Tomás Gutiérrez Alea is widely regarded as the best film-maker in the Cuban film industry. Without overlooking the collective nature of film production in Cuba, as well as the extraordinary contributions of other talented directors, the national and international success of Gutiérrez Alea's films has increased his reputation within Cuba and abroad. Gutiérrez Alea was one of the founders of ICAIC, director of the first cinematographic production of the revolution (the documentary Esta tierra es nuestra, 1959) and co-ordinator of one of three Grupos de Creación launched by ICAIC in 1988 to support young Cuban writers and film-makers. In 1993 he directed Fresa y chocolate, a film that certainly brought to light, both within Cuba and abroad, the repressive and homophobic policies implemented during the Fidel Castro regime.

Gutiérrez Alea was one of those intellectuals who decided to stay in Cuba and contribute to the development of the revolutionary project from within the country, as the result of an endless negotiation with the politicians who controlled cultural institutions in Cuba. Gutiérrez Alea's work can therefore be taken as an indication of the relationship between politics and part of the film industry under the Cuban regime. ICAIC supported Gutiérrez Alea's projects, which shows his commitment to the politics of the regime. Nevertheless, his declarations in the El País during the promotion of Fresa y chocolate demonstrate his frustration and his lack of hope for the present – as it was then – and the future of Cuba (Salas and García 1994).

In spite of his success abroad, Gutiérrez Alea was not unsympathetic to the effects that the collapse of socialism in Europe and the so-called Periodo Especial were having on the Cuban population during the first years of the 1990s. Writing to Mirta Ibarra's son in October 1991, for instance, Gutiérrez Alea referred to the impossibility of human beings sorting out their problems in clear terms when ‘la sociedad en la que vives no funciona bien’ (Ibarra 2008: 206): ‘En medio de tanta injusticia, de tantos sinsabores, de tanto sufrimiento, de tanta tristeza’, he concludes, ‘siempre llega un momento en que uno se pregunta qué sentido tiene la vida’ (Ibarra 2008: 206).

Reinaldo Arenas (1943–1990) is considered to be one of the most extreme nonconformist authors in Cuban literature. It is particularly relevant to approach his novels, articles and short stories in the light of the circumstances of his life. His creative life was dominated by tension with the Cuban authorities and the struggle to preserve his sexual and intellectual identities. Arenas often fused autobiographical material with a political denunciation of Fidel Castro's regime in his fictional works, articles and personal letters. An extraordinary writer, Arenas managed to leave behind a legacy describing his personal struggle to overcome his invisibility in Cuba. Since his well-publicised suicide in 1990 in New York, Arenas's discourse has become one of the most important points of reference for the younger generations of Cuban homosexuals in exile.

In reference to visibility and invisibility, the aim of this chapter is to show that Arenas passed through three stages following his arrival in Havana in 1960. His intermittent presence in the Cuban cultural panorama in the second half of the 1960s is referred to as his ‘penumbra’. His mythical ubiquity was perceived as a destabilising element for the ethos of the revolution. Arenas's ability to move within the margins of visibility in the years before his imprisonment has fascinated critics. Barnet and Diego wrote positive reviews of Arenas's literary works in La Gaceta de Cuba in 1967 (Barnet 1967; Diego 1967: 162–6). A dramatic adaptation of Celestino antes del alba was broadcast on national TV. Arenas read two significant papers at Havana University that had been published in the literary magazines Casa de las Américas and La Gaceta de Cuba (Arenas 1968a: 134–8; Arenas 1968b: 13–17). Arenas's popularity in intellectual circles in France and the US, however, provoked a deep sense of unease among the leaders of UNEAC who refused to publish his work in Cuba (Arenas 1992: 43). Arenas's penumbra status turned into invisibility during the 1970s as a result of his imprisonment for a year and a half and the definitive censorship of his literary work from then onwards. Subsequently, during the 1980s he was both visible and invisible: visible to the intelligentsia in Cuba, the US and Europe, but invisible to the Cuban people. Knowing that Cubans were not allowed to read his books enhanced his sense of frustration and nostalgia for his homeland.

This book demonstrates the relevance of homosexuality and homophobia in the development of a social ethos in Cuba through the twentieth century until the mid-1990s. The issue of homosexuality and homophobia was central to the design of a Cuban national identity after the departure of the Spanish, as the discussions about the Cuban male by Enrique José Varona, Benjamín Céspedes, Pedro Guiralt and José Martí show. The issue was equally central to the portrait of the Cuban nation after Fidel Castro came to power, as the number of laws created from the mid-1960s to prevent homosexuality and control homosexuals in Cuba shows. There have been constant references to the issue in political speeches and in literature. During the revolutionary period, Cuban cultural institutions (mainly ICAIC and UNEAC) promoted those artists who, they felt, supported the revolutionary project (for example Gutiérrez Alea) and silenced those who evidenced disappointment with the measures developed by the government in the field of culture and sexual freedom (for example Arenas).

Identifying the ideological basis on which the ICAIC and the UNEAC censored some Cuban authors throughout the first three decades of the revolutionary period is essential to understanding the often complex and veiled messages behind authors’ works throughout the twentieth century. The work of the Cuban homosexual dissidents (Néstor Almendros, José Lezama Lima, Virgilio Piñera and Roger Salas, among others) provides evidence of the social and institutional rejection that turned them into social marginals. Their work ultimately opened up the debate about censorship, invisibility, stigma, social injustice and social displacement on the island.

A detailed analysis of the narrative of Reinaldo Arenas and the ideological ethos behind Fresa y chocolate (1993) reveals the continuing misunderstanding that bedevilled relationships between talented homosexual artists/writers and the leaders and supporters of the Cuban government on the island and abroad, as well as the minimal chance of reconciling opposing ideologies regarding homophobia in Cuba. Arenas and Fresa y chocolate tell the story of how some dissidents could not to adapt to a system that caused an acute sense of paranoia and isolation.

Chapters 3–5 demonstrated the existence of pervasive intimidation in German national elections to the Reichstag. Although electoral law mandated universal, direct, secret, and equal suffrage, it provided insufficient protections for voters to cast their votes freely. As a result, core principles of the electoral law were systematically violated in everyday electoral practices. As Heinrich Rickert, one of the German politicians who relentlessly pushed for the adoption of additional electoral reforms, argued: “The freedom of the vote is a guarantee of one of the most important constitutional provisions in Germany. It is the foundation of our constitutional life. If we do not provide sufficient guarantees for it, then our public life will be based on hypocrisy, on pressure, and on force, and those have never led to a good outcome” (Stenographische Berichte des Deutschen Reichstages January 15, 1896).

This violation of the cornerstone of German constitutional life to which Rickert referred resulted from two distinct factors. The first were imperfections in voting technology. As the previous chapters demonstrated, imperfections in the design of the ballot and the design of the urn allowed representatives of political parties to “pierce” the veil of secrecy of the vote and observe the electoral choices made by voters. Second, economic inequalities that increased the dependency of voters on their employers amplified the consequences of these irregularities in voting technology. These economic conditions lowered the costs of electoral repression for employers and increased the credibility of their threat to harshly punish voters who made “incorrect” electoral choices. Economic dependence also undermined the willingness of individual voters to risk their livelihoods by casting different ballots from the ones handed to them by their employers. This combination of institutional imperfections in the electoral code and economic inequalities in the labor market explains the persistence of electoral pressure in German elections.

The goal of this chapter is to examine political efforts to reform electoral institutions and to ensure greater protection of electoral secrecy.

The literature on the origin and development of the German Social Democratic Party (SPD) is vast. The puzzle about the steady increase in electoral strength of a party whose voters were subjected to systematic harassment and intimidation throughout the political history of the Empire has fueled the imaginations of every generation of historians and political scientists. What explains the rise in electoral strength of the SPD, and how did it become the largest political party in the Reich? I draw on a wealth of new economic data unavailable to previous scholars in order to reexamine this classic research question in comparative politics.

An analysis of the cross-sectional and temporal variation in the electoral strength of the SPD allows me to test a range of additional observable implications of the theoretical framework developed in this book. The first part of the book (Chapters 1–4) developed a number of hypotheses about the relationship between economic and political conditions in German districts and the “costs of electoral repression” encountered by private actors. I showed that the costs were higher in districts characterized by high levels of economic heterogeneity. The costs of electoral intimidation for state employees – such as policemen and tax collectors – also varied significantly across districts and were higher in districts characterized by high levels of political fragmentation among right- wing parties, where the outcome of a race was determined in runoffs. This chapter explores an additional observable implication of this analysis: I examine the factors that affected the variation in the willingness of voters to take political and economic risks by supporting opposition candidates. The corollary of the empirical findings presented in the first part of this book is that political support for Social Democratic candidates should be higher in districts where state employees and private actors faced higher relative costs of electoral intimidation.

Chapter 3 documented the existence of a wide variety of strategies of electoral intimidation carried out by employees of the state. In addition to policemen and tax collectors, private economic actors also played a dominant role as agents of electoral intimidation. During elections, companies turned into political battlegrounds. Private actors mobilized voters at the workplace, regimented them in columns, and marched them to the polls. Supervisory personnel of these companies ensured that voters entered the precinct equipped with the “correct” ballots. Employer representatives located in close proximity to the voting place maintained voting counter-lists (Gegenlisten) that recorded the electoral choices made by each and every employee. Using this information, employers engaged in post-electoral reprisals by punishing voters who had made incorrect electoral choices. Among the strategies of post-electoral punishments used by employers, the strategy that imposed the highest costs on voters involved being laid off for having made the wrong electoral choice.

In this chapter, I draw on the historical evidence presented in the electoral reports of the parliamentary commission charged with examining electoral irregularities to document pre- and post-electoral strategies used by private economic actors and the regional and temporal variation in these strategies. The main source of my analysis is a subset of reports on electoral fraud, which comprises 15 percent of the contested elections of the period. Using this subset of the total number of electoral reports, I code in greater detail whether instances of electoral intimidation by employers were present and the types of strategies used by private actors in elections. This qualitative evidence allows me to disaggregate the incidences of intimidation across different stages of the electoral process and examine when voters were most vulnerable to pressure from employers.

During much of the nineteenth century, the electoral marketplace in many European countries exhibited ample imperfections. At the time, electoral politics was not significantly different from the politics one encounters today in many developing countries that have recently undergone democratic transitions. In their attempts to sway voters, parties and candidates combined promises of policy benefits with a variety of non-programmatic strategies. These non-programmatic strategies included a mix of positive and negative inducements. Candidates campaigned by offering money, food, or entertainment to voters. In addition to these positive inducements, electoral intimidation, pressure, and harassment were amply used during elections. Policemen, tax collectors, and other local notables were deployed to influence the decisions of voters to support particular candidates. Imperfections in voting technology allowed candidates and their agents to engage in intense monitoring of the political choices made by voters and to punish the latter for undesirable political choices. Such threats of post-electoral punishments were highly credible because of ample imperfections in voting technology.

In many countries, political dissatisfaction with these electoral practices gave rise to efforts to curb electoral irregularities and enhance the political freedom of voters. These reforms centered on the dimension of electoral rules exhibiting the most significant imperfections: the technology of voting. Improvements in the design of the ballot and the design of the urn were regarded as central elements in the broader agenda to complete the unfinished democratic project. Consider the statements of Paul Constans, a French deputy who supported a proposal to adopt electoral reforms protecting voter secrecy:

Through these legislative changes we will not be able to eliminate the force of money – namely, the influence of the capitalist plutocracy and of the employers. But what we ask is very modest and will not modify the essential laws – there will be a moment later when we will do this. Right now, we are asking you to support these measures to liberate the voter, who is subjected to many types of influences and to allow him to express his will freely, with more freedom than the current situation permits.

The improvement in the protection of electoral secrecy had far-reaching consequences for political competition in German elections to the Reichstag. These changes in the electoral law reduced the opportunities for electoral intimidation. By emboldening voters to take electoral risks, the Rickert law made possible the spectacular political growth of the Social Democratic Party (SPD) during the final elections of the period. These gains for Social Democrats came at the expense of parties on the right. Because of the reduction in their ability to monitor voters' choices, many candidates on the right lost districts that had previously been their electoral strongholds.

This chapter investigates the implications of the changes in electoral competition for German parties on the right. My analysis focuses on one dimension of electoral competition that was salient for parties rather than candidates: the ability of parties to translate votes into parliamentary seats. I find that the Rickert law affected parties on the right in different ways depending on the district-level distribution of their support. Those right-wing parties whose voters were concentrated in particular regions – such as the Zentrum or the Conservatives – experienced no changes in the vote-seat proportionality in the aftermath of the Rickert law. The situation was entirely different for other right- wing parties, such as the National Liberals and Free Liberals. The ability of these parties to convert their votes into parliamentary seats deteriorated dramatically during the elections that were held after the adoption of the Rickert law and the onset of World War I.

This worsening of the relationship between the votes won by individual candidates and the number of seats held by representatives of the respective parties in the Reichstag led to an intense search by politicians for a solution that could allow their parties to survive electorally in the changed electoral landscape. In the process, the positions of several parties about questions of electoral reform experienced a gradual change.

Efforts to reform electoral institutions and reduce opportunities for political intimidation were not confined to national elections only; they were also on the agenda of subnational parliaments throughout the period until World War I. Electoral rules differed significantly across German regions (Mares and Queralt 2014). Many regions had much more restrictive suffrage rules than those in place at the national level. In these electoral systems, political efforts to end opportunities for electoral intimidation and provide stronger guarantees for voters' political autonomy went hand in hand with efforts to reform other dimensions of the electoral system. These included changes in the method of voting (from secret voting to open voting) and efforts to replace indirect voting with direct voting.

In this chapter, I examine political conflicts over the adoption of electoral reforms aiming to reduce opportunities for electoral intimidation in Prussia. Prussia was the largest state of the German Empire, and it comprised nearly two-thirds of the territory of Imperial Germany. Proposals to reform the electoral system along the dimensions discussed above were on the agenda of the Prussian lower house beginning in 1872. This chapter examines a quantitative analysis of the roll call votes on a subset of such bills in order to understand the economic and political determinants of support for changes in electoral rules and the composition of the political coalitions supporting reforms.

The study of electoral reforms in Prussia provides us with an ideal opportunity to reevaluate one of the most prominent explanations of democratization, which stresses the importance of rural inequality as an obstacle to electoral reform. Beginning with Alexander Gerschenkron and Barrington Moore, accounts of Germany's Sonderweg have invoked rural inequality as a factor that inhibited the adoption of democratic reforms (Gerschenkron 1946; Moore 1966). This explanation dovetails with recent theoretical accounts of democratization that emphasize how inequalities in the distribution of fixed assets act as barriers to democratic reforms (Acemoglu and Robison 2000; Boix 2003).

This chapter begins the empirical investigation of the production of electoral irregularities in German elections by analyzing the variation in electoral irregularities perpetrated by employees of the state. Electoral intimidation by state employees was pervasive in German national elections. The majority of the petitions submitted to the electoral commission of the Reichstag invoked electoral irregularities committed by employees of the state (Klein 2003). The ubiquitous presence of state employees as well as the brutal nature of their harassment and intimidation of voters added an unusually harsh character to nineteenth-century German elections. The identities of the employees of the state that supplied electoral services to politicians varied significantly across German regions. In Prussia, Landräte and policemen played a pervasive role in elections. The former cumulated important political responsibilities, serving as both the collectors of taxes (and assessors of tax liabilities) and making decisions on who was conscripted into the Prussian army. In Saxony, a region that experienced very high levels of electoral irregularities, policemen were ubiquitous in elections. By contrast, in southern German states such as Baden or Württemberg, where the electoral intervention by public employees was lower, mayors were the actors who occasionally trespassed against the provisions of the electoral law and engaged in the electoral intimidation of voters.

I begin this chapter by developing a number of hypotheses about the political considerations of politicians and employees of the state, and the incentives for the former to demand and the latter to supply electoral services conflicting with the provisions of the electoral code. In particular, I examine the relationship between political competition and the production of intimidation in German national elections.

In prewar Germany, a silent process of democratization took place during the final decades of the nineteenth century. This process unfolded through a series of piecemeal reforms of the administration of elections that attempted to provide better protection of voter autonomy. Questions about the size of the ballot, the size of the urn, the location of the urn in the voting place, and whether election officials should shake urns before counting ballots carried large political weight. Both opponents and supporters of changes in electoral institutions understood that these micro-regulations in the administration of the electoral process had profound consequences not only for political competition and the relative strength of different parties, but also for more abstract ideals such as electoral freedom.

Empirically, this book examined how this process of electoral reform un- folded in Germany between 1870 and 1912. The adoption of reforms in this electoral authoritarian context resulted from the confluence of two distinct transformations. First, economic changes such as labor scarcity and increases in economic heterogeneity raised the costs of electoral intimidation for private actors. Secondly, in political terms, the rise in strength of parties that lacked access to the repressive state apparatus and to the economic actors with the means to engage in electoral intimidation contributed to an increase in the size of the electoral coalition that supported reforms. The adoption of legislation that protected electoral secrecy, I showed, had important consequences for the rise in political strength of the largest opposition party: the Social Democrats. These reforms also had more indirect consequences for different parties on the right and facilitated the formation of an electoral coalition supporting the adoption of proportional representation.

In what follows, I consider and reflect on the contributions made in this book and their consequences for other cases of democratization. I begin in Section 10.1 by discussing the broader methodological contribution of this study to comparative historical analysis.