In the run-up to the most recent Belarusian election, conducted on August 9, 2020, a series of protests were already transpiring throughout Belarus. Many of the original protests supported Sergei Tikhanovsky, an opposition blogger who had been arrested in May two days after he announced on his YouTube channel his intention to challenge incumbent President Alexander Lukashenko (Amnesty International 2020a; Human Rights Watch 2020). After a series of additional arrests, Tikhanovsky was detained on the 29th of May and charged with plotting mass riots ahead of the election, accusations which human rights groups view as illegitimate (Human Rights Watch 2020). Following the arrest of Tikhanovsky, opposition candidate Viktar Babaryka was also arrested in June on charges he was attempting a coup against the government. According to polls, Babaryka was the most likely candidate to win the presidential election (Aris Reference Aris2021). Babaryka’s arrest is also viewed with suspicion from the international community, and among the mass public in Belarus (Amnesty International 2020b; Human Rights Watch 2020). Protests continued to swell in response to Babaryka’s arrest and the violent detainment of many protestors and critics of Lukashenko. These protests were met with repressive and violent action by the state police apparatus.
The support of the Belarusian populace for Sergei Tikhanovsky during the protests prompted Sviatlana Tsikhanouskaya, his spouse, to challenge for the presidency. Following her registration as an independent candidate in July 2020, Tsikhanouskaya was endorsed by Babaryka’s campaign and campaign manager Maria Kalesnikava. She was also endorsed and joined by Veronika Tsepkalo, whose husband Valery Tsepkalo was barred from running for president and eventually forced to emigrate from Belarus (Bennetts Reference Bennetts2020). Indeed, Tsikhanouskaya drew record crowds across Belarus to her rallies in the run-up to the election (Rainsford Reference Rainsford2020).
On election night, Lukashenko claimed victory with 80.1 percent of the electoral vote. The Central Election Committee asserted Lukashenko’s primary opponent, Svetlana Tsikhanouskaya, only received around 10 percent of the final vote share. The election is widely considered fraudulent with widespread irregularities. The night of the election, videos were posted to Twitter showing election officials climbing out of windows with bags presumably full of ballots (Kim Reference Kim2020). In addition, a video was circulated on social media of a peaceful protest demanding a recount of the vote in one precinct. The election official recounted the ballots and alleged Tsikhanouskaya won 90 percent of the vote in their precinct (Jankowicz Reference Jankowicz2020). Further, election monitors pointed out that some election officials were erroneously claiming over 100 percent voter turnout (Wesolowsky Reference Wesolowsky2020). Finally, election workers have described and admitted committing fraud when counting the results in multiple precincts across the country (Manenkov and Litvinov Reference Manenkov and Litvinova2020).
The widespread and blatant electoral fraud led to even larger and more frequent mass protests around Belarus. Unfortunately, these protests were met with large-scale, extreme levels of state-sponsored violence against the protestors, including arbitrary detainment, beatings, torture, and other ill treatment (Organization for Security and Co-operation in Europe 2020; Human Rights Watch 2021). The numerous videos posted online displaying state sponsored violence, and the abundance of evidence of torture, led Amnesty International to quickly provide a petition calling for the end of police violence in Belarus (Amnesty International 2020b). Despite the large-scale state-sanctioned violence against peaceful protestors, citizens of Belarus continued to protest and march in the streets for the next seven months.
This study sets out to explore whether historical trends in political attitudes and behavior can help us explain why citizens in Belarus participated in those large-scale demonstrations despite knowing that they would be met with state-sanctioned violence. First, did the mass demonstrations follow a historical trend of increasing dissatisfaction with the Belarusian government among citizens? Second, are the mass demonstrations a product of an increasing political participatory culture?
Using World Values Surveys (WVS) conducted in Belarus in 1996 and 2018, we explore these two questions in order to compare political attitudes and behavior in the early days of Lukashenko’s rule to the latter period. In our analysis, three findings stand out. First, we find that satisfaction with the political system and confidence in political institutions was slightly lower in 1996 when compared to 2018. Second, more individuals indicated a willingness to engage in political participation in 1996 than in 2018. However, multivariate models indicate that political participation, in particular the willingness and experience to engage in demonstrations, are equally probable for people dissatisfied with the system when comparing 1996 and 2018. Conversely, those satisfied with the political system are much less likely to participate in demonstrations in 2018. The results suggest that sustained authoritarian rule made citizens less likely to express their dissatisfaction with the regime, both in the form of public demonstrations but also possibly even in surveys. In line with the literature on political mobilization under repressive regimes (Kuran Reference Kuran1989, Reference Kuran1991), these findings suggest that preference falsification disguised underlying discontent with the regime prior to 2020. Once a number of internal and external factors pushed some actors to protest against Lukashenko’s rule in 2020, a bandwagon effect drove other groups to the streets, contributing to the large-scale demonstrations that swept the country at the time.
Citizen Dissatisfaction with the Regime
The Lukashenko government is firmly considered an authoritarian regime that restricts the political rights of the citizens of Belarus. From 1997 to today, the Polity5 dataset has scored Belarus as an authoritarian state that has not witnessed any movement towards democracy (Marshall and Gurr Reference Marshall and Gurr2021). Similarly, Freedom House, which has rated Belarus “not free” over this same time period, has noted a significant decline in the country’s overall freedom score from 2017 (score = 20) to today (score = 11). In particular, Freedom House (2021) notes a sharp curtailing of dissent by the Lukashenko regime. In fact, since 1998 the V-DEM dataset indicates that Belarus has a minimum score on the Deliberative Democracy Index, the Liberal Democracy Index, and the Participatory Democracy Index (Coppedge et al. Reference Coppedge, Gerring, Knutsen, Lindberg, Teorell, Alizada and Altman2021). Taken together, all of the indices demonstrate the Belarussian government severely curtails the political participatory rights of the citizens.
One explanation for mass demonstrations within an authoritarian state is that citizens’ dissatisfaction with the regime has reached a breaking point. In his edited volume, Tismaneanu (Reference Tismaneanu2005) shows how citizen dissatisfaction with the economic landscape of the Eastern European countries led to mass protests that eventually led to the Revolutions of 1989. From the strikes of the Solidarity movement in Gdańsk, Poland, to the continuous demonstrations in Leipzig, Germany, these public acts of political participation could be explained by deep citizen malcontent with a range of governmental institutions and behaviors. Francisco (Reference Francisco1993) shows in his study on the Revolutions of 1989 in Central and Eastern Europe that the protests were a function of mobilization based on a high level of discontent of the population and views that protest actions are a rational means to achieve fundamental change. Based on our understanding of the occurrence of the protests prior to the Revolutions of 1989, we might expect that the demonstrations in Belarus are a function of increasing dissatisfaction with overall government performance or specific political institutions.
In fact, trust in institutions was fairly high in Belarus compared to other Eastern European countries at the beginning of Lukashenko’s presidency. Mishler and Rose (Reference Mishler and Rose1997) showed that in 1994 a relatively high number of Belarussian citizens expressed trust in their political institutions (37 percent), half of respondents viewed political institutions with skepticism (50 percent), and a small share of respondents distrusted political institutions (13 percent) (Mishler and Rose Reference Mishler and Rose1997). However, we might expect this relatively high level of trust to drop over the course of Lukashenko’s rule. In fact, Abott (Reference Abott2007) argues there was a decline in positive perceptions of the government due to several economic and political changes in Belarus since 1991. The reason is that the anticipated outcomes from the political change, such as greater democracy and economic success, did not materialize for most citizens. Therefore, we might expect citizens in the latter period of Lukashenko’s rule to have higher dissatisfaction with the government.
However, it could also be the case that satisfaction with and support for the regime increased as Lukashenko’s rule consolidated. Indeed, Lukashenko’s repeated engagement in electoral fraud may point in this direction. Rundlett and Svolik (Reference Rundlett and Svolik2016) find that authoritarian incumbents may enjoy high levels of genuine popular support even when they engage in higher levels of electoral fraud. The authors argue that fraud is easier to commit for authoritarian leaders when they have a high level of genuine support, which implies that rulers might not be committing electoral fraud due to vulnerability. Therefore, we cannot simply assume that Lukashenko engages in more blatant electoral fraud due to a loss of satisfaction in his regime.
There are many reasons why we might expect satisfaction with the government to remain the same or be higher over the course of Lukansheko’s rule. A deepening of personalized, authoritarian rule might lead to stable or higher satisfaction among the public. In Belarus, Korosteleva (Reference Korosteleva2004) argues that people’s satisfaction with the government is tied to an emotional loyalty to past Soviet legacies. Leshchenko (Reference Leshchenko2008) agrees, noting that, upon his 1994 election, Lukashenko returned the Belarusian flag to the former Soviet colors and design, brought back the Soviet Belarusian anthem, as well as much of the Soviet economic structure and social safety nets: initiatives popular with the majority of Belarusians at the time. In this vein, Lukashenko also developed what the author calls “egalitarian nationalism”: a type of national identity that is both ethnically inclusive and focuses on “principles of collectivism and anti-liberalism,” upholding Belarusian “uniqueness” (Leshchenko Reference Leshchenko2008). This strengthens both the Belarusian national identity and the paternalistic authoritarian regime at once. Further, the Belarusian regime is defined by its mass approval of and reliance on the personalized power of a demagogue who governs by decree yet seeks legitimacy from the electorate directly (Korosteleva Reference Korosteleva2004, 128–129; Leshchenko Reference Leshchenko2008). Therefore, it is expected the regime will have mass support due to this appeal. Indeed, Korosteleva (Reference Korosteleva2009) attributes the failure of the protests following the 2006 election to develop into a color revolution to the full identification with the president by a large share of the population. She argues that the public’s full identification with Lukashenko legitimizes his politics and policies and thus makes it unlikely his regime is ever completely challenged (Korosteleva Reference Korosteleva2009, 336).
Besides cultural identification, Lukashenko also has developed a range of tools at his disposal to deter dissent and dissatisfaction with the government. Silitski (Reference Silitski2005, Reference Silitski, Bunce, McFaul and Stoner-Weiss2009) refers to the strategy of utilizing tools to preempt democracy in Belarus as “preemptive authoritarianism,” which involves utilizing the tools of authoritarianism to prevent democracy. Marples (Reference Marples2006) points out Lukashenko’s total dominance over the media and persecution of enemies allows him to control the population’s sentiment toward the government. Kalandadze and Orenstein (Reference Kalandadze and Orenstein2009) refer to this ability to control the narrative and prevent opponents from developing a platform to challenge Lukashenko’s ideas as the “policy of preemption.” The ability to prevent dissent from percolating through the population means there are less narratives in the mass public to form dissatisfaction with the regime. This ability to prevent dissent creates a paternalist regime model where Lukashenko, and Lukashenko alone, is perceived as capable of ruling the country (Buzgalin and Kolganov Reference Buzgalin and Kolganov2021). Finally, both Dimitrova et al. (Reference Dimitrova, Mazepus, Toshkov, Chulitskaya, Rabava and Ramasheuskaya2021) and Buzgalin and Kolganov (Reference Buzgalin and Kolganov2021) argue that Belarus has a high state capacity to provide public goods to citizens. Lukashenko’s use of this high state capacity to provide public goods is a useful tool to keep the population satisfied with his rule. In particular, the authors demonstrate how health care is better in Belarus than in a relatively more democratic country: Ukraine. Buzgalin and Kolganov (Reference Buzgalin and Kolganov2021) conclude that this paternalistic regime model helps insulate Lukashenko from broad citizen dissatisfaction. Therefore, comparing 1996 and 2018, we expect that the public’s satisfaction and confidence in the government and political institutions has remained the same over time.
Civic/Participatory Attitudes
Another potential explanation for the eruption of mass protests in 2020 is an increase in individual civic or participatory attitudes. Generally, protests represent a nonviolent campaign to achieve a particular objective. Protest activity can be seen as civil or nonviolent resistance to an authoritarian regime’s policies or existence (Roberts and Ash Reference Roberts and Ash2009). Thus, the broader protest literature states that protests represent the most visible form of political participation in authoritarian regimes.
On the one hand, some developments in Belarussian society over the past 25 years would make us expect higher levels of political participation over time. Finifter (Reference Finifter1996) finds that higher education levels in the former Soviet Union are associated with support for individual responsibility and individual action, such as political participation. Therefore, as more Belarusians continue to obtain higher levels of education on average, the country is expected to have a more robust participatory culture. Inglehart and Welzel (Reference Inglehart and Welzel2003) argue that individual support for democratic practices and engagement in political participation is associated with cultural shifts in attitudes and behavior. Since the European Commission (2021) finds the Belarusian higher education system has been growing rapidly since 1991, we might except more individuals to have a stronger sense of civic/participatory attitudes. In congruence with this idea, Kulakevich (Reference Kulakevich2014) points out there has been modest growth in the number of protestors during presidential elections. In addition, Kulakevich (Reference Kulakevich2014) argues that a growth of protestors in the future is expected due to the transnational flow of a globalizing world providing citizens with increased amounts of information and ideas about political action.
On the other hand, there are reasons to expect that the development of civil society and increased participatory behavior would not occur in Belarus during that period. Citizen reliance on the state means the prospects for the development of a civic culture and political participation against the state may be low. Levitsky and Way (Reference Levitsky and Way2005) argue that democratic institutions alone are not enough to develop a robust democratic participatory culture. In particular, elections and increases in education do not immediately lead to a civic culture and increases in political participation (Carothers Reference Carothers2002). Howard and Roessler (Reference Howard and Roessler2006) find that the allowing of elections by authoritarian regimes only sporadically lead to liberalization and an increase in opportunities for participation. Mishler and Rose (Reference Mishler and Rose1997) and Way (Reference Way2005) contend that any competitive politics in Belarus that exist are not rooted in a robust civil society, rather an inability by elites to control elections, media, and opponents. Similarly, Lenzi (Reference Lenzi2002) argues that there is an intentional absence of the development of a civil society in Belarus caused by the regime. The Lukashenko government has prevented the creation of any independent institutions that may lend to the creation of a robust civil society ultimately reducing robust vertical threats to his authority.
Indeed, there seems to be a regular cycle of protest behavior in Belarus over the past 25 years. Lukashenko triggered protests known as the “Minsk Spring” or “Belarussian Spring” in March and April 1996 as a result of his move to increase the power of the presidency just two years after winning his first presidential term. More protests were sparked in reaction to Lukashenko claiming to have won almost double the votes in the 2001 presidential election. In October 2004, the passing of a referendum to allow the president to serve more than the two terms in office provoked demonstrations that lead to violent clashes with the police, dozens arrested, and the arrest and sentencing of opposition leaders. In March 2006, Lukashenko was declared the winner of the presidential election by a landslide. Between 10,000–30,000 people responded in protests known as the “Jeans Revolution” or the “Cornflower Revolution,” and hundreds of people were arrested, including former presidential candidates. Again, countering the 2010 presidential election over 40,000 people demonstrated in Independence Square, leading to hundreds of arrests.
Along these lines, research on protest behavior suggests that, even if political attitudes change, the costs of participating in collective action may be too high for most citizens except the most committed political activists in strong authoritarian states (Kuran Reference Kuran1989, Reference Kuran1991; Lichbach Reference Lichbach1995). Fearing reprisals from a repressive regime, citizens may engage in preference falsification and be hesitant to express any negative feelings about the regime, thus disguising underlying levels of discontent.
As discussed earlier, Dimitrova et al. (Reference Dimitrova, Mazepus, Toshkov, Chulitskaya, Rabava and Ramasheuskaya2021) point out the Belarusian government has a strong state capacity, which means the state is able to control vertical attacks to its authority. For example, Rosenfeld (Reference Rosenfeld2017) finds that controlling the size of protests is partially a function of public sector employment. Since public sector employment is more extensive in Belarus when compared to countries such as Ukraine or Georgia, the government has been able to sustain a low to moderate level of protest behavior through threats of losing employment (Rosenfeld Reference Rosenfeld2017). Buzgalin and Kolganov (Reference Buzgalin and Kolganov2021) agree that workers are less likely to participate in demonstrations due to worry about loss of employment. However, they also argue that workers are concerned about the overall economic stability of the paternalistic economy in Belarus and question whether the state’s ability to provide will continue. The authors argue that the recent protests were primarily composed of younger students willing to take greater security risks, which is a trend seen in previous protests.
Based on this literature, we would expect that the number of citizens expressing willingness to engage in political actions would remain more or less stable over time – typically those who express greater levels of dissatisfaction with the regime. Conversely, we expect people satisfied with the government to be less likely to say that they might participate or have participated in political actions in 2018 because they will be more aware of the potential consequences of speaking out against the government based on a history of crackdowns to participatory activity.
In sum, it is worth asking whether long-term changes in citizens’ political attitudes and participation lay behind the dramatic explosion of political protests in 2020. In what follows, we analyze how satisfaction with the regime, confidence in political institutions, and protest behavior differed from the early days of Lukashenko’s rule to the years prior to the 2020 demonstrations, and whether the willingness to participate in political protests was more or less associated with expressed dissatisfaction with the regime at these different points in time.
Data
To examine these issues, we utilize data from Wave 3 of the World Values Survey (WVS) and Wave 7 of the Joint World Values Survey/European Values Study (WVS/EVS) conducted in Belarus in 1996 and 2018 respectively (Inglehart et al. Reference Inglehart, Haerpfer, Moreno, Christian Welzel, Diez-Medrano and Lagos2014; European Values Study 2017). The WVS conducted in 1996 was the first version of the survey conducted after the adoption of the Belarusian national constitution and Lukashenko’s election as president. The WVS/EVS survey conducted in 2018 was the last version of the survey to be conducted in Belarus prior to the 2020 presidential election and is the most recent large-scale data collection on attitudes among citizens in the country. Unfortunately, Belarus was not included in Waves 4 or 5 of the WVS. In Wave 6, the satisfaction with the political system and political participation survey questions were not asked of respondents in Belarus. Since the political participation questions represent the main dependent variables of interest for the multiple regression models, it was not possible to utilize Wave 6 in those analyses. That being said, we do include data from Wave 6 as a robustness check for our analyses of citizens’ confidence in the government and other institutions in the Appendix, and provide a discussion in the analysis section.
It is important to highlight how notoriously difficult it is to carry out public opinion research under repressive authoritarian regimes. Besides the obvious logistical and security challenges posed for researchers in those contexts, survey respondents are also more likely to hide their true preferences for fear of repression. While this means that some of our results should be taken with caution, these two survey datasets allow us to compare how satisfaction with the government, confidence in governmental institutions, and political participatory behavior differed between the early days of Lukashenko’s regime and the years prior to the 2020 demonstrations. Furthermore, this comparison also allows us to produce some tentative remarks about whether respondents were more or less likely to engage in preference falsification as the regime consolidated.
Methods
We conduct the analysis in three stages. First, we explore whether levels of satisfaction with the political system and confidence in various political institutions differed between 1996 and 2018. For satisfaction with the political system, respondents are asked, “how satisfied are you with how the political system is functioning in your country these days?” Then, respondents are provided a 1–10 scale to place their satisfaction with the political system where 1 represents “not at all satisfied” and 10 represents “completely satisfied.” For political institutions, we explore confidence in six governmental institutions, which include the government, parliament, political parties, civil service, courts, and police. The prompt for the questions states, “I am going to name a number of organizations. For each one, could you tell me how much confidence you have in them: is it a great deal of confidence, quite a lot of confidence, not very much confidence, or none at all?” We estimate bivariate linear regression models in order to test whether the level of satisfaction and confidence in these entities is statistically different in the two years. All the statistical models include post stratification survey weights, which account for the main socio-demographic variables. In the Appendix, we present the results of additional analyses using the 2011 data from Wave 6 of the WVS.
The second stage of the analysis compares levels of political participation in 1996 and 2018. In particular, we explore differences in the rates of respondents indicating that they “would never do,” “might do,” or “have done” four political participatory activities: signing a petition, joining in boycotts, joining an unofficial strike, and attending a lawful/peaceful demonstration. Here, we estimate bivariate ordered logistic models utilizing survey weights to test whether there are statistical differences between the two years.
In the final stage of the analysis, we analyze how the relationship between dissatisfaction with the regime and political participation differed between these two periods. Since in the second stage of the analysis we find that willingness to engage politically appears to be higher in 1996 when compared to 2018, we estimate multiple regression models to further investigate the potential cause of these differences. We include a number of common control variables that seek to predict political participation, including age, sex, education, income, and political interest, as well as survey weights.Footnote 1 We estimate ordered logistic regression models because the participatory variables are coded as 0 = “would never do,” 1 = “might do,” and 2 = “have done.”Footnote 2 Variable coding and descriptive statistics are provided for the control variables in Appendix A. Our main independent variable of interest is satisfaction with the political system. Based on the literature discussed above, we expect respondents who are satisfied with the government to be less likely to indicate that they “would do” “or have done” a political participatory activity in 2018 when compared to 1996. This is because people who maintain some satisfaction with the government should be less likely to participate in political activities as the regime consolidated, and they became aware of the potential consequences of raising their voices against Lukashenko’s rule.
Results – Comparing Satisfaction and Confidence in the System and Institutions
In Figure 1, we plot the density of respondent selections for the variable that asks about satisfaction with the political system in 1996 and 2018. Output from a bivariate linear regression model indicates that respondents were statistically more satisfied with the political system in 2018 when compared to 1996. As Figure 1 displays, a larger density of respondents chose values 1–5 representing dissatisfaction in 1996 (88.05 percent) when compared to 2018 (55.25 percent). In comparison, in 1996, only 11.94% of respondents chose values 5–10, indicating some level of satisfaction with the political system compared to almost half of respondents in 2018 (44.76 percent). The differences in respondents’ level of satisfaction with the political system are quite striking when comparing the early rule of Lukashenko to the latter time period. The results here indicate that people are more satisfied with the political system in 2018, which provides some support for authoritarian deepening or acceptance of the regime. It is important to highlight that Lukashenko held a controversial referendum on Russification of Belarus in 1995. While the official results show that the referendum obtained overwhelming support, the referendum is widely seen as fraudulent and as the Lukashenko’s first power overreach. The lower level of support in the political system in 1996 could be related to the referendum conducted the prior year. As discussed earlier, The V-DEM and Freedom House indices both indicate a deepening of authoritarianism in Belarus over the course of Lukashenko’s presidency (Coppedge et al. Reference Coppedge, Gerring, Knutsen, Lindberg, Teorell, Alizada and Altman2021; Freedom House 2021).
Figure 1. Satisfaction with the Political System – 1996 and 2018
Bivariate analysis indicates a statistically significant difference.
A similar trend is present when exploring confidence in individual political institutions. In Figure 2, we plot the density of respondent selections for the variables that ask about confidence in the six political institutions in 1996 and 2018. Bivariate linear regression output indicates that for five out of the six institutions there is a statistically significant difference between attitudes in 1996 and 2018. There is no statistically significant difference in confidence in the government between the two years. In comparison, respondents have a slightly higher level of confidence in the civil service in 1996 when compared to 2018. Overall, 51.34 percent of respondents indicate some level of confidence in the civil service (selecting values 6–10) in 1996 compared to just 33.8% of respondents in 2018. That being said, when evaluating the remaining four institutions, respondents have statistically higher levels of confidence in these institutions in 2018.
Figure 2. Confidence in Country’s Institutions – 1996 and 2018
* Indicates a statistically different bivariate difference between the years.
In 1996, around one-third of respondents (29.9 percent) expressed some positive level of confidence in parliament (selecting values 6–10). In contrast, in 2018, just over a majority of respondents (50.9 percent) indicated a positive level of confidence in parliament, which is a fairly large difference of 20 percent. When exploring confidence in the courts and political parties, just over 10 percent more respondents indicate a higher level of confidence in these institutions in 2018 compared to 1996. Less than a majority of respondents (48.8 percent) had some level of confidence in the courts in 1996. Contrasting 2018, we note increased confidence in courts to almost two-thirds of respondents (62 percent). In both years, positive evaluations of confidence in political parties are relatively low at 16.4 percent in 1996 and 29.7 percent; however, confidence was notably higher in 2018. Finally, there is a strikingly large difference between the years when exploring confidence in the police. In 1996, just over one-third of respondents (37.8 percent) indicate a positive level of confidence in the police. Surprisingly, almost two-thirds of respondents (62.2 percent) indicate a positive level of confidence in the police in 2018. The 24.4 percentage difference between the two years is sizable. Comparing 1996 and 2018, the public’s satisfaction and confidence in the government and political institutions has remained the same or increased over time. The results provide some evidence that people were more likely to express acceptance of the authoritarian government and political institutions as the regime consolidated.
As a robustness check, we include Figure C1 in the appendix. Figure 1 displays confidence in the six institutions for 2011 in comparison to 1996 and 2018. The same trends exist. Confidence in the government is statistically similar in all three years. Confidence in parliament, political parties, civil service, courts, and the police is statistically higher in 2011 when compared to 1996. In addition, when comparing 2011 and 2018, there is only one institution where there is a statistically significant difference in confidence. In 2011, confidence in the civil service is statistically greater than it is in 2018. It is important to point out that the 2011 survey was conducted just a few months after the 2011 Minsk Metro bombing. Therefore, if survey respondents are identifying first responders as members of the civil service, the high level of confidence in the civil service could be a product of their performance following the attack. For example, the United States witnessed high levels of support for police and firefighters in the wake of 9/11.
Taken together, the results provide some evidence for Silitski’s (Reference Silitski2005, Reference Silitski, Bunce, McFaul and Stoner-Weiss2009) conceptualization of “preemptive authoritarianism.” Lukashenko has been able to use the tools of authoritarianism to preempt democracy in Belarus. For example, as stated prior, Lukashenko was able to utilize the tools of authoritarianism to create a healthcare system that is rated better than neighboring democratic countries. These types of activities were useful in limiting public expressions of dissent and maybe even creating contentment with the regime.
Results – Comparing Rates of Participatory Activity
Figure 3 displays the rates of political participatory activity in 1996 and 2018. The top panel presents the percentage of respondents indicating they “might” or “have” engaged in each of the four participatory activities. In the bottom panel, the percentage of respondents that indicate they “might do” or “have done” these activities are presented with the groups combined. For all four participatory activities, respondents were statistically less likely to indicate that they “might do” the activity in 2018. Between 5–7 percent more respondents indicated they might sign a petition, boycott, or participate in a strike in 1996 compared to 2018. A significant gap (11.4 percent) exists between the two years regarding the percentage of people indicating they might participate in a demonstration.
Figure 3. Descriptive Statistics of Participatory Activity – 1996 & 2018
In 2018, respondents were statistically less likely to respond that they had participated in a boycott or demonstration. There is a small gap (around two percent) when it comes to participation in boycotts. In contrast, the gap (12.5 percent) is quite large when it comes to demonstrations. Given that Belarus has witnessed an increase in the number of people who have participated in political demonstrations in the past two years, the result is quite surprising.
Overall, the findings from the first two sets of analyses indicate that Belarussian respondents tended to express higher levels of satisfaction with the political system, higher levels of confidence in most institutions, and a lower willingness to participate in political activities in 2018 compared to 1996. Drawing on the literature on protest behavior, one potential explanation for these results is that as the regime consolidated, citizens became less likely to express their discontent and their willingness to participate in political activities against the regime for fear of reprisals.
Results – Ordinal Logistic Regression Models Predicting Political Participation
To examine the relationship between dissatisfaction with the regime and willingness to participate in political activities in more detail, Table 1 displays model output from a series of ordinal logistic regression models predicting the four forms of political participation in 1996 and 2018. For the most part, the control variables appear to have the same relationship on predicting political participation in both years. Age is negatively related to signing a petition, boycotting, or participating in a strike in both years. Age has no relationship with demonstrating in either year. Women are less likely than men to sign a petition, boycott, or strike. In 2018, women were less likely than men to participate in a demonstration. However, women and men were equally likely in 1996. Given that both women and youth have played an important role in the 2020 demonstrations, these findings call for further analysis, but they may suggest that it was precisely those groups that were more likely to engage in preference falsification as Lukashenko’s rule consolidated. Education is positively correlated with signing a petition, boycotting, or striking in both years. However, education is only positively related to participating in a demonstration in 1996. Additionally, income only has positive relationship with the four participatory acts in 1996. Again, this provides some evidence that people who are doing better under Lukashenko’s rule have learned that there are personal and economic consequences to their political participation. Individuals at higher income levels might realize they could lose their wealth and status by participating in activities against the government. Finally, political interest has a positive, statistically significant relationship with all four participatory activities in both years. Those who monitor government and political actions are the people most likely to participate in activities against the government.
The focal independent variable in the models in Table 1 is satisfaction with the political system. In both years, the model output indicates that people who are satisfied with the political system are statistically less likely to sign a petition, boycott, or strike. However, people satisfied with the political system are less likely to participate in a demonstration in 2018 only. The results illustrate that those who were satisfied with the political system were equally as likely as those who were not satisfied with the political system to demonstrate if an issue important to them arose. Given the model output does not directly convey the impact that satisfaction with the political system has on those who respond “might do” compared to “have done,” we plot predicted probabilities holding all other variables at their survey-weighted means in Figure 4.
Table 1. Ordered Logit Models Predicting Political Participation – 1996 & 2018
* indicates statistical significance at p > 0.05. Standard errors in parentheses. Survey weights utilized.
Dependent variable: 0 = “Would never do”; 1 = “Might do”; 2 = “Have done”
Figure 4. Impact of Satisfaction w/ the Political System on Participatory Activity – Ordered Models
Predicted probabilities calculated holding all independent variables at their survey weighted means. 95% confidence bounds displayed.
Predicted probabilities calculated holding all independent variables at their survey weighted means. 95% confidence bounds displayed.
In the top panel of Figure 4, we plot the probability a respondent indicated they “might do” each of the four political activities for respondents at the lowest and highest levels of satisfaction with the political system in 1996 and 2018. The plot in the top panel shows there is no statistical difference in the probability of predicting a respondent “might do” any of the four activities for respondents at the lowest level of satisfaction between the two years. The result suggests that those dissatisfied with the system would be equally likely to participate in activities against the government in both years. In contrast, the probability a respondent indicates they “might do” in terms of signing a petition or participating in a demonstration is statistically lower for respondents at the highest level of satisfaction with the political system in 2018 when compared to 1996. Respondents who are completely satisfied with the political system are five percent less likely to indicate they might sign a petition and 15 percent less likely to indicate they might participate in a demonstration. This finding suggests that people who are satisfied with the political system might be more aware of the deleterious effects to their security if they participate in activities that antagonize the government. Whether these results are driven by preference falsification of respondents when answering the survey question or by their “true” attitudes, they suggest that as the regime consolidated, more satisfied citizens were less likely to express a willingness to engage in political activities.
The bottom panel of Figure 4 presents the probability a respondent has done each of the four participatory activities in the two years for respondents at the lowest and highest levels of satisfaction with the political system. There is an equally low probability of a respondent having indicated boycotting or striking when comparing both the years and the level of satisfaction with the political system. For people at low levels of satisfaction, respondents were slightly more likely (three percent) to indicate they had signed a petition in 2018. At high levels of satisfaction, there were no differences in the probability of having signed a petition. A particularly interesting result can be found when we explore the differences in having participated in a demonstration. In 1996, respondents who exhibited the lowest level of satisfaction with the political system were slightly more likely (two percent) to indicate they demonstrated. In contrast, there is a fairly large gap (15 percent) between the two years when examining the probability that a respondent has demonstrated at the highest level of satisfaction. In 1996, respondents satisfied with the political system were much more likely to indicate that they demonstrated in the past when compared to 2018. Again, the results suggest that people may have learned over time that they face significant threats to their security for demonstrating in Belarus. Therefore, people indicating satisfaction with the system became less likely to state that they have engaged in such behavior. As authoritarian rule deepens, this creates a feeling of futility in vertical action as being ineffective as preference falsification for the regime increases (Kuran Reference Kuran1989). In other words, citizens may show an increasing support for a regime they may privately hate, making it difficult for others in society to gauge support for the opposition.
Overall, the results from the ordinal logistic regression analyses explain why a lower percentage of people in 2018 replied they “might do” or “have done” when asked about demonstrating: whereas those highly dissatisfied with the regime continued to express their willingness to protest, those claiming to be highly satisfied were less likely to express any willingness to engage in political action. When taken against the backdrop of the 2020 protests in Belarus, these results suggest that a large bandwagon effect took place at the time, triggered by a number of short-term factors (i.e., the mishandling of COVID-19 and evidence of electoral fraud) that pushed groups that only two years before were less likely to express their willingness to protest – e.g., youth, women, more educated citizens – out into the streets.
Conclusion
This study set out to explore whether historical trends in political attitudes and behavior contributed to explain why citizens in Belarus engaged in mass demonstrations in 2020. First, were the mass demonstrations a product of increasing dissatisfaction with the Belarusian government over time? Second, were the mass demonstrations a product of a growing political participatory culture in the country?
Using World Values Surveys (WVS) conducted in Belarus in 1996 and 2018, we explored these two questions in order to compare political attitudes and behavior in the early days of Lukashenko’s rule to the latter period. We highlight three findings. First, satisfaction with the political system and confidence in political institutions was slightly lower in 1996 when compared to 2018. The results indicate that sustained authoritarian rule made citizens less likely to publicly express negative attitudes towards the system. Second, a greater percentage of individuals indicated a willingness to engage in political participation in 1996 when compared to 2018. This result is explained by our third finding from exploring the attitudes of the individuals willing to engage in political participation in the two years examined. Dissatisfied citizens were equally likely to express their willingness to engage in and experience with political action through demonstrations in 1996 and in 2018. However, we find that people who report satisfaction with the political system were much less likely to express their willingness to participate in demonstrations in 2018, likely due to the evidence over time of the personal consequences of acting against Lukashenko, as well as the seeming futility of vertical action.
Based on the results of our analysis, we find the recent large-scale mass protest is yet another entry in a long-term trend of regular demonstrations against the government. The results lead us to conclude that the scale of the demonstrations in 2020–2021 was exacerbated by a number of conjunctural factors that triggered a bandwagon effect, bringing into the street protestors beyond the core group of dissatisfied citizens that would routinely protest in the face of state injustice. As Kulakevich (Reference Kulakevich2014) argues, the weakness of the mobilizing structures and framing processes had previously resulted in the absence of large-scale protests in Belarus. In the current situation, the issues that citizens are demonstrating about are plentiful (economy, handing of COVID-19, corruption, electoral fraud, state-sponsored violence) and touch multiple segments of the society. Once the most dissatisfied citizens took to the streets in the context of the 2020 elections, these issues offered new frames to voice discontent with the government. These factors compounded to mobilize new groups and individuals, even in the face of potential negative personal consequences.
This study utilized survey data to explore political attitudes and political participation in Belarus. While the analysis is straightforward, it is important to recognize its limitations. First, this study only explores two (sometimes three) time periods and it would be helpful if data were available to provide a more comprehensive time series analysis. Second, the use of survey data is restricted to the questions that were asked by the original researchers and to the challenges of measuring political attitudes under repressive authoritarian regimes. Even though preference falsification is likely to be driving some of our results, that in itself is an important finding, since it shows that citizens become less likely to express their discontent with the political system and their willingness to engage in political actions as authoritarian regimes consolidate, even on the eve of a major wave of political protests. Future research incorporating interviews into the analysis would greatly help our understanding and provide much needed thick description to these findings.
Supplementary Materials
To view supplementary material for this article, please visit http://doi.org/10.1017/nps.2022.64.
Acknowledgements
The authors would like to thank the editors of Nationalities Papers and the anonymous reviewers for their feedback throughout the peer-review process. The authors would also like to give a special acknowledgement and expression of gratitude to Agustín Goenaga for his invaluable feedback.
Disclosures
None.
Open access
