The last two decades have witnessed the gradual expansion of the attention that Cuban media pay to citizen letters. Currently, such letters form the basis of the radio talk show Hablando claro and TV programs like Libro acceso, Papelitos hablan, and Cuba dice. Several print media also feature regular sections based on citizen letters: the biweekly Bohemia prints Puntillazos; the weekly La tribuna de La Habana has Tribuna del lector; and the daily Juventud rebelde runs Acuse de recibo. Finally, since March 14, 2008, Granma has been publishing a weekly section called Cartas a la dirección, which prints not only letters to the editor containing complaints, criticisms, and suggestions, but also responses to these letters and editorial commentaries to perfunctory responses. The section rapidly grew in popularity and the number of letters received increased fourteenfold in the first year of its existence.1 To accommodate this explosion in reader interest, the newspaper increased the size of the section from one to two printed pages only two months after the publication of its first installment. In the editors’ own assessment (which is confirmed by what one hears on the streets of Havana), Cartas a la dirección is the most closely read portion of the Friday paper (Granma, January 4, 2013, 10). The section stands as one of the most visible symbols of Raúl Castro’s reform policies, in particular his emphasis on openness to criticism (apertura a la discusión y la crítica).
This attention to citizen letters raises three related questions that have broader relevance beyond the specific case of Cuba. First, why would media outlets in a communist regime solicit citizen letters? Second, why would some of these letters be publicized, either by being printed or by being read on radio or TV programs? And third, why would the media seek out responses to the letters and comment on unsatisfactory responses? This essay argues that in Cuba, as in other communist regimes, published complaint letters have two functions: the simple printing of select letters facilitates the collective letting off of steam, whereas the publication of responses to the letters by the authorities that were responsible for the infractions outlined in the initial complaint allows the regime to demonstrate that it takes popular input seriously. Therefore, Cartas a la dirección serves as a nonelectoral mechanism of accountability (Dimitrov 2019).
Though regime insiders may value appearing accountable by printing letters, citizens may be unwilling to complain either because they fear retaliation or because they do not believe that their letters will have any impact. This essay argues that by publicizing critical letters, the editors of media outlets signal that those who write will not be punished for expressing their views. Furthermore, complaining becomes more meaningful with the publication of responses that specify what measures have been taken to redress the grievances identified in the letters and to punish those who are responsible for them. However, this essay argues that the openness to input is limited: only letters about the issues covered under the social contract are publicized. Although this serves the ultimate goal of signaling to citizens that the social contract is enforceable and thus securing their quiescence in exchange for the provision of goods, services, and the protection of legal rights, it limits the scope of accountability by excluding grievances about civil and political rights from the range of the permissible complaints.
This study analyzes the role of citizen complaints in Cuba on the basis of an original dataset that was constructed by the author by compiling and hand coding all letters that were printed in Cartas a la dirección between March 2008 and September 9, 2016 (thus, collection was completed prior to Fidel’s death on November 25, 2016). The decision to construct the dataset on the basis of these letters was driven by several considerations. First, the systematic collection of transcripts of radio and TV shows based on citizen complaints is exceedingly difficult; although compiling the letters printed in Granma was not easy (the principal challenge was that no single library has complete holdings of the newspaper), it was nevertheless feasible on the basis of a combination of sources: the newspaper’s own partial online archive; the holdings of the José Martí National Library in Havana; the collection of the Ibero-Amerikanisches Institut in Berlin; and library repositories in the United States. Second, in contrast to other news outlets, which use citizen letters in redacted form to construct a story, Cartas a la dirección publishes unredacted letters and responses to them by the authorities (Granma, March 14, 2014, 10). This feature of the section allows for a more detailed coding of the letters and responses than would have been possible for the letter excerpts printed in other outlets. Third and most important, to the extent that the publication of citizen letters reflects editorial choices that are determined by regime priorities, the official organ of the Central Committee of the Cuban Communist Party is the print outlet that is most likely to give us insight into the political logic of publishing certain types of complaint letters and seeking out responses to them.
The dataset contains a total of 2,452 letters, 634 responses, and 126 editorial postscripts (coletillas). These are supplemented by letters published in Granma prior to 2008 and by a selection of letters from other Cuban print media. The essay argues that the analysis of this corpus of letters (in combination with findings from eight interviews with journalists and editors conducted in Havana in 2015) can shed light on a subject that has not yet received systematic scholarly attention: namely, the functions of letters to the editor in contemporary Cuba.
One question that is raised by this study concerns the role of the traditional press in the age of digital media. Although Cuba has a relatively low Internet penetration rate when compared to a communist regime like China, the share of the population that has access to the web has been rapidly expanding.2 The Internet has led to a “pluralization of the public sphere” (Hoffmann 2016), which has challenged the traditional role of the press as a “militant force in the service of … the Revolution and the Party” (García Luis 2013, 148). Despite the fact that all media in Cuba are state property (Article 53 of the 2002 Constitution), the rise of thousands of blogs and the circulation of the paquete semanal (the weekly digital entertainment package sold for about US$2) have exposed vast segments of the population to alternative sources of information (Díaz 2015). Scholars have highlighted Raúl’s characterization of the media as boring and have stressed that traditional media need to become more sensitive to public opinion in order to remain relevant in the age of blogs (Garcés 2014). These challenges explain why all major press outlets maintain webpages that feature numerous comments on citizen complaints. Interview evidence indicates that these moderated discussion boards are operated with the specific purpose of allowing traditional media to compete with the new media for the attention of readers (Cuba interviews, June 13 and June 15, 2015, Havana).
The citizens of autocracies regularly write letters to the authorities to register their dissatisfaction with the quality of services and to make demands for social assistance, the provision of welfare, or the guarantee of their legal rights. This is true for autocracies as distinct as Taiwan under the Kuomintang (KMT) (Chen 1983), Russia under Vladimir Putin (http://letters.kremlin.ru), and contemporary China (Dimitrov 2015). In the context of Latin America, studies of citizens demanding social assistance by writing to the president exist both for Peronist Argentina and for Mexico under the Partido Revolucionario Institucional (PRI) (Guy 2016; Nava 1994). One consequential difference between single-party communist regimes and electoral autocracies is the relative frequency with which such letters are written. For example, although Mexico had a population that was five times greater than that of East Germany, the annual number of letters received by President Carlos Salinas de Gortari was almost identical to that of the letters received by Erich Honecker when he was the leader of the German Democratic Republic (GDR).3 A second important difference concerns letters to the press: whereas in communist autocracies all media are owned by the state and thus letters to the media are not different from letters to other party-state organs (both are typically requests for services), letters to the media in electoral autocracies are likely to register opposition to government policies. There is no better illustration of this point than a comparison between letters to the Soviet press under Stalin or Brezhnev (Inkeles and Geiger 1952, 1953; Dimitrov 2014) and during the final years of perestroika (1989–1991) (Riordan and Bridger 1992), when competitive elections were introduced. Readers’ letters in Cuba today are more similar to letters to the media in the pre-perestroika Soviet Union than to letters to the media in electoral autocracies.
Two research strategies can be pursued by those who study citizen letters in communist autocracies, depending on the type of materials that are available to them. The first would involve analyzing all letters that were received by the party-state, including letters specifically directed to the media. The second strategy is to focus only on published letters. Analysis of all letters received requires access to government-generated documents that reveal how regime insiders utilized these letters to assess popular opinion. This research strategy is not viable for Cuba. Examining published letters to the press allows us to make arguments about how the regime uses these letters to guide popular opinion so that it is perceived as accountable to the public.
Authoritarian regimes face challenges when trying to collect information on popular discontent (Friedrich and Brzezinski 1965; Wintrobe 1998). The fundamental problem is the inability of dictatorships to overcome preference falsification, which involves citizens publicly declaring their support for the regime, while being privately opposed to it. Political scientists have attributed the rise of sudden coups and revolutions to preference falsification (Kuran 1991; Lohmann 1994). The empirical expectation created by this theoretical literature is that nondemocracies will be inherently unstable and short-lived. Although some of them are, single-party communist regimes have emerged as the most durable subtype of authoritarian regime.4 This raises the question of how they have been able to mitigate the information problem.
In recent years the literature on comparative authoritarianism has proposed several solutions to the information problem. One is the holding of competitive elections, which can reveal to the incumbent his share of popular support (Magaloni 2006; Greene 2007; Miller 2015). Another is the encouragement of protests, which can be used to locate the geographic incidence of discontent (Lorentzen 2013). The third involves encouraging investigative journalism (Lorentzen 2014) and monitoring the content of commercialized print media and social media (Stockmann 2013; King, Pan, and Roberts 2013; Pan and Chen 2018). Although these channels undoubtedly generate information, they do so at a cost: the information on popular discontent is available both to regime insiders and to other discontented citizens, who can use it to engage in collective action that can destabilize the regime. Therefore, autocrats need to foster other channels that allow for the private transmission of information only to regime insiders. What might these channels be?
The archival revolution that followed the collapse of communism in Europe produced new insights regarding the internal government understandings of the information problem. Research based on previously classified government-generated materials demonstrated that in contrast to the standard arguments in the literature, communist regimes were aware of the information problem and actively worked to develop strategies that would allow them to mitigate it (Fulbrook 2005; Betts 2010). Various studies have emphasized how information on public opinion was collected through techniques like state security surveillance (Izmozik 1995); opinion polling (Friedrich, Förster, and Starke 1999); and the analysis of rumors and jokes (Hertle and Saure 2015). In addition, citizen complaints both in the form of letters to the editor and as signals directed to various party and government organs have been analyzed as an avenue for information collection (Bos 1993; Eberle 2016; Suckut 2016); complaints were so highly prized in the GDR that districts (Bezirke) were encouraged to compete for handling the highest volume of citizen complaints (Eingaben) (Betts 2010, 175). Although each of these channels had its biases and limitations, when taken as a whole, the information they yielded made it possible to produce remarkably nuanced assessments of public opinion in communist single-party autocracies (Dimitrov 2014).
These studies of other communist autocracies generate hypotheses about information gathering in Cuba. We know that like other communist regimes, reform-era Cuba employs a variety of techniques to assess popular opinion. Despite the lack of official acknowledgment, it is reasonable to assume that the Dirección General de Contrainteligencia (DCI) and various technical divisions of the Ministry of the Interior (MININT) conduct surveillance of dissidents and monitor telephone, post, and electronic communications, as did their counterparts in other communist regimes (Kallinich and de Pascuale 2002). In contrast, public opinion research by the Centro de Estudios Sociopolíticos y de Opinión (CESPO) of the Central Committee of the Cuban Communist Party has been publicly recognized. CESPO studies public opinion by carrying out surveys; by conducting individual and group interviews with citizens; and by monitoring the so-called opinión espontánea, which consists of spontaneous popular reactions on matters like employment, transportation, food shortages, and government policies and is typically expressed in long queues and overcrowded buses.5
What is the role of letters to the press in Cuba? One hypothesis that cannot be tested with the currently available materials from Cuba, but which seems warranted given what we know about other communist single-party regimes, is that the real purpose of Cartas a la dirección is to stimulate citizens to send letters to Granma so that these unpublished letters can be analyzed and information about public opinion can be extracted from them and transmitted vertically to the top leadership. After all, the value of citizen complaints as a source of information is officially acknowledged in print sources (Castellanos 1988; Juventud rebelde, January 1, 2006, 4; Granma, January 4, 2013, 10), in public commentaries by editors in chief (Último jueves, May 30, 2013), and in interviews with journalists in Havana (Cuba interviews, June 9, June 13 and June 16, 2015, Havana). However, testing this hypothesis would require access to a source of data that is currently unavailable: the almost 100,000 letters and emails that Granma receives on an annual basis (Granma, January 4, 2013, 10), rather than the 250 or so that it decides to publish. Until such data become available, we can focus on gaining a more complete understanding of the function of published letters to the editor in Cuba.
The standard understanding of critical letters published in newspapers in communist regimes is that they function as a social safety valve (Inkeles and Geiger 1952, 1953). Although letters in Cuba also allow for the collective letting off of steam, the safety valve explanation does not shed light on why the regime would find it worthwhile to authorize print media to seek out responses to published letters and to comment on those responses that they consider inadequate. Granma’s practice of printing not only letters but also responses and coletillas gives rise to a different hypothesis about the role of citizen complaints in Cuba: namely, that complaints serve as a mechanism for accountability. In a communist dictatorship, the central government can institute a system of proxy accountability (Dimitrov 2013). In such a system, the central government acts as the proxy of citizens, holding local officials accountable on their behalf. If local officials fail to respond to citizen complaints, higher levels of government can instruct lower levels of government to resolve the problems referred to in the complaint. In the end, local officials are more likely to respond to citizen complaints when higher levels of government are involved. This essay argues that as long as the public trusts the central government (or Granma, which represents the party-state) to intervene on its behalf, it will provide information through citizen complaints.
Accountability rests on two pillars: responsiveness and sanctioning (Rubin 2005; Philp 2009). Responsiveness refers to the requirement that officials provide an explanation or a justification for their behavior. Sanctioning refers to the ability of citizens to punish officials for being unresponsive or corrupt. This essay argues that Granma, which is the official organ of the Central Committee of the Cuban Communist Party, allows the regime to demonstrate its commitment to proxy accountability by authorizing the publication of critical letters in Cartas a la dirección. By seeking out responses to such letters, the editorial staff of Granma signals regime commitment to responsiveness. In turn, the printing of coletillas to inadequate responses indicates that the sanctioning of irresponsible officials (who are typically party members) is also taken seriously. The cumulative effect of Cartas a la dirección is to enhance citizen loyalty by presenting the regime as accountable to the public.
Even if the party-state values complaints, it would be unreasonable for citizens in an autocracy to expect that the government would be willing to hear all types of grievances. This raises the question of how citizens know the scope of permissible claims. In a communist state, the range of acceptable demands is determined by what scholars have called “the social compact” or “the social contract” (Pravda 1981; Millar 1985; Hauslohner 1987; Cook 1993). The conventional use of the social contract concept in contemporary scholarly research on authoritarian politics diverges from that articulated by Rousseau insofar as it does not assume that citizens enter into the contract voluntarily. Nevertheless, there are expectations of how the two parties to the contract should behave: namely, citizens will remain quiescent as long as the regime supplies material benefits and safeguards their legal rights. The contract is binding on the regime and citizens can punish it by withdrawing their contingent consent if the regime reneges on its commitment to provide goods and social services and to guarantee legal rights (Cook 1993; Cook and Dimitrov 2017; Dimitrov 2018).
One may wonder what is left of the social contract in contemporary Cuba. At the Sixth Party Congress in 2011, Raúl Castro laid out an ambitious plan for economic reform, featuring price liberalization, the end of rationing, economic decentralization (although large enterprises would continue to be owned by the state), a massive expansion of market activities with the concomitant limiting of the role of the planned economy, and the layoffs of 1.3 million people (25 percent of the Cuban labor force) (Castro 2011). These reform plans called for nothing short of the dismantling of the socialist social contract (Mesa-Lago 2012; Mesa-Lago and Pérez-López 2013). Nevertheless, eight years after the Sixth Party Congress, many entitlements have been preserved. The layoff targets have not been achieved (although the number of those employed in the private sector increased from 12 percent of the labor force in 2007 to 20.3 percent in 2017, Torres 2016, 1688; Oficina Nacional de Estadística e Información 2018), price liberalization has not occurred, and rationing persists in the form of the ubiquitous libreta de abastecimientos (supplies booklet), which entitles every Cuban citizen to purchase from one of the specialized groceries (bodegas) their rations of sugar, rice, beans, bread, eggs, poultry, minced meat, pasta, salt, matches, soap, toothpaste, and coffee. These items are sold significantly below cost, with monthly rations costing no more than $3.00, which is about an eighth of the average monthly salary. In addition, phone, electricity, and gas for cooking are available for another $2.50 a month per household. Public transportation, basic medicines, newspapers, books, and theater tickets are also heavily subsidized. Finally, few Cubans pay rent, since most of them own their housing. Overall, the party-state has proven reluctant to quickly implement bold reforms that would lead to a dismantling of the remnants of the socialist social contract. By signaling to citizens that the regime takes the social contract seriously, published letters to the editors help increase regime legitimacy.
Letters to the editor are not a new phenomenon in Cuba. Granma published them decades before it established Cartas a la dirección in 2008. And in reform-era Cuba, Granma is not the only periodical that features such letters. This section of the essay puts the letters printed in Granma today in historical context and compares them with those published in other contemporary news outlets. These comparisons allow us to highlight what is unique about the letters in Cartas a la dirección: although letters that were published by Granma prior to 2008 and letters that appear in other print media today act as a social safety valve, they have limited utility as a mechanism for accountability.
Granma started publishing responses to citizen queries in 1970 in a tiny column entitled Buzón económico (Economic Mailbox). Citizens would send in questions (inquiring about such items as the female participation rate in economic activity; the nutritional characteristics and the quality of various types of rice; different ways of organizing the supervision of sugarcane harvesting; and the number of artificially inseminated cattle prior to the triumph of the Cuban Revolution) and Granma would answer (the answer to the last query was none). These were essentially requests for hard-to-find statistical information rather than criticisms, complaints, and suggestions. Another section that was published at irregular intervals throughout the 1970s was Luz roja (Red Light). It was based on citizen letters but focused exclusively on traffic accidents, traffic safety, and poor transportation infrastructure. Because it did not publish responses to letters, Luz roja (as well as Cartas de viaje and Cartas sin sobre, which were introduced in the 1980s and promptly discontinued) could not transcend the function of a social safety valve.
A vuelta de correo (By Return Post) offers an important hybrid form of a section on readers’ letters that also featured responses from government agencies. Granma introduced this section in February 1975, when preparations for the First Congress of the Cuban Communist Party were in full swing, and published it irregularly until 1984. The section featured articles based on readers’ letters (usually redacted), which were accompanied by a photograph, and infrequently, by a redacted response. The articles typically focused on questions like waste of state resources or identified instances of abandoned machinery (bulldozers, tractors, trucks). Occasionally, letters drew attention to issues like shortages (of toothpaste, bath soap, and batteries and spare parts for VEF, Sokol, and Orbita radios), the irregular supply of carbonated drinks, and the poor customer service in upscale hotels. Some letters highlighted abysmal delays in providing services: one reader complained that the watch repair shop Slava in Havana took two years to clean his Swiss-made Prisma Lux (Granma, October 10, 1978, 5). In another case, readers were alarmed by bottled beer that contained “insect eggs, soil particles, and other impurities,” including “a crayfish 4 centimeters in length” (Granma, March 22, 1984, 3). Articles referencing such letters enabled the collective letting off of steam. Typically, no response from those responsible for the infractions was included in the article. Occasionally, there was a reply, but it deflected responsibility or cited “objective difficulties.” Other times, the response letter vaguely reported that “necessary steps were taken” to redress the situation. Overall, although the articles may have helped improve responsiveness, they did not function as a full-fledged mechanism of accountability, because it was highly unusual for letters to be specific concerning who was responsible and how he or she was sanctioned. One exception that proves the rule is the case of a postman who negligently discarded twelve telegrams in the yard of the Freyre Andrade hospital in Havana; he was terminated and legal proceedings were initiated against him (Granma, January 19, 1978, 3).
In 1988, Granma ceased printing letters to the editor altogether. Heavily editorialized letters were occasionally published, starting in 1994 in Abrecartas, but the section was discontinued in the early 2000s. Given the grievances that accumulated in Cuban society during the ensuing years, the inauguration of Cartas a la dirección in 2008 was met with keen interest by the citizenry (Cuba interview, May 30, 2015, Havana).
Several media outlets publish letters to the editor today. One is Bohemia, which runs Puntillazos; the section features several paragraph-long critical commentaries on negative phenomena (noise, accumulation of uncollected trash in the streets, or the poor quality of services) that are extracted from readers’ letters and are typically accompanied by a photograph. In contrast to Cartas a la dirección, which insists on printing the complete name and address of letter writers, complaints excerpts in Puntillazos feature only the first name of the reader. Most importantly, no responses are printed, so the section is best understood as a contemporary equivalent of Granma’s Luz roja.
Three other print media publish readers’ letters: one is Trabajadores, the other La tribuna de La Habana, and the third Juventud rebelde. All three feature articles reminiscent in style to those that appeared in A vuelta de correo in the 1970s and 1980s: they typically include a redacted letter, a response from the entity that is criticized in the letter, and some editorial commentary. Of the three, Juventud rebelde is the most widely read, because it has the second-highest print run after Granma (Juventud rebelde, July 25, 2010, 4). Juventud rebelde has published a section entitled Acuse de recibo (Acknowledgment of Receipt) since 1997. Currently, the section appears five times a week. According to internal assessments, it is the second most popular section of the newspaper after Sexo sentido (which focuses on sexuality) (Cuba interview, June 15, 2015, Havana).
In many ways, Acuse de recibo and Cartas a la dirección resemble each other: they receive and publish a similar number of letters on largely identical issues (various irregularities in the provision of municipal services, excessive bureaucratic regulations, or shortages of certain goods); they keep track of the government agencies that do not respond to complaints that have been published; and they have a similar understanding of the role of complaints as a window onto public opinion (Juventud rebelde, August 21, 2005, 4; Granma, June 27, 2008, 10). Nevertheless, there are important differences between the two sections in terms of their capacity to act as mechanisms of accountability. One is that Acuse de recibo is a section managed by two journalists (José Alejandro Rodríguez and Jesús Arencibia Lorenzo), whereas Cartas a la dirección represents the anonymous editors of Granma. The fact that Juventud rebelde articles are attributed to specific journalists diminishes their potential to be immediately perceived as reflecting efforts by the party-state to establish responsiveness and sanctioning. Another concerns the way that letters are presented. Providing excerpts from letters and responses is fully consonant with the traditional journalistic style of socialist media, where letters were always accompanied by a commentary that told readers what is the correct way to interpret them. This practice diminishes the potential of Acuse de recibo both to act as a social safety valve and to signal government commitment to accountability. In contrast, letters and responses published in Cartas a la dirección look more authentic, because they are not redacted and editorial commentary (which is minimal) is clearly identified as such (Granma, March 14, 2010, 10). This might explain why the number of letters received by Acuse de recibo has been declining in recent years, whereas the volume of letters directed to Cartas a la dirección has been increasing (Juventud rebelde, January 17, 2010, 4; Granma, March 16, 2012, 10). The last and most consequential difference is that Juventud rebelde is published by the Communist Youth League (Unión de Jóvenes Comunistas), whereas Granma is the official organ of the Central Committee. This means that content in Cartas a la dirección is rightly perceived to represent the official position of the regime. As a result, Granma is the only Cuban print media source that can enforce proxy accountability on behalf of the party-state through the mere act of publishing responses and coletillas.
This section uses a detailed coding of readers’ letters, responses, and editorial postscripts published in Granma to assess whether the content of these letters has evolved since the inception of Cartas a la dirección. To have maximum leverage on change over time, the essay analyzes all letters printed in the first year after the new section was introduced (March 14, 2008–March 6, 2009) and in the most recent complete year for which published letters were collected (2015).
We cannot make sense of the letters published in Granma without systematic coding rules. After each letter was read, it was assigned to one of two categories: noncritical or critical. Noncritical letters were subdivided into specific subcategories such as letters praising the party and the revolution, opinions on the pace and direction of reform, opinions on the importance of establishing standardized accounting practices, suggestions about improving work discipline, praise of the educational system and the standard of medical care, and letters expressing satisfaction with the quality of goods and services. Critical letters included complaints about the poor quality of goods and services, complaints about shortages of goods, complaints about excessive bureaucratic regulations, complaints about speculative trade, complaints about violations of social order, and signals about environmental pollution. Classification was straightforward: critical letters, even when written by individuals who described themselves as loyal to the party and the revolution, contained complaints about the implementation of government policies and the provision of services. The largest subcategory of such letters—complaints about the quality of services—was subdivided into specific issues, such as transportation services, medical services, customer service in hard-currency stores, mail delivery services, telephone services, and trash collection. Grouping the letters into these categories allows us to track the critical orientation of Cartas a la dirección over time.
Tables 1, 2, and 3 reveal three important differences between the letters that appeared in Cartas a la dirección in 2008 and those published in 2015. First, Table 1 indicates that the share of noncritical letters has dropped from 35.9 percent in 2008 to 16.4 percent in 2015. Especially noteworthy is the decline in letters praising the party and the revolution from 12.2 percent in 2008 to 1.9 percent in 2015 (Table 3). Second, the ratio of published letters to published responses has decreased more than fourfold, from 7.5:1 in 2008 to 1.7:1 in 2015 (Table 2). We should also note that in December 2013 Granma started printing “postscripts” (coletillas), which are commentaries that are attached to responses that the editors consider deficient in some way. The postscripts have led to a marked improvement in the quality of the responses. And third, the share of letters focusing on shortages and the poor quality of goods and services has increased significantly, from 28.5 percent in 2008 to 50.7 percent in 2015 (Table 3). These three trends indicate that the party wants to send increasingly stronger signals about its openness to criticism and its commitment to establishing accountability.
|Total letters||Noncritical letters||Critical letters|
|2008||270||97 (35.9%)||173 (64.1%)|
|2015||213||35 (16.4%)||178 (83.6%)|
|Letters||Responses to letters||Letters to responses ratio||Postscripts|
|% of letters praising the party and the revolution||% of letters about shortages and the poor quality of goods and services|
One of Raúl Castro’s strategies for increasing the legitimacy of the Cuban Communist Party was to declare its openness to criticism in early 2008 (Mesa Lago 2012, 242–243). The very establishment of Cartas a la dirección within weeks of Raúl’s speech stands as tangible evidence of such openness. Another indicator is offered by the editorial decisions about the letters that are printed in the section, in particular the gradual decline of noncritical letters between 2008 and 2015. Table 4 offers a breakdown of the noncritical letters printed in Granma in 2008 and 2015. We need to highlight two trends. One is the already noted sharp reduction in letters praising Fidel, Raúl, the Cuban Communist Party, the achievements of the Cuban Revolution, and international solidarity. Although thirty-three such letters were printed in 2008, by 2015 their number had been reduced to a mere four. Without a doubt, some Cubans must believe in the party and the achievements of the revolution. Yet, even if they are genuine, such letters can easily appear doctored, as when the daughter of a convict expresses gratitude to the penitentiary system and testifies to the value of the revolution for her family; or when an inmate thanks the party for converting prisons into schools, thus allowing him to study the revolutionary ideas of Fidel and Raúl and to bring joy to his mother’s heart; or when a cancer survivor proclaims that this much love and care for the sick can only exist in the Cuba forged by Fidel (Granma, October 3, 2014, 11; Granma, May 2, 2014, 11; Granma, October 17, 2014, 11). Such overt displays of loyalty to the revolution are suspect; printing them less often can help increase the overall credibility of the letters published in Cartas a la dirección. Another policy that makes content look more authentic is the insistence of the editors on printing the full name and address of the readers who send letters.
|Absolute number 2008||As % of all letters in 2008||Absolute number 2015||As % of all letters in 2015|
|Praise of the party and the revolution||33||12.2%||4||1.9%|
|Opinions on the pace and speed of reform||33||12.2%||10||4.7%|
|Opinions on the need to improve control, enterprise budgeting and introduce uniform accounting standards||17||6.3%||4||1.9%|
|Praise of the quality of goods and services||5||1.8%||6||2.8%|
|Praise of the educational system||4||1.5%||3||1.4%|
|Praise of the medical system||1||0.4%||4||1.9%|
The decision to print letters from each of Cuba’s sixteen provinces can also be understood as an effort to demonstrate openness to public opinion from all parts of the island. Coding of the letters in the dataset indicates that 49.8 percent of the letters that were printed in 2015 came from Havana. Granma publishes less than 5 percent of the letters that reach it (Granma, October 2, 2015, 10), but it does not report on the volume of letters it receives from various provinces; therefore, we cannot know how representative published letters are of all letters sent to Cartas a la dirección. However, Acuse de recibo reports that 47.3 percent of the letters it received between April and October 2015 came from Havana (Juventud rebelde, January 10, 2016, 4); the letters received by Juventud rebelde from other provinces are very similar to the share of letters from these provinces printed in Granma. This suggests that Cartas a la dirección may be following internal rules mandating that letters from all provinces be printed. To the extent that it gets noticed, this representation of each province may build trust among readers and incentivize them to direct their grievances to Granma rather than air them on the Internet.
This essay has argued that letters to the editor can reduce social tension by disseminating carefully managed information about widespread popular grievances. This is a counterintuitive claim, as the dissemination of such information can, of course, lead to instability by creating a focal point that allows for collective antiregime mobilization (Kuran 1991; King, Pan, and Roberts 2013). Importantly, Cuban media do not print complaints that directly challenge the legitimacy of the party by making demands for improvements in political rights and civil liberties. Race is also a taboo topic. Labor issues identified by scholars as likely to spur collective mobilization in other communist regimes that transitioned to a market economy (enterprise closure, nonpayment of wages, and unemployment) (Cai 2010) received no coverage in Granma in 2008 and accounted for only 0.9 percent of the letters printed in 2015. Finally, sensitive questions like the quality of education or medical care were absent in 2015 and accounted for only 1.9 percent of all letters printed in 2008. Cartas a la dirección publishes letters on “safe” issues that do not allow for collective mobilization (service quality, shortages of nonessential items, and excessive regulations) and appears to be taking these grievances seriously by stressing responsiveness. Under those conditions, publishing letters to the editor can reduce social tension by allowing citizens to collectively let off steam. Because these issues are a source of daily consternation for Cubans, printing letters on such themes is very popular, as indicated by interviews with journalists (Cuba interviews, May 30 and June 9, 2015, Havana), by audience research conducted by the Equipo de Investigaciones Sociales of Juventud rebelde (Cuba interview, June 16, 2015), and by what one hears on the streets of Havana.
As Table 5 demonstrates, Granma has begun to print more critical letters over time. Thus, in 2015, complaints about excessive regulations, shortages, and the poor quality of goods and services constituted 61.5 percent of the letters in Cartas a la dirección, a significant increase over 2008, when they only made up 36.7 percent of the total. Readers writing about excessive regulations focused on the difficulty of obtaining permits to repair or build new homes; the challenges involved in transferring property; and the difficulty of building a successful business as a cuentapropista (self-employed individual). These complaints maintained a roughly similar share of all letters printed in Cartas a la dirección: 8.1 percent in 2008 and 10.8 percent in 2015.
|Absolute number 2008||As % of all letters in 2008||Absolute number 2015||As % of all letters in 2015|
|Excessive bureaucratic regulations||22||8.1%||23||10.8%|
|Poor cadre quality/poor management||16||5.9%||5||2.3%|
|Poor work discipline||8||3%||0||0%|
|Waste of resources||5||1.8%||6||2.8%|
|Use of state assets for private gain||2||0.7%||0||0%|
|Enterprise closure, nonpayment of wages, and unemployment||0||0%||2||0.9%|
|Violations of public order||13||4.8%||1||0.5%|
|Poor quality of goods (including food)||3||1.1%||7||3.3%|
|Poor quality of services (nonmedical and noneducational)||63||23.3%||86||40.4%|
|Poor quality of medical services||1||0.4%||0||0%|
|Criticism of the quality of education||4||1.5%||0||0%|
With regard to shortages, the number of letters printed increased threefold, from 2.2 percent of the total in 2008 to 7 percent in 2015. It is unlikely that this represents a dramatic worsening of the existing shortages in Cuba. Instead, it probably indicates an editorial decision to give a higher profile to this problem, knowing full well how much time ordinary citizens devote to looking for scarce items. In 2008, letter writers focused on the shortages of tampons, contact lenses, eyeglasses, and orthopedic shoes. In 2014, beyond tampons and orthopedic shoes, citizens complained about the scarcity of condoms, toilet paper, school uniforms, chocolate mix, salt, deodorant, potatoes, pots and pans, light bulbs, batteries for electric bikes, motorcycle tires, animal feed, toothbrushes, and white tooth fillings and resins. In 2015, letters highlighted the shortage of orthopedic shoes, gas-permeable lenses, digital TV converter boxes, water filters, building materials, bleach, and insecticides. In 2015, Granma also published letters about the shortages of basic foodstuffs distributed through the infamous libreta de abastecimientos; letters focused on the poor quality of rice (Granma, January 9, 2015, 9; January 30, 2015, 7) and the shortage of soy yogurt (Granma, April 17, 2015, 11; July 17, 2015, 11). For most of those products, explanations from the responsible suppliers were printed, offering specific plans for meeting the existing demand. However, the ongoing chronic shortages of many items suggest that these plans have had limited effectiveness.
Table 6 offers a detailed breakdown of the letters focusing on the poor quality of services, which constituted the single largest group of complaints in both 2008 (25.2 percent of all letters) and 2015 (40.4 percent of all letters). Some items appeared on the list in both years, accounting for a roughly similar share of the total number of service complaints: transportation (irregular bus service, rude drivers pocketing bus fare, buses infested with vermin), telecommunications, mail delivery, and rude customer service in hard-currency stores. Others, like service quality at the ice cream parlor Coppelia, appeared in 2008, but not in 2015. Finally, one category of service complaints (inconsistent prices at state-owned stores) was present in 2015 but not in 2008 (complaints about speculative trade by cuentapropistas appeared in both 2008 and 2015); this is consistent with the emphasis at the Seventh Party Congress in April 2016 on combatting price speculation (Castro 2016). In the absence of reliable opinion polling results, it would be unreasonable to treat this data as representative of the relative prevalence of different types of concerns about poor customer service. The data do, however, illustrate the striking range of poor customer service in contemporary Cuba.
|Absolute number 2008||As % of all letters about services in 2008||Absolute number 2015||As % of all services in 2015|
|Hard currency stores||6||8.8%||5||5.8%|
Granma also publishes letters that indicate satisfaction with the quality of goods and services (see Table 4). Whereas in 2008 the letters of this genre seemed exaggerated (as, for example, when a complaint about the quality of coffee and chocolate mix resulted in an angry riposte from another reader who believed it was “unjust to evaluate the chocolate mix negatively”) (Granma, August 15, 2008, 11), by 2014–2015 these letters were toned down. And yet, by expressing thanks for exemplary service, they highlighted how rare it is to see a clean bus (Granma, June 27, 2014, 11); to have one’s wallet or other lost valuables returned (Granma, September 18, 2015, 9); or to encounter a medical professional who uses gloves (a dental student thanked a veterinarian for draining an abscess on her dog with enthusiasm but without gloves, “his hands covered in blood and pus”) (Granma, April 11, 2014, 11).
As printing letters represents the choices of the editorial staff of Cartas a la dirección, we cannot infer with any certainty whether some daily life problems have become worse between 2008 and 2015. We can only conclude that Granma has made a decision to publicize these grievances and that, instead of fearing that shining a light on these popular grumbles would destabilize the system, it knows that it can reduce social tensions by allowing citizens to collectively let off steam.
The chief problem that communist regimes encounter when soliciting citizen complaints is how to ensure a continuous flow of such letters. Even though the introduction of a section that regularly publishes letters to the editor could induce readers to write to newspapers, over time this willingness might taper off, unless citizens believe that their complaints can have some meaningful impact on redressing the grievances they have highlighted. For this reason, the party-state needs to ensure that responses to some letters are printed. Furthermore, these responses have to specify what measures were taken to investigate the complaints and how those who were responsible were punished for their infractions. To the extent that they do that, responses serve as an instrument of accountability.
In 2008, Granma printed 270 letters and only 36 responses. The responses were often pro forma, pointing out objective difficulties or the US blockade as the source of the problem. Sometimes the response would indicate that the letter writer was a troublemaker or would entirely negate the complaint, as when Cubacafé said that coffee quality was good or when the post office indicated that an employee suspected of negligence in delivering Granma to a subscriber was in fact exemplary (Granma, October 24, 2008, 11; Granma, August 29, 2008, 10; Granma, August 22, 2008, 10). Even when the response acknowledged that there were problems, the solutions offered were nonspecific: following a complaint about poor customer service at the ice cream parlor Coppelia, the management blamed the shortcomings on the long-term legacies of the Special Period, stated that the collective was fervently dedicated to its work, and concluded by saying that, nevertheless, workers were instructed to commit to offering sterling service (Granma, July 25, 2008, 10). These measures were ineffective, as demonstrated by the ongoing complaints about Coppelia that appeared in Granma throughout 2008. Overall, the responses printed in 2008 did not convey the impression that the letters in Cartas a la dirección could function as a mechanism for accountability in Cuba. What made matters worse is that over time, responsiveness declined even further: in 2011, for example, only 8.8 percent of the 373 letters published resulted in a response.
Granma took several measures to improve the level of responsiveness to citizen complaints. In December 2013, it introduced postscripts to selected readers’ letters and, especially, to inadequate responses. For example, when commenting on the response of a water supply company to a complaint about flooding in Habana del Este due to sloppy maintenance work, Granma stated that “objective problems are one thing and poor customer service quite another” (Granma, July 18, 2014, 10). Such editorial comments were meant to publicly shame government agencies that were shirking responsibility. Another measure that aimed to increase responsiveness was an overview once every four months of the responses received and a list of the entities that did not respond to printed letters in a timely manner. The result of these efforts was impressive: in 2015, only 30 percent of the letters that were published did not receive a response; in the first six months of 2016, this figure was reduced to 23 percent (Granma, July 8, 2016, 11).
The quality of the responses also improved rapidly: response letters were more likely to specify who was determined to have been responsible for a certain infraction and how he or she was punished. Another important feature of responses was that by 2015, they typically specified that a visit by the authorities to the home of the complainant had taken place in order to get a better sense of the issues involved in the complaints; readers reported that they were satisfied with this new policy that demonstrated concern for their grievances (Granma, April 3, 2015, 9). Coletillas were used as a mechanism to induce the sanctioning of officials who were responsible for the initial infraction (and to send them a message that the party takes discipline seriously); they demanded that specific measures be implemented to punish those responsible for the infractions committed (Granma, May 15, 2015, 10; Granma, November 27, 2015, 8). As most businesses against which citizens complain are state-owned, punishing employees demonstrates that the state is concerned about public opinion. Overall, then, we can conclude that the responses printed in Granma in 2015 project a much more convincing image of accountability than did those that appeared in Cartas a la dirección in 2008. They attest to the efforts of the traditional press to adjust to the twin challenges of Raúl’s reforms and the competition of digital media.
This essay has analyzed the functions of readers’ letters published in Granma since 2008. It has argued that these letters improve the quality of governance in Cuba by signaling the openness of the party to criticism, by allowing citizens to collectively let off steam, and by creating a mechanism for accountability. At the same time, these letters also indicate that the openness to criticism is limited to certain issues like the provision of subsidized goods and services and the protection of rudimentary property rights that constitute a part of the social contract. The experience of the East European communist regimes that collapsed shortly after they tried to dismantle their social contracts indicates that the party may be able to maximize popular support by further delaying bold reforms that would eliminate the entitlements associated with the social contract. Thus, though economically inefficient, the social contract is politically expedient.
In many ways, the weekly publication of readers’ letters in the official organ of the Central Committee is an anachronistic phenomenon in the era of digital media. These letters are handpicked by the editors of Granma to reflect current regime priorities and to manage public opinion in ways that are favorable to the party-state. However, as the experience of China indicates, high Internet penetration allows for the rapid horizontal dissemination of information about popular grievances through social media like Weibo and WeChat and establishes requirements for prompt responsiveness (Dimitrov 2017). Thus, further Internet penetration could eventually make Cartas a la dirección obsolete and create a new challenge for the communist leadership: namely, how to control a technology that allows for the instantaneous transmission of information on popular opinion and requires a degree of responsiveness that is higher than the one the party has been providing so far.
1The citizen letters department at Granma received 468 letters in the one-year period prior to the inauguration of Cartas a la dirección and 6,364 letters and emails in the one-year period following (Granma, March 13, 2009, 10).
2Over 1,000 Internet access points existed as of 2016 (up from 118 telepuntos in 2013), but Internet penetration remains low at 32.4 percent (http://www.internetlivestats.com/internet-users/cuba/, accessed May 6, 2017). We should note that the falling price of Internet access and the growing number of public Wi-Fi spots are positive trends.
3Salinas received 432,996 letters in the first four years of his presidency (MacGregor 1993, 15), whereas Honecker received 2 million letters in his eighteen years in office (Deutz-Schroeder and Staadt 1994, 6), resulting in 108,249 letters per year for Mexico and 111,111 for the GDR. The average population of Mexico in 1988–1992 was 85.6 million, whereas the GDR had an average of 16.7 million citizens in 1971–1989.
4Communist regimes are the most durable subtype of nondemocratic regime, outlasting both noncommunist single-party regimes and nondemocratic monarchies. As of 2000, the average lifespan of noncommunist single-party regimes (like Mexico under the PRI) was 28.51 years and that of nondemocratic monarchies 34.75 years. In contrast, communist single-party regimes had an average lifespan of 46.2 years as of 2000. My dataset includes thirty-nine noncommunist single-party regimes (based partially on Smith 2005), twenty nondemocratic monarchies, and fifteen communist regimes. As of 2019, the five remaining communist regimes had an average lifespan of 62 years.
5As stated at an Último jueves roundtable “Opinión pública y toma de decisiones,” organized by the magazine Temas on May 30, 2013, in Havana and attended by the author. Subsequent references to this roundtable will be Último jueves, May 30, 2013.
I would like to thank the Stone Center for Latin American Studies, the Cuban and Caribbean Studies Institute, the Center for Inter-American Policy and Research, and the Department of Political Science at Tulane University for financial and logistical support. This manuscript was finalized during a fellowship year at the Princeton Institute for International and Regional Studies. For feedback, I want to thank Ana López, Carolina Caballero, Ludovico Feoli, Marilyn Miller, Virginia Oliveros, Rubén Gallo, Rachel Price, Scott Morgenstern, Bert Hoffmann, and audiences at seminars at Tulane, Pittsburgh, Princeton, and at the 2015 Annual Meeting of LASA in New York City. The personnel at the Latin American Library at Tulane, at the Ibero-Amerikanisches Institut in Berlin (especially Frau Roldán), and at the José Martí National Library in Havana were exceptionally helpful. My interviewees in Havana must remain anonymous.
Martin K. Dimitrov is Associate Professor of Political Science at Tulane University. His books include Piracy and the State: The Politics of Intellectual Property Rights in China (Cambridge University Press, 2009); Why Communism Did Not Collapse: Understanding Authoritarian Regime Resilience in Asia and Europe (Cambridge University Press, 2013); and The Political Logic of Socialist Consumption (Ciela Publishers, 2018). He has held residential fellowships at Harvard, Princeton, Notre Dame, the University of Helsinki, the American Academy in Berlin, and the Woodrow Wilson International Center for Scholars. He is currently completing a book entitled Dictatorship and Information: Autocratic Regime Resilience in Communist Europe and China.
Betts, Paul. 2010. Within Walls: Private Life in the German Democratic Republic. New York: Oxford University Press. DOI: https://doi.org/10.1093/acprof:oso/9780199208845.001.0001
Bos, Ellen. 1993. Leserbriefe in Tageszeitungen der DDR: Zur “Massenverbundenheit” der Presse 1949–1989. Opladen: Westdeutscher Verlag. DOI: https://doi.org/10.1007/978-3-322-93595-3
Castro, Raúl. 2011. “Informe Central presentado por el compañero Raúl.” http://www.granma.cubaweb.cu/secciones/6to-congreso-pcc/artic-04.html, April 17, 2011.
Castro, Raúl. 2016. “Informe Central presentado por el compañero Raúl.” http://www.cubadebate.cu/noticias/2016/04/17/informe-central-al-vii-congreso-del-partido-comunista-cuba/#.V-ltSJMrL_Q, April 16, 2016.
Cook, Linda J., and Martin K. Dimitrov. 2017. “The Social Contract Revisited: Evidence from Communist and State Capitalist Economies.” Europe-Asia Studies 69 (1): 8–26. DOI: https://doi.org/10.1080/09668136.2016.1267714
Dimitrov, Martin K. 2013. “Vertical Accountability in Communist Regimes: The Role of Citizen Complaints in Bulgaria and China.” In Why Communism Did Not Collapse, edited by Martin K. Dimitrov, 276–302. New York: Cambridge University Press. DOI: https://doi.org/10.1017/CBO9781139565028.015
Dimitrov, Martin K. 2014. “Tracking Public Opinion under Authoritarianism: The Case of the Soviet Union under Brezhnev.” Russian History 41 (3): 329–353. DOI: https://doi.org/10.1163/18763316-04103003
Dimitrov, Martin K. 2015. “Internal Government Assessments of the Quality of Governance in China.” Studies in Comparative International Development 50 (1): 50–72. DOI: https://doi.org/10.1007/s12116-014-9170-2
Dimitrov, Martin K. 2017. “The Political Logic of Media Control in China.” Problems of Post-Communism 64 (3–4): 121–127. DOI: https://doi.org/10.1080/10758216.2017.1318346
Dimitrov, Martin K. 2018. The Political Logic of Socialist Consumption [Politicheskata logika na sotsialisticheskoto potreblenie]. Sofia: Institute for Studies of the Recent Past and Ciela Publishers.
Dimitrov, Martin K. 2019. “Socialist Social Contracts and Accountability.” In Reforming Communism: Cuba in Comparative Perspective, edited by Scott Morgenstern and Jorge Pérez-López. Pittsburgh: University of Pittsburgh Press, forthcoming. DOI: https://doi.org/10.2307/j.ctvbqs9hf.8
Garcés, Raúl. 2014. “La actualización del modelo y la (des)actualización de la prensa: Consensos, disensos y silencios mediáticos en torno a la Reforma cubana.” CLACSO Observatorio Social de América Latina 14(36): 47–60.
Greene, Kenneth F. 2007. Why Dominant Parties Lose: Mexico’s Democratization in Comparative Perspective. New York: Cambridge University Press. DOI: https://doi.org/10.1017/CBO9780511509803
Hauslohner, Peter. 1987. “Gorbachev’s Social Contract.” Soviet Economy 3 (1): 54–89. DOI: https://doi.org/10.1080/08826994.1987.10641271
Hoffmann, Bert. 2016. “Bureaucratic Socialism in Reform Mode: The Changing Politics of Cuba’s Post-Fidel Era.” Third World Quarterly 37 (9): 1730–1744. DOI: https://doi.org/10.1080/01436597.2016.1166050
Inkeles, Alex, and Kent Geiger. 1952. “Critical Letters to the Editors of the Soviet Press: Areas and Modes of Complaint.” American Sociological Review 17 (6): 694–703. DOI: https://doi.org/10.2307/2088047
Inkeles, Alex, and Kent Geiger. 1953. “Critical Letters to the Editors of the Soviet Press: Social Characteristics and Interrelations of Critics and the Criticized.” American Sociological Review 18 (1): 12–22. DOI: https://doi.org/10.2307/2087843
King, Gary, Jennifer Pan, and Margaret E. Roberts. 2013. “How Censorship in China Allows Government Criticism but Silences Collective Expression.” American Political Science Review 107 (2): 326–343. DOI: https://doi.org/10.1017/S0003055413000014
Kuran, Timur. 1991. “Now Out of Never: The Element of Surprise in the East European Revolution of 1989.” World Politics 44 (1): 7–48. DOI: https://doi.org/10.2307/2010422
Lohmann, Susanne. 1994. “The Dynamics of Informational Cascades: The Monday Demonstrations in Leipzig, East Germany, 1989–91.” World Politics 47 (1): 42–101. DOI: https://doi.org/10.2307/2950679
Lorentzen, Peter. 2013. “Regularizing Rioting: Permitting Public Protest in an Authoritarian Regime.” Quarterly Journal of Political Science 8 (2): 127–158. DOI: https://doi.org/10.1561/100.00012051
Lorentzen, Peter. 2014. “China’s Strategic Censorship.” American Journal of Political Science 58 (2): 402–414. DOI: https://doi.org/10.1111/ajps.12065
Magaloni, Beatriz. 2006. Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico. New York: Cambridge University Press. DOI: https://doi.org/10.1017/CBO9780511510274
Millar, James R. 1985. “The Little Deal: Brezhnev’s Contribution to Acquisitive Socialism.” Slavic Review 44 (4): 694–706. DOI: https://doi.org/10.2307/2498542
Miller, Michael K. 2015. “Elections, Information, and Policy Responsiveness in Autocratic Regimes.” Comparative Political Studies 48 (6): 691–727. DOI: https://doi.org/10.1177/0010414014555443
Pan, Jennifer, and Kaiping Chen. 2018. “Concealing Corruption: How Chinese Officials Distort Upward Reporting of Online Grievances.” American Political Science Review 112 (3): 602–620. DOI: https://doi.org/10.1017/S0003055418000205
Philp, Mark. 2009. “Delimiting Democratic Accountability.” Political Studies 57: 28–53. DOI: https://doi.org/10.1111/j.1467-9248.2008.00720.x
Pravda, Alex. 1981. “East-West Interdependence and the Social Compact in Eastern Europe.” In East-West Relations and the Future of Eastern Europe: Politics and Economics, edited by Morris Bornstein, Zvi Gitelman, and William Zimmerman, 162–187. London: George Allen and Unwin.
Smith, Benjamin. 2005. “Life of the Party: The Origins of Regime Breakdown and Persistence under Single-Party Rule.” World Politics 57 (3): 421–451. DOI: https://doi.org/10.1353/wp.2006.0004
Torres, Ricardo. 2016. “Economic Transformations in Cuba: A Review.” Third World Quarterly 37 (9): 1683–1697. DOI: https://doi.org/10.1080/01436597.2016.1177454
Wintrobe, Ronald. 1998. The Political Economy of Dictatorship. New York: Cambridge University Press. DOI: https://doi.org/10.1017/CBO9781139174916