“As we were saying in our last class” was the sentence with which Gaetano Salvemini, political scientist and historian from southern Italy, opened his class at the University of Florence on November 15, 1949, after 20 years in exile, after having his Italian citizenship taken from him, after seeing all his family and many of his students dead, after being a political prisoner of the Italian fascism and seeing the world change substantially and unrecognizably. It was also like this that Filippo Turati, socialist politician and journalist, started his first lecture after imprisonment, and how Norberto Bobbio addressed his students in Rome after a long exile in New York. It was how the humanist and Spanish friar Luis de León started his class after being freed from the prison in which the Inquisition had put him, in the sixteenth century, and how Gustave Cohen, the famous medievalist, also started his class upon returning to the Sorbonne after having escaped the Nazi extermination camp.
It is also like this that Rui Tavares presents us to these figures in his brilliant podcast “Agora, agora e mais agora”. Rui Tavares himself says it, in the hope of being able to say it again at the end of the Covid-19 pandemic, as if challenging reality, as if laughing at the tremendous misfortune he experienced. I also hope to say it one day, opening class for students who will no longer live the third wave of autocratization or a world going through the challenge to the ruling global order. Students who will live in times of political and social stability, and who will live less afraid of the state of things. Together we will ignore, in this multigenerational joke, the risks we lived, and together we will study this dark period as a stumble of world history, which no longer exists.
I long for this moment, but I know that, when it comes to politics, nothing comes without effort, without action, and without agency. For this reason, I do my part, which in these parts manifests itself in the texts I write and in the ideas I seek to raise. I then usually point out problems in governance, analyze causal chains to explain why events occur as they occur, and hopefully, provide to those who vote, participate, or decide, with facts and understanding sufficient for concrete and effective action. In these next two columns, however, I will deal with these things only to the extent necessary to introduce suggestions about what can and what should be done. I thus contribute, I hope, in an even more concrete way.
I shall focus my attention on the fields of democratic resilience and resistance – respectively how to ensure that democracies can survive the blows they receive, perhaps even becoming stronger and more resilient with them, and how to offer a democratizing counter-force, or as a colleague, a former Hungarian politician, says, how “not only to take hits, but hit back.” There are thousands of ways to think about resilience and resistance, but here I propose only two: how to strengthen the institutions of democracy, and how to deal with voters who, ironically, vote for its end. In this column I speak of the first; in the next, titled “What to do when they vote against your right to vote”, we will speak of the second.
Institutions are like dams against strong waves
The first and second waves came strong and collided against the sands of a violent world. In these waves of autocratization, autocrats took power through military coups, foreign invasions, or the famous self-coups – in which leaders are democratically elected but soon afterwards abolish democratic institutions, close parliaments, and stop elections. The third wave came differently, hard to see, and therefore hard to stop. In the third wave – the wave we are drowning in today as I write – leaders take power gradually. They gradually concentrate powers around the executive – a tactic called in academic literature “executive aggrandizement” –, and then propose “promissory coups” – which means saying that they begin to dismantle institutions such as free media, the judiciary, NGOs, civil society, and the opposition under the pretext of defending democracy. All done legally and with increments that are not so easy to notice – an NGO hunted here, a journalist made into a joke there, and so on.
When the process gets more advanced, autocrats begin to change election rules, altering norms and the number of administrative units, changing the place where people vote, painting opponents as enemies, etc. These details seem silly, too small to incite a popular response, sometimes too technical for ordinary citizens to understand. When the opposition draws attention to them, it is usually framed as exaggerated. This is how autocratization goes unnoticed.
Here is the first element of concern: autocrats learned from the failures of the first two waves, and those of the third wave continue to learn from each other. This is not to say that the most radical forms of authoritarianism no longer exist – see Madagascar, for example, which recently suffered a military coup à la first and second waves – but only that they are no longer the most common form of authoritarianism in the current world. The far right, for example, dominates the political game on social media like no one else, pointing out simplistic answers for complex problems and offering someone to blame when life does not bring what is expected of it. They are agile and proactive, while the center, the old right, and the democratic left continue to preach moderation and caution, or worse, just shout “fascists!” as if that were enough. Politicians with old strategies from a world in which information came from printed newspapers, reactive, will never obtain the success of far-right politicians. Democrats also need to learn from each other and from the past. They also need to be proactive, learning to play the social media game, learning how to present their messages in a clear, engaging, and mobilizing way. There have already been changes in this in Brazil, like the working group at the Ministry of Human Rights and Citizenship which includes names such as Felipe Neto and Manuela d’Ávila, or the CRIA G20 with Nath Finanças and Licypriya Kangujam, but those are nowhere near enough.
Besides social media, there are also other areas in which institutional learning is vital, such as the intra-regime support network and institutional responses to attacks against democracy. As I described in more detail in the column “Interdependence and its empty promises”, autocrats help each other, sending troops, information, technology, logistics, money, weapons, and whatever else is necessary to ensure that other autocrats can not only maintain their regimes, but prosper. However, the same is not done today by democrats – or at least not to the same intensity. Democracies need to unite, cooperate with each other, and understand that the success of one, against autocracies, is the success of all.
As Applebaum proposed, not only is an international democratic league necessary, but also a domestic league, composed of lawyers, civil servants, and activists with practical experience in dealing with autocrats. These leagues need to be supported by military coalitions and a shared intelligence network that anticipates and stops violent state actions against civilians and shares information with said leagues. Surveillance systems need to be implemented to ensure sanctions against autocrats are effective, and a task force active in combating Fake News needs to be created. And when a country acts progressively in defending democracy, it is necessary that this action be protected by others, and learned as an example to be applied. This is the case with the conviction of former president Bolsonaro by the Supreme Court, a democratic measure that should serve as an example for the United States.
As Riedl, McCoy, Roberts, and Somer demonstrate in their seminal research, “executive aggrandizement” can occur essentially in four ways institutionally: legislative capture, plebiscitary bypass, override of other institutions, and elite collusion. Of course, once it begins by any of these, it is common that the others appear. Let us address what to do in each of them.
Legislative Capture
Legislative capture is the process in which the legislative power is given, formally or informally, to the executive. It usually happens in the case of supermajorities in congresses or parliaments (when the leader’s party is the large majority in these houses and therefore votes according to what the leader prefers), and it is generally presented to the people as the ultimate representation of democracy (after all, they were elected by the people) and representatives of the people’s will. When they persecute judges, for example, they argue that these were not elected, and therefore are less democratic, ignoring the role of judges in democracy and that, if they were elected, they would not be able to perform their function with parsimony.
Here it is crucial to understand the source of the problem: without a supermajority, any government needs to come to an agreement with other parties, form coalitions, and offer positions, which ensures that more ideas will be represented in the government. Without the need for a coalition, only one group will be represented in the government – the voters of the party leading the supermajority. It is a distorted view of democracy that understands that the interests of one group in particular are more important than those of the rest of the country simply because they are the majority. In a full democracy, electoral and social minorities also need a voice and to be protected, and coalitions are the way protection is conducted.
It is in legislative capture that narrative wars are usually seen (the opposition claims that the leader is a threat to democracy, while the leader claims the threat is another and justifies the centralization of power in themselves: the history of corruption, fascism, communism, the wave of immigrants, homosexuals, the elites, any “other” that can be vilified).
The speed with which autocratization occurs in this case depends on the degree of personalization of politics (the leader who creates a new movement – like Hugo Chávez, for example – usually autocratizes the political system faster than one who takes over a pre-existing party or movement – like Trump or Erdoğan).
As in the case of legislative capture the opposition has been vilified, it may need new names. Here there is a crucial difference between the presidential and parliamentary systems. In the presidential system, the president naturally concentrates greater political personalization – since voters voted for them to occupy the office, and not their party, as in the parliamentary case – it is common for the opposition to remain strong even after some legislative capture. That is why Lula can oppose Bolsonaro, while in Hungary there are practically no strong political names sufficient to compete electorally with Viktor Orban. On the other hand, the presidential system also makes it more difficult for new names to enter politics, naturally making periodic government change harder – an important feature of democracy. Without new names, we can think of new coalitions among old names. A good example was Lula’s choice of vice president, uniting the old right and the old left in a democratic league.
Besides the opposition, an important path to escape autocratization during legislative capture is the electoral path. Since they use their broad majority or legislative control as a narrative instrument, leaders in this situation need to maintain the system, at least to some degree. This is how Poland managed to vote an autocratic party – Law and Justice (PiS) – into power in 2015, and then vote it out in 2023.
In fact, the opposition depends on the electorate, since alone, its chances of stopping such a contentious movement are small. In successful cases in Brazil, Poland, and Moldova, where autocrats were voted into power and then out, the union between electoral efforts – with support from social movements and civil society as a whole – and the opposition ensured such success. Protests encouraged the opposition to seek coalitions and showed them that the power calculation tilted in their favor (here I refer to the calculation made by the opposition in which the risk of opposing the government is measured against the opportunity to succeed).
Here I only deal with the electorate as opposition support and institutional responses, but I do not go deeper into the topic. For this, read the next column titled “What to do when they vote against your right to vote.”
Plebiscitary Bypass
When the autocrat is very popular but cannot achieve a substantial majority in the legislature, they may resort to plebiscites and referenda. Letting the people decide directly which policies should or should not be executed may be democratic, but the autocrat usually uses these instruments as a way to avoid having their bills vetoed by the legislature – also elected by the people – or to avoid using executive decrees – less popular and with expiration dates.
This was how Hugo Chávez used referenda to approve a new constitution that expanded his powers, and how Nicolás Maduro continually ignored the opposition-majority legislature. This was also how Rafael Correa transformed Ecuador into a low-quality delegative democracy, using the direct voting instrument to pass a package of multiple laws at once (known as omnibus).
Plebiscitary bypass is more common in presidential regimes, which exploit political, social, or economic crises to sow a cult of personality and distrust in the political class. Even though it is more common in presidential regimes, it can also occur in parliamentary ones.
Given the difficulty of perceiving plebiscites and referenda as dictatorial weapons, traditional media occupies a vital position in resistance attempts. It will be through reliable and quality media – which here does not mean media that agrees with your worldview, but simply reports events reliably, following a standard and journalism ethics – that people can perceive the instrumentalization of plebiscites and referenda. Therefore, it is vital for democracies to maintain traditional media.
Courts are another element of democratic resistance in this case, contesting not the plebiscite and referendum instruments themselves, but their undemocratic use. Autocrats often tend to attack the courts, accusing them of judicial activism or silencing the will of the people. Therefore, it is vital that there be greater trust in the courts, a role that again falls under the jurisdiction of the media.
Curiously, there is a third form of resistance rarely mentioned: the plebiscites and referenda themselves. This was the case in Ecuador, for example, when the population voted against indefinite re-elections.
Override of Other Institutions
Also known as autogolpe, this form of “executive aggrandizement” usually occurs through the declaration of a “State of Emergency.” Still, it can also occur gradually, with an increase in executive decrees, the appointment of supporters to high government positions, misuse of the State’s coercive powers to cause public panic, and other forms as well. It is therefore the most extra-institutional, most openly dictatorial way of aggrandizing the executive.
When it occurs in the third wave of autocratization, it is a clear sign of weakness, since the autocrat neither counts on broad popular support to hide behind referenda, nor has a legislative majority to pass reforms without depending on the people. And it is precisely for being a sign of political weakness that this form of “executive aggrandizement” only succeeds if the opposition is divided or incapacitated and if the armed forces allow the coup.
This was the case, for example, in Tunisia under Kais Saied, where the secular parties refused to form a coalition with the Islamic party Ennahda, the people rejected the political class, the courts had little power, and the media was not free. It was not the case, thanks to strong courts and a unified opposition, in Brazil.
Given its extra-institutionality, emergency character, and the leader’s weakness, this form is the most dangerous in terms of reaction time. Extra-institutional actions like demonstrations and popular uprisings are especially recommended at this stage, since a response of equal strength will have great chances of reversing the dictatorial scenario, but only before the leader achieves success – a short window of opportunity.
Analogously, the opposition also has at this moment a short window of opportunity to unite and act more extremely, with impeachment requests, blocking the executive, or other forms. The media occupies an important role in providing information, but without compromising its journalistic integrity.
Elite Collusion
This last form is distinct from the others because it is not a direct form of “executive aggrandizement,” but rather an instrument to achieve the other forms. It concerns the support that certain elites give to autocrats – see, for example, the macabre dinners and speeches supporting Trump given by Mark Zuckerberg, Jensen Huang, Jeff Bezos, and company. Due to the economic power they carry, and unfortunately the informational power of big techs, elite collusion often confers special power to the autocrat.
Here, there will hardly be a better suggestion than the one given by Anne Applebaum. According to her, economic transactions need to be more transparent, meaning owners of financial enterprises in general and their stakeholders need to be named in contracts visible to the public. Furthermore, the political consequences of such investments also need to be clear and objectively defined. The sending of dividends to tax havens should be prohibited, and practicing lawyers prohibited by law from interacting with such localities and governments.
Applebaum makes legal suggestions, but it is also important to think about the reach of non-legal measures, such as greater technological sovereignty – so that democracies are not so dependent on big tech – or even greater individual control and knowledge of one’s own information. Redistributive justice could also be considered, since it does not seem fair that a few individuals have so much control over the fate of entire nations, but the debate ends up taking on distinct contours and being co-opted by other agendas, losing credibility.
Final Considerations
I hope I have managed to present a little of what is discussed today about the topic of democratic resistance and resilience, proposing concrete solutions and pointing out useful distinctions. I hope I have done my part by addressing the institutional forms of executive aggrandizement, although I suspect that, without the next column on the electorate, this one inevitably remains incomplete.
I hope we can get through this turbulent period in our history and emerge better on the other side. And if you, reader, are one of those responsible for this change, I wish you freedom, since resistance will probably not bring you peace.
References
Applebaum, Anne. (2024). Autocracy, Inc.: The Dictators Who Want To Run The World. Random House: USA.
Bermeo, N. (2016). On democratic backsliding. Journal of Democracy, 27(1), 5–19.
Bueno de Mesquita, B., Smith, A., Siverson, R. M., & Morrow, J. D. (2003). The logic of political survival. MIT Press.
Cravo, A. (2023, 22 de fevereiro). Com Felipe Neto e Manuela D’Ávila, governo cria grupo para combater discurso de ódio e extremismo. O Globo.
Encarnación, O. G. (2025, January 16). A tale of two caudillos. Foreign Affairs.
Ferreira, E. (2025, October 17). The far-right tells a story people buy — so why don’t progressives? EUobserver.
Gamboa, L. (2024). Plebiscitary override in Venezuela: Erosion of democracy and deepening authoritarianism. The ANNALS of the American Academy of Political and Social Science, 712(1), 124–136.
Kisilowski, M. (2023). How Poland won back its democracy. Project Syndicate.
Linz, J. J. (1990). The Perils of Presidentialism. Journal of Democracy, 1(1), 51–69.
Lührmann, A., & Lindberg, S. I. (2019). A third wave of autocratization is here: What is new about it? Democratization, 26(7), 1095–1113.
Merkel, W., & Lührmann, A. (2021). Resilience of democracies: Responses to crises and challenges. Democratization, 28(5), 869–889.
Molinero Junior, G. (2025, 30 de maio). Interdependência e suas promessas vazias. DPolitik.
Moncagatta, P., & Pazmiño, M. (2025). A tale of a failed recovery: Ecuador’s democratic stagnation. The ANNALS of the American Academy of Political and Social Science, 712(1), 137–151.
Riedl, R. B., McCoy, J., Roberts, K., & Somer, M. (2025, March 21). Pathways of democratic backsliding, resistance, and (partial) recoveries. The ANNALS of the American Academy of Political and Social Science, 712(1), 8–31.
Savage, R. (2025, October 14). Elite military unit says it has seized power in Madagascar. The
Guardian. Retrieved from https://www.theguardian.com/world/2025/oct/14/elite-military-unit-says-it-has-seized-power-in-madagascar.
Schwartz, M., & Montague, Z. (2025, March 2). Federal judges are ordering Trump to slow down. Will he listen? The New York Times.
Secretaria de Comunicação Social. (2024, 13 de novembro). CRIA G20 traz Felipe Neto, Nath Finanças, Eduardo Moreira, Janja e convidados internacionais para discutir fome, economia e clima. gov.br.
Tavares, R. (2020). Agora, agora e mais agora [Podcast]. Público.
