) | (0.134) | (0.127) | (0.119) | |
Education (1=University) | 0.517*** | 0.459*** | 0.354*** | 0.294*** |
(0.050) | (0.047) | (0.045) | (0.043) | |
Settlement type (1=urban) | 0.225*** | 0.218*** | 0.189*** | 0.131** |
(0.052) | (0.049) | (0.046) | (0.043) | |
Party: TS-LKD (vs. FP) | 0.124 | 0.070 | 0.048 | 0.056 |
(0.113) | (0.106) | (0.100) | (0.094) | |
Party: LRLS (vs. FP) | −0.021 | 0.026 | 0.056 | 0.117 |
(0.149) | (0.140) | (0.133) | (0.124) | |
Party: LVŽS (vs. FP) | −0.505*** | −0.365** | −0.268* | −0.076 |
(0.127) | (0.120) | (0.113) | (0.106) | |
Party: DP (vs. FP) | −0.919*** | −0.697*** | −0.393** | −0.207 |
(0.149) | (0.141) | (0.135) | (0.127) | |
Party: LSDP (vs. FP) | −0.204 | −0.082 | 0.032 | 0.045 |
(0.143) | (0.134) | (0.127) | (0.119) | |
Party: other (vs. FP) | −0.485*** | −0.332** | −0.224* | −0.085 |
(0.119) | (0.113) | (0.107) | (0.100) | |
Party: non-voters (vs. FP) | −0.681*** | −0.483*** | −0.340*** | −0.142 |
(0.102) | (0.097) | (0.092) | (0.087) | |
Trust in the EU (0-10) | 0.103*** | 0.080*** | 0.023* | |
(0.008) | (0.008) | (0.009) | ||
The use of Russian media: sometimes (vs. often) | −0.0001 | −0.084 | −0.136 | |
(0.082) | (0.078) | (0.073) | ||
The use of Russian media: never (vs. often) | 0.235*** | 0.165** | 0.090 | |
(0.063) | (0.060) | (0.056) | ||
The use of Russian media: NA (vs. often) | 0.265 | 0.265 | 0.365 | |
(0.212) | (0.200) | (0.187) | ||
Like – dislike Vladimir Putin | (0-10) | −0.021 | 0.013 | 0.011 |
(0.011) | (0.011) | (0.010) | ||
Communist nostalgia (1-5) | −0.287*** | −0.199*** | ||
(0.021) | (0.021) | |||
Political trust (0-10) | −0.007 | |||
(0.013) | ||||
Trust in media (0-10) | 0.025** | |||
(0.009) | ||||
Trust in scientists (0-10) | 0.049*** | |||
(0.010) | ||||
Populist attitudes (1-5) | −0.422*** | |||
(0.037) | ||||
Constant | 4.708*** | 3.315*** | 3.808*** | 5.000*** |
(0.218) | (0.235) | (0.225) | (0.269) | |
Observations | 1,529 | 1,529 | 1,529 | 1,529 |
Adjusted R2 | 0.246 | 0.336 | 0.407 | 0.485 |
Residual Std. Error | 0.847 (df = 1512) | 0.795 (df = 1507) | 0.751 (df = 1506) | 0.700 (df = 1502) |
F Statistic | 32.201*** (df = 16; 1512) | 37.793*** (df = 21; 1507) | 48.581*** (df = 22; 1506) | 56.243*** (df = 26; 1502) |
Note:*p < 0.05; **p < 0.01; ***p < 0.001.
Entries are OLS regression coefficients with standard errors in parentheses. Survey weights are applied.
As the populist attitudes scale was not included in the shorter phone version of the survey, 2.7 percent of respondents who were interviewed by phone were excluded from the main analysis, presented in Table 1. The models with the full dataset controlling for the survey mode (excluding populism scale variable) are presented in Table A4 in the Appendix. As the results of the analysis with or without respondents interviewed by phone are very similar, I limit myself to discussing the results of the main analysis excluding phone interviews.
As can be seen from the table, Model 1 with socio-demographic variables and party preference accounts for 25 percent of variance of the dependent variable (Adjusted R2 = 0,246). As in many other countries (see van Mulukom et al. 2020), except the US (Cassese et al. 2020), gender is a statistically significant factor: women endorse COVID-19 conspiracies more than men; the effect holds in all subsequent models. Age has a curvilinear U-shape effect on the disbelief in CTs in Model 1; the relationship, however, diminishes in Model 2 and disappears in other models. Education and settlement type has a significant effect in all models: urban population and people with university education endorse CTs less than others. Religiosity is not a significant predictor of the belief in COVID-19 conspiracies.
Party preference has a significant effect in Model 1, revealing the difference between governing and oppositional political parties. Nonvoting and support for oppositional parties – the Union of Farmers and Greens (LVŽS), the Labour Party (DP) and other smaller parties (category ‘other’) – predicts increased belief in CTs in comparison to voting for the Freedom Party (FP, reference category), one of the governing parties. The effect of partisanship, however, decreases in subsequent models and finally disappears in Model 4, where trust and populist attitudes are controlled.
In Model 2, I add variables of geo-political attitudes and behaviour: trust in the EU, like – dislike of Putin and the use of Russian media. Adding these variables significantly improves the model, in general confirming the findings of Achimescu et al. (2021). Trust in the EU has a statistically significant positive effect on the disbelief in COVID-19 CTs. Russian media effects are statistically significant while not linear: those who never use Russian media are less likely to believe in COVID-19 CTs than heavy users of Russian media (those using it at least once a week), but occasional usage of Russian media has no statistically significant effect. Liking Putin does not have a statistically significant effect on the belief in COVID-19 CT, seemingly because this variable has a very low variation in Lithuania (the mean of this variable is 1.09 on the 0–10 scale).
In Model 3, I add soviet nostalgia. F-test statistics show that adding this variable improves the model significantly; Adjusted R2 increases from 34 to 41 percent. As can be seen from the table, the communist nostalgia variable has a statistically significant (p < 0,000) and substantive effect on the belief in COVID-19 CTs, as expected: people less prone to the soviet nostalgia are less inclined to endorse COVID-19 CTs, even controlling for their trust in the EU and the use of Russian media. An increase of communist nostalgia score by 1 point (on a 5-point scale) diminishes the index of disbelief in COVID-19 CTs by 0,29 point. Therefore, controlling for other factors, an increase from 1 to 5 in the communist nostalgia scale would reduce the disbelief in CTs by 1.2 point – an effect much greater than that of university education or party preference. The effect of soviet nostalgia also exceeds the effects of the exposure to Russian media and trust in the EU. It is worth noting that adding communist nostalgia reduces the coefficients of Russian media and trust in the EU. This implies that communist nostalgia might be one of the determinants of anti-Western attitudes.
Does the effect of communist nostalgia hold when controlling for other attitudinal variables? In Model 4, I add trust variables (political trust, trust in media and trust in scientists) and populist attitudes. As can be seen from the table, Adjusted R2 of the model rises from 0,41 to 0,49, mostly due to the substantial effects of populist attitudes scale. As in other countries, populist attitudes appear to be the strongest determinant of the beliefs in COVID-19 conspiracies in Lithuania. Controlling for other attitudes, political trust has no statistically significant effect (it is not surprising as political trust is strongly correlated with populist attitudes), while the effects of trust in media and trust in scientists are modest but statistically significant. Adding trust and populism variables in Model 4 reduces the coefficient size of communist nostalgia from 0,29 to 0,2 (p < 0,000); it still, however remains the second-best predictor of COVID-19 CT beliefs after the populist attitudes.
Conclusions
The spread of conspiracy theories related to COVID-19 is a global phenomenon, but it has its own specific features in different regions and countries. In order to understand what fosters beliefs in CTs in a specific political and cultural environment, it is crucial to know what pre-existing attitudes and behavioural patterns are conducive to the diffusion of conspiracies. For example, in the US, the spread of conspiracies is affected by the polarisation of the political elites and media between the conservative and liberal camps; in some Western European countries it is induced by populist attitudes and/or the support for radical right parties. In other regions, there might be some other political divisions and specific political constellations that encourage the spread of conspiracy beliefs.
This article focuses on the case of Lithuania, which represents the region of post-communist Central and Eastern Europe. A significant feature of this region is widespread communist nostalgia, which is the result of a painful post-communist transformation process that drastically reshaped peoples’ lives and produced a large stratum of people who have experienced political marginalisation and objective or subjective deterioration of their social status. The article argues that communist nostalgia provides a fertile ground for conspiratorial narratives, as it fosters a sense of helplessness and alienation, distrust of political institutions and an opposition to mainstream narratives.
While specific mechanisms linking communist nostalgia and propensity to believe in COVID-19 CTs could not be tested in this analysis, empirical examination of the survey data from Lithuania revealed a strong and significant correlation of communist nostalgia with the conspiracy beliefs related to COVID-19. Even when controlling for such factors as trust in the European Union, the use of Russian media, like – dislike of Putin, political trust, confidence in media and scientists, communist nostalgia remains one of the strongest predictors of beliefs in COVID-19 CTs. It is very likely that this relationship would hold in other post-communist countries that experienced the trauma of post-communist transformation, as Lithuania serves a good critical case for such analysis.
The Lithuanian data demonstrates that communist nostalgia is closely interlinked with populist attitudes, distrust in authority, media and scientists. Further analysis is needed to disentangle the causal effects between these variables and to test these relationships in other post-communist countries. The paper suggests that the mind-set of post-communist citizens feeling nostalgic about the communist past is comparable to the dispositions of the supporters of populist parties in the West, as both groups are characterised by ‘nostalgic deprivation’ – a discrepancy between the perceived current social status and perception of the past (Mudde and Rovaira Kalwasser 2018). The correlations among communist nostalgia, populism and conspiracy beliefs revealed in this paper urge for closer inspection of the phenomenon of nostalgia in Central and Eastern Europe.
A closer look is also needed at the link between conspiracy beliefs and an aspiration for counter-narratives of marginalised groups in different regions. The paper suggests that higher susceptibility of Eastern Europeans to the conspiracy theories might be related to the prolonged suppression of grievances that contradicted the official story of success of the post-communist transformation. The distrust in the dominant modernisation narratives might also be an important complement of populist attitudes in the West explaining a predisposition of populists to conspiratorial thinking.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Footnotes
The data is available via the Lithuanian Data Archive for Social Sciences and Humanities (LiDA), see Ramonaitė et al., 2022.
National lockdown in Lithuania was introduced on 7 November 2020, as a wave of new infections started to grow rapidly since the end of October 2020. The wave reached the peak by the end of December 2020, with around 3,000 new cases per day in a country of population with 2 795,7 thousand inhabitants. By the end of January 2021, the average number of new cases dropped down to about 800 per day. The lockdown was cancelled on 1 July 2021 (LRV 2021).
As the first item (‘Coronavirus was probably created in a laboratory’) might be criticised for being too vague, a four-item index was also created excluding this item and used for analysis as a robustness check.
An alternative measure capturing the evaluation of the communist regime in relation to the present can be found in the Baltic Barometer 2014. It asks respondents of the survey in the three Baltic states to identify the time period when their country was best off – the interwar era, the Soviet era, the years between the restoration of independence and EU membership in 2004; or the current time period from 2004 (see Duvold et al. 2020).
See Table A5 in the Appendix for the results of the analysis with the four-item the index of disbelief in COVID-19 conspiracy theories (excluding an item ‘Coronavirus was probably created in a laboratory’).
References
Ainė Ramonaitė is Professor at the Institute of International Relations and Political Science, Vilnius University. She is the principal investigator of the Lithuanian National Election Study. Her research focuses on political attitudes and electoral behavior in post-communist countries.
Author notes
Supplemental data for this article can be accessed online at https://doi.org/10.1080/14616696.2022.2132525.