Digital activism and democratic culture: can digital technologies help save democracy?

This study investigates digital activism and democratic culture among citizens of São Paulo, Brazil. It aims to understand: 01. whether digital participation is becoming a surrogate instance of other forms of participation; and 02. whether digital activists share a specific political culture regarding democratic attitudes and behaviors. Drawing on a representative sample of 2,417 interviews, we apply multivariate analysis techniques to characterize digital activists in terms of political participation and democratic culture. Our findings show that digital participation complements – rather than replaces – other forms of participation. We also found that although the levels of democratic culture among digital-only activists were lower than among activists who participate in several arenas, digital-only activists embrace democratic culture more significantly than non-activists. This finding suggests that digital-only participation could be an important first step in developing democratic attitudes in individuals, albeit this form of participation is not sufficient to foster the highest level of democratic culture.

his article addresses the issue of digital activism among citizens of São Paulo, the main Brazilian metropolis. Throughout the country's history, the city of São Paulo has been portrayed as one of the most important venues for political participation, as illustrated by the Constitutionalist Revolution in 1932, the March of the Family with God for Liberty in 1964, the civil unrest demanding direct presidential elections (Diretas Já) in 1984, and the June Journeys in 2013, just to mention a few (LACERDA and SIMONI JR., 2021). In face of growing risks of democratic deconsolidation around the world (FOA and MOUNK, 2017), we investigate the digital technologies' potential to restore the ordinary citizens' appreciation for democracy, as these technologies provide new means for political participation and engagement in social causes. However, the synergy between the digital world and the democratic regime has never been taken for granted within political theory. In fact, even the question of the extent to which political participation is beneficial for democracy has been the subject of heated debate among political theorists.
In democratic theory, political participation typically refers to the behavior of the citizenry intended to affect politics, and it has been seen as a factor of vitality (PUTNAM, 2015), legitimacy (FISHKIN, 2015), and influence (VERBA and NIE, 1972) in the regime. Although the definition of political participation is disputed, most contemporary scholars agree that, by and large, it refers to a voluntary activity in which people engage as citizens to affect the political domain, although it might not occur within that domain per se or be directed to politics in the strict institutional sense (DETH, 2014). Scholars in the field disagree -according to their theoretical perspective -about the extent to which citizen participation in the political regime is beneficial. For instance, participatory theorists claim that political participation should be widespread in democracies, while institutionalist theorists argue that it would be better for democracy to have rather limited channels for participation available to the general public.
The participatory theory of democracy has for a long time advocated in favor of expanding people's influence on politics (FISHKIN, 2015;PATEMAN, 1992). Such influence should be promoted mostly by advancing a myriad of arenas for political participation, including conventional forms of participation, such as voting in elections or engaging in political parties' activities, and unconventional ones, such T Diego Moraes Silva, Thaise Kemer (2022) 16 (2) e0006 -3/31 as joining civil society associations or public demonstrations (VERBA and NIE, 1972). More recently, a new arena for political participation has attracted much of scholars' attention due to its pervasiveness in contemporary societies. We refer to digital participation, which gained momentum after the emergence and diffusion of the new information and communication technologies in the last decades of the 20 th century. Currently, no political scientist could ignore the role of technology in shaping world politics via digital transformation, although we still cannot be sure if digital technologies are best characterized by the idea of a 'new era' or a 'false hope' for democracy (DIAMOND, 2010).
Either way, the rise of digital democracy is currently an unavoidable reality.
One of its basic tenets is precisely to foster participation within civil society (GOMES, 2016). However, it is still not clear if this new political arena complements other arenas or grows at their expense. As technological change and political modernization go hand-in-hand (BIJKER, 2006), technologically-advanced societies transfer innovative and knowledgeable outlooks to politics, favoring democratic institutions and forms of participation that are more dynamic and demanding (INGLEHART and WELZEL, 2009). Nevertheless, users of mass media technologies might also be seen as potentially more isolated, apathetic, and distant from community life (PUTNAM 2015).
At the same time, a growing number of governments have been using digital tactics to harass opponents and control digital media in their countries, raising concerns over the use of technology to authoritarian ends. 1 After all, scholars have not yet reached a consensus about whether the net result of such technological change has been mostly positive or negative for democracy (FUKUYAMA, 2020). In particular, most scholars have not properly addressed the relationship between the new digital technologies available for political participation and the democratic culture of its users, although a few recent studies have shown that different political cultures -both in consolidated and in new democracies -affect the use of these technologies by the citizenry differently (VACCARI and VALERIANI, 2018).
______________________________________________________________________________________________ 1 For instance, the 'Election watch for the digital age', a new research initiative led by the Freedom House, has been investigating the interplay between digital platforms and election integrity. After analyzing 40 elections and referendums between 2018 and 2020, it found that 88% of them had been subject to some sort of digital interference (see ˂https://freedomhouse.org/report/electionwatch-digital-age˃). In addition to this Introduction, the article is structured as follows: Section 02 reviews the literature on the intersection of political participation and digital democracy; Section 03 details the data and methods used in this study; Section 04 presents the results and discusses their main implications; Section 05 concludes by providing some final remarks and pointing out the limitations of our analysis and possible avenues for future research.

Political participation and digital democracy
There is wide consensus in political science that no regime could be a democracy without guaranteeing formal rights for all its adult citizens to participate in politics (DIAMOND and MORLINO, 2017). The basic premise here is that people who are affected by political decisions must, to some extent, be a part of the decision-making process. However, how extensive participation should be is precisely the point of disagreement among two general perspectives in political theory. On the one hand, scholars associated with a more minimalist or institutionalist view of democracy argue that the political participation of ordinary citizens should be minimal, confined mostly to suffrage and electoral activities (SCHUMPETER, 2008). From this perspective, limited participation -and even public apathy -plays a key role in buffering the shocks in public opinion, which is usually uninformed and uninterested (POSNER, 2005). Therefore, politics should be left to the political elites and representative institutions who would remove the risk of tyranny of the majority by making the political process less passionate and more professional (HAMILTON, MADISON, and JAY, 2011).
On the other hand, scholars associated with a more maximalist or participatory view of democracy claim that the minimalist understanding of ordinary citizens is a sort of self-fulfilling prophecy: by refusing to foster more engaged citizens, the institutionalist approach eventually produces apathetic individuals who are uninterested in politics. The premise of the participatory approach, in contrast, is that a participative system eventually becomes selfsustainable, as the qualities required to be a good citizen are those promoted by the participatory process itself (PATEMAN, 1992). In this vein, the core argument for a participatory democracy relies on the idea that the act of participating might have an "educational function", insofar as people who participate politically learn to be good citizens by enhancing their sense of efficacy, getting more information on public issues, increasing their tolerance toward opinions other than their own, hence acquiring some form of "public spirit" (FISHKIN, 2015).
Although the minimalist/institutionalist approach has been widely acknowledged as the contemporary mainstream view on democracy (GAMA NETO, 2011;PERES, 2008), it has been challenged by the growing crises of democracy around the globe, which are mostly institutional crises of representative democracy (MOISÉS, 2019;MOISÉS and WEFFORT, 2020;NOGUEIRA, 2014). As expected, the participatory approach inspires many of the The issues of political polarization and fake news in the digital world are particularly concerning for democracy in the contemporaneity. Bakshy et al. (2015) identified significant polarization on Facebook, where liberal and conservative individuals share "hard" stories mainly in groups of ideologically -aligned users, thus reinforcing their preexisting political inclinations. Lazer et al. (2018) also found this "echo chamber" effect in social media environments, where In his study of this issue using survey data in the United States, Dalton (2017) found that, while online participation has, to some extent, replaced offline participation, the former contributes to an overall increase in political participation.
However, since individuals who are best educated and resourced tend to be more Going back to the institutionalists' concerns regarding people's broad participation in politics, we notice that the fear of tyrannic behavior from the masses was one of the main arguments used to limit political participation (HAMILTON, MADISON, and JAY, 2011). Therefore, it could be that scholars from the skeptical perspective are applying to digital democracy the same reasoning that long ago led institutionalist theorists to fear widespread political participation. However, it would be a mistake to attribute this digital "participatory despotism" behavior to every citizen or nation who is more engaged virtually due to the new digital technologies available. To make sense of the thin line between democratic and authoritarian digital participation, we propose to examine the cultural context in which engaged citizens are immersed, i.e., their political culture.
The modern political culture approach was inaugurated in the post-war period by Gabriel Almond and Sidney Verba's seminal book The Civic Culture: Political Attitudes and Democracy in Five Nations. In this book, the authors conceive of political culture as a set of cognitive, affective, and evaluative orientations that citizens have toward social and political objects (ALMOND and VERBA, 1989). To the extent that such orientations make the citizenry prone to engage actively in the political life and adhere to democratic norms and values of moderation, trust, tolerance, and the like, a so-called "civic culture" (or "democratic culture", as some contemporary authors would prefer to call it) would prevail in a given context (ALMOND, 1989;DIAMOND, 1994DIAMOND, , 2015. Moreover, according to this body of literature, the proper functioning and survival of democracy at its macro-level is deeply linked to the values and orientations at the micro-level, in other words, a democratic culture is an important condition to a sustainable democratic consolidation (INGLEHART and WELZEL, 2009). Hence the need to examine the political culture of digital activists and assess whether their values and behaviors are consistent with those expected from democratic citizens; in doing so, we may identify the actual potential of digital technologies in helping save contemporary democracy.
In this vein, Vaccari and Valeriani (2018) found that institutional legacies and political culture affect the extent to which active citizens take up new opportunities for informal political conversation in the digital world, as political discussion in social media is more strongly associated with participation -be it institutional or extra-institutional -in established democracies than in "third wave" democracies.
Similarly, Mounk (2018) argues that the internet and, particularly, social media have only had such a corrosive impact on democracy around the world because the moral foundations of our political system are getting more and more fragile, even in countries where democracy was supposed to be consolidated, so that anyone who wants to contribute to revitalizing democracy will have to help rebuild it on more stable cultural norms and values.
This article focuses precisely on the relationship between digital political participation and democratic culture to answer the following research questions: Therefore, the hypotheses to be tested are the following: H1 -Digital-only activists are more culturally democratic than non-activists.
H2 -Wide-spectrum activists are more culturally democratic than digital-only activists.
We applied a PCA technique to reduce the number of participatory arenas to a few dimensions and use them as dependent variables in regression models in which the predictors are the democratic culture variables and socio-demographic controls.   (Table A3 in the Appendix).   in any arena -they are non-activists. The proportion of individuals who are activists in only one arena is 14,53%, and only about 2% of the individuals are activists in four or all five arenas. These numbers show how scarce political participation still is in the context of a major Brazilian city, although it has probably increased in the last years, as since 2013 we have seen frequent large street protests across the country, especially in large metropolises.

Results and discussion
As for the data on arena-specific activists in Table 2  Interestingly, all factor loadings of PC1 are positive and higher than 0,4, which shows a high correlation between the different participatory arenas and a significant number of individuals who are politically active in several of these arenas. On the other hand, only three factor loadings of PC2 exceed the threshold of 0,4: electoral and institutional participation (which are negative) and digital participation (which is positive). Such results indicate that there is also a significant   Nevertheless, when compared with non-activists, the only statistically significant mean differences for digital-only activists are being more educated, employed, and richer. Interestingly, the socio-demographics of digital-only activists are closer to those of non-activists than the critics of digital political participation would expect and more distant from the traits of wide-spectrum activists. Indeed, digital-only activists seem to be situated in the middle of two categories, having slightly more resources than the individuals who do not participate politically, but also a slightly lower social status than those who participate on several fronts.   These results suggest that despite the significant democratic culture gap that still exists between digital-only activists and wide-spectrum activists (the former being considerably less advanced than the latter), the exclusively-digital form of political participation seems to be an important first step in enhancing the civic skills and democratic values of individuals, since digital-only activists are significantly more culturally democratic than non-activists, even though both groups have fairly similar socio-demographic attributes. As our final statistical procedure, we did a regression analysis with multiple imputations to estimate the strength of the ass ociation between political participation and democratic culture, differentiating between a more general political participation and a more specific digital participation. For this purpose, we used the two principal components retained from the PCA presented in Table 03. As mentioned above, PC1 is characterized by high factor loadings across all participation arenas, and it represents a variable for a more general participation (named "General Participation"). PC2, in turn, is characterized by a high and positive factor loading only for digital political participation, and by substantive factor loadings -although negative -for electoral and institutional participation; PC2 thus represents a variable for digital-only participation (named "Digital Participation"). With these two components as dependent variables, we ran multiple linear regression models using the democratic culture variables from Table 05 as independent variables and the socio-demographic variables from Table   04 as controls. We also included regional dummies for each of the eight administrative regions to control for regional fixed effects and account for the city's territorial heterogeneity; finally, we estimated robust standard errors to avoid heteroskedasticity issues.
Graph 01 presents the main results from the regression models limited to the coefficients and confidence intervals of the independent variables (full results are presented in Table A2 in the Appendix). We found a significant association between general participation and various democratic traits like willingness to get politically informed, openness to dialogue, openness to change, and rejection of democracy relativization. The coefficients obtained are significantly high, as well as their statistical significance. Digital participation, on the other hand, is significantly associated with political knowledge, higher diversity of information sources, and openness to dialogue; however, except for diversity of information sources, the coefficients -and their statistical significance -are considerably low. These results demonstrate that the relationship between digital participation and democratic traits is rather weak in a controlled empirical test. Nevertheless, the figures point in the same direction as the one previously found in the descriptive statistics. The combination of the regression results with the mean difference tests thus gives us the confidence to confirm our two hypotheses: digital-only participation is associated with a slight improvement in the democratic culture of the citizenry (H1), while a broader form of political participation correlates more significantly and deeply with democratic values and attitudes (H2).

Final remarks
This article addressed the relationship between digital political participation and democratic culture among citizens in the city of São Paulo, particularly digitalonly activists. As the new information and communication technologies became so pervasive, and therefore so impactful in shaping politics, digital political participation became an essential research topic. Aware of the importance of being cautious with generalizations, we contend that the case of São Paulo, a very sociodemographically complex environment, might shed some light on the interplay between digital technologies and political culture. Moreover, studying the case of São Paulo is innovative as it allows for the issue of digital participation to be addressed from the perspective of a major city in a developing country, one that combines a young democracy with a tradition of authoritarian movements throughout its history, something typically overlooked by the mainstream literature focusing on the developed world.
Our study reveals that political participation is scarce in the city, given that 74,03% of the interviewees in our sample do not participate regularly in any Focusing on the digital-only activists, data analysis on socio-demographic and democratic culture variables also provided some interesting findings. Results showed that digital-only activists share a similar proportion of socio-demographic traits with non-activists and wide-spectrum activists, thus positioning them in the middle of these two categories. Since one of the main concerns in the literature refers to the socio-economic gap between activists (typically more privileged in terms of resources and status) and non-activists (typically more marginalized, lacking both resources and status), this result suggests that the digital arena might indeed be an important channel for the inclusion of minority groups in politics.
As for the democratic culture variables, our study confirms that digital-only activists share 01. lower levels of democratic culture traits than widespectrum activists but 02. higher levels of democratic culture traits than nonactivists. Therefore, hypotheses H1 and H2 were confirmed by the multivariate analysis applied in this study. Since digital-only activists and non-activists have fairly similar socio-demographic attributes, these results suggest that the digital arena may be an important channel for fostering democratic culture. However, as shown by the far better results obtained by wide-spectrum activists, this channel is possibly only a first step toward improving our democratic culture, not the ultimate means to achieve sustainable democracy.
These results also open possibilities for future investigations, for instance, to address: 01. the relationship between different forms of digital political participation and specific variables of democratic culture that we found to be  How often the individual participates in electoral activities, such as rallies, electoral debates, caucuses, political parties' meetings and conventions, among others (1 -Never; 2 -Rarely; 3 -Sometimes; 4 -Often; 5 -Always)

Institutional participation
How often the individual participates in institutional activities such as plebiscites, referendums, popular consultations, public hearings, among others (1 -Never; 2 -Rarely; 3 -Sometimes; 4 -Often; 5 -Always) Associative participation How often the individual participates in associative activities through civil society organizations such as neighborhood associations, trade unions, social movements, NGOs, among others (1 -Never; 2 -Rarely; 3 -Sometimes; 4 -Often; 5 -Always) Demonstrative participation How often the individual participates in demonstrative activities such as public demonstrations, protests, marches, caravans, among others (1 -Never; 2 -Rarely; 3 -Sometimes; 4 -Often; 5 -Always) Digital participation How often the individual participates in digital activities such as political engagement on social media, online discussion forums, e-government consultations and polls, among others (1 -Never; 2 -Rarely; 3 -Sometimes; 4 -Often; 5 -Always) Political knowledge about the municipal government Whether the individual knows who is the main responsible for formulating the government plan and the budget plan for the city (1 -Does not know/Gave the wrong answer; 2 -Knows about the existence, but does not recall the name; 3 -Mentioned the mayor or City Hall) Political knowledge about the City Council Whether the individual knows who is the main responsible for passing laws in the city (1 -Does not know/Gave the wrong answer; 2 -Knows about the existence, but does not recall the name; 3 -Mentioned the city councilors or the City Council) Political knowledge about the judiciary branch Whether the individual knows who is the main responsible for guaranteeing the individual, collective, and social rights and for solving conflicts between citizens, entities, and the state (1 -Does not know/Gave the wrong answer; 2 -Knows about the existence, but does not recall the name; 3 -Mentioned judges or the judiciary branch) Political knowledge about the Public Prosecution Service Whether the individual knows who is the main responsible for filling charges to defend social interests and ensure public heritage protection (1 -Does not know/Gave the wrong answer; 2 -Knows about the existence, but does not recall the name; 3 -Mentioned public prosecutors or the Public Prosecution Service) Political knowledge about the Court of Accounts Whether the individual knows who is the main responsible for overseeing the use of public money and approving public expenditures (1 -Does not know/Gave the wrong answer; 2 -Knows about the existence, but does not recall the name; 3 -Mentioned the ministers of the Court of Accounts or the Court of Accounts) Political knowledge about the mechanisms by which the people wield influence Whether the individual knows the three main mechanisms by which the people can influence the government, i.e., Access to Information Law, Popular Initiative Law, Popular Action Law (1 -Does not know/Gave the wrong answer; 2 -Knows about the existence, but does not recall the name of any of the mechanisms; 3 -Mentioned at least one of the three mechanisms) Willingness to get politically informed How often the individual is informed about political issues, especially about his/her city (1 -Never; 2 -Rarely; 3 -Sometimes; 4 -Often; 5 -Always) Legitimacy of the law Degree to which the individual agrees with the statement: "It is important to comply with the laws and the government regardless of whether the politicians in power are those I voted for" (1 -completely disagrees; 2partially disagrees; 3 -partially agrees; 4 -completely agrees)

Rejection of democracy relativization
Degree to which the individual disagrees with the statement "When there is a difficult situation, it doesn't matter if the government overrides the laws, Congress, and institutions in order to solve problems" (1 -completely agrees; 2 -partially agrees; 3 -partially disagrees; 4 -completely disagrees) Preference for democracy Degree to which the individual agrees with the statement "Democracy is preferable to any other form of government, regardless of the circumstances" (1 -completely disagrees; 2 -partially disagrees; 3 -partially agrees; 4completely agrees) Sex (Male) Whether the individual is female or male (0 -Female; 1 -Male) Age Current age of the individual in years Educational level Which is the highest education level attained by the individual (1 -No formal education or incomplete primary school; 2 -Complete primary school or incomplete secondary school; 3 -Complete secondary school or incomplete higher education; 4 -Complete higher education and more) Occupational status Whether the individual worked in the last 7 days as an employee, selfemployed, employer, or unpaid worker (0 -Unemployed; 1 -Employed) Income bracket Which is the average nominal income bracket in the individual's household (0 -Up to 1 minimum wage; 1 -From 1 to 2 minimum wages; 2 -From 2 to 5 minimum wages; 3 -From 5 to 10 minimum wages; 4 -from 10 to 20 minimum wages; 5 -More than 20 minimum wages)

Religiosity (nonreligious)
Whether the individual belongs to any religion or religious group (0 -Belongs to some religion; 1 -Does not belong to any religion) Marital status (