The concept of a culture of opposition (CO) under repressive political systems, most fully developed by Scott (1990), alerts us to the importance of culture, particularly language, other collective symbols, and disguised social norms and values used by persecuted people to adapt to undesirable political situations, to facilitate the occurrence of protests and other forms of collective action, and that are enriched by them (see for example, Johnston, 1991 on the Catalonian resistance to Franco; Canel, 1992, for Uruguay under the 1973-84 military dictatorship; Fantasia and Hirsch, 1995, for the Algerian Revolution; for the labor movement see Fantasia, 1988; on the related concept of movement culture see Lofland, 1993).
As Scott (1990) points out, cultures of oppositions are social organizations, with cultural codes and patterned, ongoing structures of social relationships. They give expression to an imagined, anti-hegemonic viewpoint of oppressed people, representing socially restricted collective reactions to shared indignities. They are punctuated by the action of “heroes” who dramatize the grievances of collectivities and are fueled by the elaboration of alternative ideologies and the occurrence of shared experiences of domination. Cultures of opposition are located in unstable social spaces to the extent that they are temporarily insulated from the control of the dominant group and peopled by trusted, known others. The participation of people in the activities and shared experiences of the CO creates individual and collective alternative identities. The CO creates and sustains anti-hegemonic interpretations of events and standards of rights among people. It affects their daily activities and social organization. It creates and preserves a collective memory of their experience with prior collective actions. It teaches participants how to deal with the system of repression to which they are exposed.
Not all hidden acts of political discontent are connected to and enacted in COs, nor are such acts necessarily linked to collective behavior such as protests and other forms of oppositional collective surges (Lofland, 1993). The CO is composed both by the explicit, often organized political dissidence of a nation state as well as the countercultures and subcultures in the society to the extent that significant proportions of these communities evince an oppositional political consensus of opinion regarding the desirability of social change. Although not all subcultures and countercultures have such oppositional political consensus, many deviant acts follow deviant norms that are consciously understood by their enactors as expressions of political dissent.
As developed here, the concept of CO is analogous to M. Weber’s concept of political parties in situations of political repression. Empirically, COs are composed of people who participate to varying extent in the organizing of alternative realities the CO represents but who may or may not be members of formally organized collectivities; they are parts of a social movement without social movement organizational counterparts (Danns, 1986). They would like to exert control of the bureaucracy of the state to bring about social change but cannot act openly without fear of state repression as political parties usually do in democratic polities. Whatever their differences, members of COs are opposed to varying degrees to the government, to the political party or parties represented by the government, and to its policies and programs. A culture of opposition perspective on protest provides a useful corrective to the overemphasis on social movement organizations as the primary mechanism bringing about social and political change (Danns, 1986).
Analytically, COs are different from and yet interrelated with general political culture, cultures of social movements, and civil society. The general political culture of a nation-state impacts the characteristics of its CO. In turn, the CO of a nation-state shapes the cultures of social movements and social movement organizations that are often parts of it (Fine, 1995; Lofland, 1995). Conversely, the cultures of social movement organizations help create and transform COs which in turn bring about changes in the general political culture (Fantasia and Hirsch, 1995). While the strategic emphasis in this paper is on cultures of opposition and their measurement and correlates, very fruitful research is done at the other three levels of analysis. The concept of civil society identifies a dimension of CO, for it connotes the relative degree of autonomy of society from the state and its political instrumentalities.
THE ATTRIBUTES OF THE CULTURE OF OPPOSITION
While Scott (1990), among other scholars, identifies the presence or absence of qualities said to typify COs, it is nevertheless possible to turn these qualities into continuous variables, or key dimensions of the COs of national political systems to form with them a classificatory multidimensional space. Ten of these key continuous variables are: (1) the proportion of the population of a nation-state that participates in the CO; (2) relative ease of communication among members of the CO; (3) strength and viability of the collective memories of heroes, heroic acts and instances of collective suffering in the CO; (4) number and variety of places in a society in which the CO is practiced; (5) degree of connectedness of the CO with institutions and organizations in the society which are independent of the control of the state; (6) extent to which groups and organizations in the CO are involved in international cooperative projects and activities with other national and international associations and movements; (7) presence in the CO of iconic leaders with wide followings in the society; (8) degree of conceptual sophistication of the ideologies of resistance articulating the values and goals of the CO; (9) average degree of knowledge of these ideologies among participants; and (10) degree of connectedness of the CO to the central constitutive historical experiences, beliefs, values and myths of the nation.
In one of the regions of the space formed by these variables are COs with high loadings. A historic example would be Poland’s CO and the centrality in it of the Catholic Church and the Solidarity Movement immediately prior to the collapse of the communist government. At the other extreme are societies like Cuba. Its CO belongs to a different set with comparatively low scores in many of these variables and in which strong, politically influential social movement organizations do not occur (Aguirre, 1998). It is a culture of opposition with low connectivity (Scott, 1990, see next).
Scott’s Domination and the Arts of Resistance (1990) presents a sophisticated model in which social control systems, treated as an exogenous variable in the model, impact the hidden transcripts of oppressed and repressed sub-communities, the likelihood and form which the open expression of their hidden transcript takes, and the type of collective action that may materialize. While in the main Scott’s work does not emphasize the diffusion of oppositional frameworks, nevertheless, he discusses the transformation of the hidden transcripts into open expressions of political defiance. The third endogenous variable in the model—types or forms of collective action that may result from the transformation of shared hidden political grievances into open expression of dissent— receives relatively less sustained treatment, although logically it is the outcome of the model.
One of Scott’s contributions is to document that much is going on when oppressed people are not openly in protest. But he also specifies (pp. 223-224) the conditions in which a larger population of aggrieved people members of various sub-communities adopts a shared understanding of overt acts of dissent. Such conditions are that they share the same subordination regime, that the various sub-communities share similar hidden transcripts, and that they think of themselves “as acting within some larger frame of reference,” such as nation, language, and religion. Once these conditions are met, there is, metaphorically, “a single power grid…(although not every) declaration of the hidden transcript will ramify through the entire grid …the grid itself, as defined by the hidden transcript, delimits the maximum possible symbolic reach of such acts, the population for whom such acts carry comparable meaning” (p. 224).
The rest of this paper considers the distribution of this open transcript in the population. Its objective is to offer an answer to the question of how is knowledge of the culture of opposition distributed in Cuba and what are the correlates of such knowledge. Can culture of opposition theory predict the distribution of such knowledge? The results show that it does. The next sections discuss the characteristics of Cuba’s CO and the hypotheses guiding the research. Sections presenting the methods and results of the investigation follow them.
QUESTIONS IN SEARCH OF ANSWERS
Cuba is a good test case of the power grid hypothesis, for it satisfies at least two of the three conditions imposed by Scott for a culture of opposition to be widespread in the population: (1) people in Cuba share a relatively homogeneous culture; (2) think of themselves as a nation; and (3) are subjected to a nationwide and long-reigned system of political domination. However, in contrast to societies where the general political culture, the mass media, and the electronic networks provide people with ready sources of multiple information about social and cultural life at home and abroad, Cuba’s state-created and guided political culture and effective system of social control (Aguirre, 2002) makes it difficult for people to know about events and social actors that are considered by the government to be contrary to its interests. To paraphrase Scott (1990, p. 223), in Cuba, as well as in other contemporary state polities, the “unveiling of the hidden transcript” may not have widespread repercussion in public opinion, for the unveiling may go unnoticed; the “grid” is ineffective, disjointed, lacking in connectivity. Thus, it is this context of political repression in Cuba—in which knowledge of anti-hegemonic acts and actors is censored, and in which many people, long subjected to the reigning ideology, are indifferent to the state-controlled public life—that provides the setting for the empirical question driving this paper, namely, how is knowledge of the culture of opposition distributed in the population and what are the correlates of such knowledge?
Six hypotheses, derived from the work of Scott and other scholars in the CO tradition, guide the empirical analysis. The relationship between the dependent variable in this study, knowledge of the culture of opposition, and its covariates are not causal but functional, for the CO theoretical framework assumes that there are feedback effects among these constructs. The hypotheses represent likely dimensions around which shared understandings of hidden transcripts of sub-communities can be assumed to materialize.
1. Knowledge and participation in the culture of opposition facilitate the social and psychological detachment of people from the prevailing system of domination. The prediction is that knowledge of the culture of opposition will be positively associated with the extent to which respondents feel repressed by the state and with their tendency to perceive fellow citizens to be opposed to the government’s repressive tactics.
2. Knowledge of the culture of opposition should also be positively associated with the extent to which respondents adopt dissidents’ self-conceptions and participate in dissident organizations. As I have argued previously, political dissidence is not necessarily equivalent to culture of opposition, but it is reasonable to expect an overlap between the two, such that one supports the other.
3. Access to news from abroad that has not been censored by the government should also be positively correlated with knowledge of the culture of opposition. As stated earlier, the CO springs from conditions of political repression where the government censors such knowledge, so that access to uncensored information should facilitate the knowledge of the culture. Conversely, participation and knowledge of the CO can be expected to facilitate access to uncensored news.
4. Respondents who are more highly integrated with the organizations of the state—who are more invested in orthodox political life—should have less time, opportunity, and desire to interact with people and arrangements that would give them information about the culture of opposition.
5. Since cultures of opposition are presumed to foster critical views of existing systems of domination and alternative interpretation of social life, it is also reasonable to expect respondents who are critical of the state and its leaders, and who are unhappy with other aspects of society and culture, to be more knowledgeable of the culture of opposition than their counterparts who are not critical of political orthodoxy.
6. It is not possible to predict the relationship between knowledge of the culture of opposition and the socio-economic status of people subjected to political repression. Scott assumes a negative relationship in his examples using slavery, such that economically marginal people will be more prone to know and participate in the culture of opposition. It is possible, however, to argue that well-to-do people have greater cultural resources at their disposal to obtain knowledge of the culture of opposition.
This study is based on information collected during December 1998 to April 1999 from 1023 recently arrived Cuban immigrants who had been in the United States for 90 days or less (for the methodology used in the original survey see Roberts et al., 1999, pp. 11ff.; Roberts, 1999; for a discussion of the univariate findings see Betancourt and Grenier, 1999). The Measuring Public Opinion Project (MPOP) used a convenience sample of Cuban émigrés to describe their perceptions of the Cuban government and their experiences in Cuba (for critical comments on the original use of the sample to generalize to the entire population of Cuba see Seligson, 1999). Because of important social and demographic differences among earlier and the most recent Cuban immigrants to the United States, it is not possible to generalize the results of this study to the former category of Cuban foreigners. Moreover, rather than generalizing them to the total population of Cuba, the results of our statistical model using MPOP can be generalized only to the population of would-be immigrants in Cuba, or about 20 percent of the total population of the island.1
Knowledge of culture of opposition (0-15, alpha=.78) is the dependent variable in this study. It is a seven item scale that includes information on explicit acts of dissent and on leaders and organizations of the dissidence: a. whether respondents had knowledge of independent union leaders; b. personally knew Cuban dissidents; c. knew about the “Support for a Democratic Transition in Cuba” document released by the Clinton administration in 1994; d. knew the substance of “The Fatherland Belongs to All” declaration by four prominent Cuban dissidents; e. knew about the 1998 street demonstration in support of dissident Reinaldo Alfaro; f. could identify eight major leaders of the dissident movement in the island; and g. knew of the existence of the “Cuban Council” (an organized, peaceful important effort to bring about political and social change). As shown, this scale taps knowledge of the general culture of opposition that prevailed in Cuba rather than of hidden transcripts of localized micro types of resistance.
Six dimensions are used in this test to group the sixteen covariates included in the OLS models to test the hypotheses.
The first two items measure dimensions of oppositional politics. The first of these items measures the Extent of repression felt immediately prior to leaving Cuba. It ranges from 0 to 7. Respondents were asked if they felt the presence of state repression in streets; neighborhoods; work places; schools; because of the activities of state security; due to the repudiation acts by government-led mobs against dissidents; and for other reasons. The second item measures if their perception is that People are against repudiation acts (1. yes, 0. No).
A second dimension included in this test is the respondents’ perceptions and activities vis-à-vis the dissidence in the island. The third item asks if respondent Considers her or himself a dissident (1. Yes, 0. No). The fourth asks if respondent Belongs to dissident organizations (0=no, 1=1, 2=2 or more). The fifth asks their views about whether People respect dissidents (1. Yes, 0. No).
A third dimension includes mass media use indicators. The sixth item measures the Extent of dependence on state media for news (0-9, high). The item measures respondents’ dependence on national radio, television and press to receive the most complete information about recent important events (these events were the political crisis surrounding the execution of General Ochoa; the mass migration of the Guantanamo boat people; the attack against Brothers to the Rescue airplanes; changes in 1994 migratory agreements with the US; the Malecon crisis and protests of 1994; the crisis in tonnage production in the recent sugar harvests; the approval of the Helms-Burton Law; and the sinking of the “13 de Marzo” tugboat). The seventh item measures the Source of News, (ranging from 0 to 5, high, to indicate respondents who used only national sources of news). This variable includes information about the extent to which the respondents got their information from national radio, foreign radio, national newspapers, foreign newspapers, and watched TV Marti (compare to Schock, 1999).
A fourth dimension includes two indicators of integration with the state. The eighth item taps Integration with state organizations scale (0-6, standardized item alpha=.75). It is a three-item scale that includes information as to whether the respondents were members of the CTC (official labor union); worked in the government sector; and belonged to the 7 most important mass organizations of the state. It is important to stress that this measure does not necessarily imply respondents’ degree of acceptance of the legitimacy of the political regime in place. The ninth item measures whether respondents Belonged to an official organization prior to leaving Cuba: (1. Yes, 0. No).
The fifth dimension represents evaluation of life in Cuba. The tenth item is A problem of daily living scale (0-6, standardized item alpha=.85). It is a six items scale measuring whether the respondents were bothered by the two-faced attitude of everyone in daily life; absence of food and other necessities; problems with transportation; religious freedom; the corruption and privilege of government officials; and the preference for foreigners and the discrimination against Cubans. The eleventh item is a Hatred of all leaders and organizations scale. (0 to 2, standardized item alpha=.74). This scale scores as 2 respondents who hated all 12 top government leaders as well as the 15 most important regime organizations (for the names of the leaders and organizations see questions 78 and 79 in Roberts et al., 1999). The twelfth item presents information about whether the respondents thought there had been Accomplishment in health and education (0=yes, accomplishment in both areas, 1=mixed, 2. No accomplishment in either area; for an extended treatment of both state programs during the present economic crisis see Pearson, 1997).
The sixth dimension included in this test measures the economic situation of the respondents. The thirteenth item is a measurement of the respondents’ Monthly family income in Cuban pesos (1. <164 pesos, 2. 164-300, 3. 301-450, 4. >450). The fourteenth item is a scale measuring Contacts with family abroad and gifts of money scale (0-4, standardized item alpha=. 64). It includes two items: the number of telephone contacts with relatives living abroad, and whether respondents received dollars from their kin living abroad. As in other countries in the developing world, in Cuba the receipt of financial assistance from kin living abroad is a very important source of income for many families (see Díaz-Briquets, 1997). The fifteenth item tapping the economic status of the respondents is a scale measuring the Receipt of humanitarian aid and assistance from Caritas (0-2, standardized item alpha=.87). The sixteenth item in the test presents information about the work of the respondents and its “distance” from accepted practice. It measures whether the respondent is Self-employed and breaking the law (0=Not self employed, 1=Yes, and licensed by state, 2=Yes and without state license).
Nine statistical controls are also included in the OLS analysis. It is reasonable to assume that they may be related to both the dependent variable and the covariates. The first is the Age of respondent, grouped in seven categories (1=15-20, 2=21-30, 3=31-40, 4=41- 50, 5=51-60, 6=61-65, and 7=65>. The second is race, scored 1 for white and 0 for blacks (includes black, mulatto, mixed races). The third control is whether they came to the United States with Lottery visas (0=Yes, 1=Other). The fourth is Catholic (0=Yes, 0=Other). The fifth is their formal Education (1. < high school (HS), 2. HS graduate, 3. Pre-college, Technical school, 4. University and Graduate training). The sixth is whether they had Havana Residence (0=Yes, includes city and province of Havana, 1=Other areas of the country). The seventh is their Present marital status (0=Married, 1=Other). The eighth control is Gender (1=Male, 0= Female), and the ninth is Months since emigrating from Cuba (1-25 months).
There is evidence showing that young males are very often the target of state repression and may be more prone to know and participate in the CO. Catholics have been historically a persecuted category of religious believers. Havana, due to its status as capital of the country and the most important cultural hub, is the most prominent site of anti-government incidents and thus it can be assumed that its residents are better informed about the CO. Blacks protest much more now than in the past, in part due to their relatively greater inability to access dollars during the present economic crisis, the collapse of government programs, and the resulting economic difficulties (Aguirre and Bonilla Silva, 2002) and perhaps this activism is related to knowledge of the CO. Finally, the inclusion in the model of Months since emigrating from Cuba is also necessary to control for the possible effect of the passing of time on the accuracy of the recall of the respondents about their experiences, knowledge and attitudes while in Cuba.
Cuba’s Culture of Opposition
Cuba’s political system is in a pre-transitional stage (O’Donnell and Schmitter, 1986; Munck, 1994). Its political dynamics are very different, for example, from those of East Germany and Czechoslovakia in 1989, in which the worker brigades refused to back the police against the demonstrators (Oberschall, 1994). It is similar to Bulgaria and Romania before the disappearance of the USSR (Chilton, 1994; Linden, 1995) and to contemporary events in China, in which the state attempts (with mixed success, see Locay, 1995; Centeno, 1994) to keep political power while managing economic and social change. So far, there has been no access of challenging groups to political institutions, a weakening of the state’s repressive apparatus and the ideological opening of the Castro regime (Gras, 1994). The top leadership of the government remains stable. There have been no advances in the rights of categories of persons or the enforcement of constitutionally protected citizenship rights. Changes in criminal laws used by the security system to justify its activities have not occurred (Alfonso, 1994; Hidalgo, 1994, pp. 292-299; Murray, 1994). Instead, in the aftermath of the 1991 Fourth Party Congress, the Cuban Communist Party (CCP) became more homogeneous and pliable to the policy dictates of Fidel Castro (Bengelsdorf, 1994, pp. 169- 173; Domínguez, 1994; del Aguila, 1994). Alternative ideologies to those of the state are undeveloped and relatively unknown by the Cuban people. The extent of appropriation by political dissidents of the constitutive myths and symbols of the Cuban nation is incomplete, contested by the nationalist ideology of the state and its organs of social control. Modally, sub-communities whose ideological voices have not been sufficiently articulated and broadcast inside Cuba have embodied its CO. Nor have iconic dissident leaders been able to operate for long in the island. Voluntary organizations and institutions independent of the state that could provide support to it are mostly absent. Organized collectivities have until recently lacked systematic, ongoing cooperative contacts with organizational affiliates outside Cuba. It is a culture of opposition that facilitates the occurrence of loosely structured, un-institutionalized forms of collective action (Aguirre, 1998).
Knowledge of the Culture
Figure 1 presents the distribution of the respondents in the knowledge of the culture of opposition scale. As expected given the hypothesized effects on knowledge of the culture of opposition of social control processes and the widespread alienation of people from public life, the distribution is negatively skewed, for the vast majority of respondents has a rather incomplete knowledge of the CO: 26.5 percent had no knowledge whatsoever of the CO, while 50 percent scored 3. Only 87 respondents scored 12 or more in this scale. This state of generalized ignorance of the culture of opposition can be hypothesized to be partly the result of the great effectiveness of the systems of social control in Cuba (Aguirre, 2002) and of a generalized sense of disinterestedness in public affairs and distrust of public institutions in the population (for the case of Russia see Rose, 1995).
Table 1 presents partial ordinary least squares (OLS) regression models in which the variables included in each of the dimensions previously identified are included separately, a full OLS model in which all of these variables and controls are included, and a stepwise reduced OLS model in which only the statically significant variables are retained. There is hardly any variation in the results from these various models, so that the discussion of the findings is limited to the reduced OLS model.2
Four of the six hypotheses previously identified receive support from this test.
• As predicted, knowledge of the culture of opposition is positively associated with the extent to which respondents feel repressed by the state (B=.06) and with their tendency to perceive people to be opposed to the government’s repressive tactics (B=.12). Thus, a change of one standard deviation (1.94) in extent of repression is associated with a change of .28 of a standard deviation (4.59) in the knowledge variable; and a change of one standard deviation (.50) in the perception that people are opposed to repressive tactics is associated with a .55 of a standard deviation in the knowledge variable.
• As predicted, knowledge of the culture of opposition is positively associated with the extent to which respondents develop dissidents’ self-conceptions (B=.14), participate in dissident organizations (B=.29), and perceive others as respecting the dissidents (B=.16). The corresponding changes expressed in terms of fractions of a standard deviation of the knowledge variable are .64, 1.33, and .73, respectively.
• As predicted, lack of access to news from abroad, measured as dependence on state media, is also negatively correlated (B=-.16) with knowledge of the culture of opposition. It is a change of -.73 if expressed as a fraction of a standard deviation of the knowledge variable.
• Also supporting the prediction, respondents who belonged to state organizations prior to emigrating from Cuba had less knowledge of the culture of opposition (B=-.09). One change in this covariate’s standard deviation (.49) is associated with a change of -.41 expressed as a fraction of a standard deviation in the knowledge variable.
• As predicted, respondents who hated the leaders and organization of the Cuban state (B=.08), and who thought that there had not been advances in public health and education (.07) were also more knowledgeable of the culture of opposition. The corresponding changes in the knowledge variable, expressed as a fraction of a standard deviation are .37 and .32.
Contrary to the prediction, “problems with daily living” was negatively associated with knowledge of the culture (B=-.16). One increase of a standard deviation (2.24) of this covariate is associated with a change of -.73 standard deviation of the knowledge scale. Reasoning after the fact, it may be that contrary to the theory, people who face great difficulties and who are engaged in marginal, quasi-illegal occupations (see below) do not have the time and resources to learn about the general culture of opposition.
Surprisingly, indicators of the economic situation of respondents were not statistically significant covariates of knowledge of the culture of opposition. This set of covariates had the smallest partial adjusted r square (.02) and proved statistically insignificant addition to R square if entered as a block of variables in a hierarchical solution (not shown). The exception to this lack of statistical significance is “self-employment and illegal work activities”, which was negatively associated with knowledge of the culture (B=-.06, sd=.71). Importantly, knowledge of the culture is less likely among this category of marginal people. A finding also supporting this conclusion is that education, one of only four statistically significant controls, is positively associated with knowledge of the culture (B=.09), so that an increase of one standard deviation (.99) in education is associated with an increase of .41 of a standard deviation in the knowledge variable.
Finally, knowledge of the culture of opposition also increases with age (B=.12) and if respondents were white (B=.07) and had not come to the U.S. with lottery visas (B=.07) (for an extended examination of the importance of the lottery of visas carried out by the U.S. Department of State in Cuba see Aguirre and Bonilla Silva, 2000).
The results of this research document that the vast majority of respondents had very incomplete knowledge of the CO scale used in this research; approximately one-fourth had no knowledge whatsoever of it. Moreover, the results also show the usefulness of culture of opposition theory in identifying and predicting the relationships among a set of covariates and knowledge of the culture of opposition in Cuba. Knowledge of the culture of opposition is: a. positively associated with the extent to which respondents feel repressed by the state and with their tendency to perceive people to be opposed to the government’s repressive tactics; b. is positively associated with the extent to which respondents develop dissidents’ self-conceptions, participate in dissident organizations, and perceive people as respecting the dissidents; c. negatively correlated with lack of access to news from abroad; d. negatively associated with membership in state organizations prior to emigrating from Cuba; e. positively associated with hatred of the leaders and organization of the Cuban state and if respondents thought that there had not been advances in public health and education. Contrary to the prediction, however, knowledge of the culture declines among respondents who faced great difficulties and were engaged in marginal, quasi-illegal occupations.
It is useful to think of these findings regarding the impact of oppositional politics, dissidence, mass media use, integration with state, evaluation of life in Cuba, and the economic situation of respondents as indicating points in Scott’s grid, inflections related to life styles and life chances which facilitate or impede the knowledge of the culture of opposition. Or, expressed differently, at present the various sub-communities in the Cuban nation subjected to the domination of the present-day government do not share a uniform hidden transcript, and these dimensions, derived from culture of opposition theory, helps us understand the distribution of such knowledge in the population.
It is plausible to assume that the prevailing lack of knowledge of the CO is partly explained by the special nature of social control systems in Cuba (Aguirre, 2002; Mahoney-Davis, 2000). The basic quality of these systems of control is a combination of both formal and informal arrangements and practices that simultaneously emphasize openness and rigidity (See Figure 2). Both formal and informal systems are guided by an ideology, a long-reigning political culture socially engineered by the state, and the operations of a centrally planned society. Elements of the informal system are education, residential patterns, humor, myths and rituals, and charismatic authority. The formal system is geared to block all anti hegemonic acts of individuals and organizations, particularly those acts that if left unchecked could become symbolic acts encouraging similar patterns of behaviors perceived as undesirable by the authorities (compare to Sztompka, 1994). Greater importance is given nowadays to reactive rather than proactive approaches—as exemplified by rapid action brigades formed by the security system to repress people. This system of social control thrives in the indifference of the citizenry to matters of public interest such as those tapped by our CO scale.
The generalized indifference to matters of public interest as well as the prevailing lack of knowledge of the culture of opposition among Cubans (in the MPOP and generalizable to a population of approximately 2 million would-be emigrants in the island) explain in part the absence of sustained organized protests in the island. Previous research has shown that the most frequent forms of collective action occurring in Cuba exemplify loosely structured, un-institutionalized collective action (Oberschall, 1980). These forms are the heroic actions of protest by individuals, prison and food riots, brutality-generated protests, mass behavior, and rumors. Movement organizers or entrepreneurs do not usually plan them, nor can these activities be accurately described as carried out by transitory teams led by the staff of social movement organizations.
Indeed, most social movement organizations in Cuba have a small number of members, are restricted to a city or region, and are the targets of very effective state repression. They sponsor a pacifist ideological stance, engage in petitioning the government to allow for peaceful change to take place, and try to convince members of the ruling elite about the need for a multiparty political system and respect for human rights. So far, their strategy has failed to have any measurable effect in bringing about significant changes to the political system.
Nevertheless, the organized collective action of dissidents in Cuba and among Cubans in South Florida, Puerto Rico, and other communities of the Cuban Diaspora is facilitated by the new means of mass communication such as the Internet and the recently emerged transnational system of nongovernmental voluntary organizations (Chilton, 1994). This increase in cooperation with, and resource acquisition from transnational voluntary and social movement organizations may precede a more important role for social movement organizations in the future (Aguirre, 1998).
The Cuban case study suggests the value of the study of COs in national polities dominated by repressive state systems. An emphasis on culture is necessary if we are to understand people’s knowledge of alternative realities in repressive political systems as well as the repertoires of forms of collective action that take place in these polities in the absence of the activities of social movement organizations (Danns, 1986). Thus, this paper emphasizes, as do Piven and Cloward (1978; 1992) and Fox and Starn (1997, p. 3), the importance of the “cultural politics of protest” and of emerging cultural understandings of the subjective meanings of anti-hegemonic collective political participation (see Piven and Cloward, 1978; Fox and Starn, 1997). Alternative available conceptualizations do not seem as fruitful.
1. Seligson (1999, p. 250) points out that the MPOP may be susceptible to a “social desirability response set” in which the respondents are prone to answer in ways that they think will please their U.S. based interviewers; they may wish to “ingratiate themselves with their new society.” As a reliability-based measurement error, this problem is minimized in the present analysis through the use of multi-item scales to measure most of the most important concepts used in it. While questions of validity are more difficult to resolve, a scrutiny of the Pearson correlation matrix shows that the variables relate as they should based on what we know about Cuban society. Thus, reflecting what is known about immigration to the United States from Cuba, Black respondents come with lottery visas more often than White respondents (Pearson’s r=.23), in part due to their lack of family relations in the U.S.; married respondents are older than single respondents r=.25), the greater the respondents’ education the greater their income (r=.20); the greater their integration with state organizations the lesser their self employment without license (r=-.25; see Puerta, 1999). In contrast to the original investigators, we do not wish to estimate specific values for population parameters from the MPOP sample estimates; we do not weigh the cases to correct for the overrepresentation of whites or similar problems with the sample when compared to characteristics of the population of Cuba (Comité Estatal de Estadísticas, 1991, pp. 24-30). Instead, we wish to ascertain an analytical rather than a descriptive question, namely the extent to which a set of theoretically specified predictors can help us understand the distribution of knowledge about the culture of opposition among Cubans in the island.
The results are largely invariant (not shown, available upon request) if the MPOP sample of respondents is restricted to respondents who came to the U.S. with lottery visas. This reproducibility establishes the applicability of our findings based on MPOP to the population of would-be immigrants in Cuba, for immigrants with lottery visas are randomly selected by U.S. officials in Cuba from a much larger population of petitioners participating in the Cuba Visa Lottery Program instituted soon after the migratory accord of 1994 between the two governments (Travieso-Diaz, 1998). While the precise size of the present day population of would-be immigrants in Cuba is unknown, it is certainly a very large number of people. At the beginning of the lottery system in 1994, approximately 189,000 Cubans applied. In the second lottery, in 1996, there were approximately 435,000 applications. During the last lottery, in 1998, more than 525,000 applied. Even if we assume a somewhat smaller rate of growth, the next lottery that will take place in 2001 or 2002 may include approximately 800,000 to 1,000,000 applicants from a population of 11 million. Moreover, this relatively large number would need to be doubled to roughly two million to include the segment of the would-be population that does not apply to the lottery but still wishes to leave Cuba. This excluded segment is large although perhaps diminishing in importance as a result of the recent migratory accords. Rodríguez Chavez (1997) estimates that of the 47,500 Cubans who arrived in the United States during 1990-1993, only 10 percent had obtained immigrant U.S. visas before leaving Cuba. Instead, 20% came as refugees, 17% came from third countries, 20% came in rafts or through the U.S. base in Guantánamo, and 30% came as visitors who then decided to stay. Indeed, Urrieta Barroso (1997) estimates that in 1994 there were 11,222 legal and 13,270 illegal immigrants from Cuba to the United States, or about half of the total, again furnishing grounds to double the estimate of the size of the population of would-be emigrants. Empirically, some key characteristics of the MPOP sample are close to roughly known parameters of the population of would-be immigrants. Thus, 46 percent of the respondents lived in Havana. Montiel Rodríguez (1996, 264) estimates that in the 1990’s, 60 to 70 percent of Cuban emigrants had resided in Havana and its environs. Since Havana accounts for approximately 26 percent of the population of the island (Morejón Seijas and Molina Soto, 1997), it is fairly certain that there is a chain migration process taking place in which a large proportion of would-be emigrants move to Havana and later emigrate from it. Indeed, Díaz-Briquets and Pérez-López (2000) have recently shown that during periods of rapid emigration from Cuba, the population of Havana and its environs declines temporarily only to be replenished by gains from internal net migration. Males accounted for 59 percent of the sample. The largest age group, accounting for 93% of the respondents, was 15 to 50 years of age. Both characteristics correspond with what is known about the gender and age selectivity of the would-be immigrant population (Morejón Seijas and Molina Soto, 1997; Rodríguez Chávez, 1997; Urrieta Barroso, 1997). Eighty-seven percent of the MPOP sample was white, again reproducing the known tendency for white Cubans to immigrate in larger proportions than blacks (Urrieta Barroso, 1997; Rodríguez Chávez, 1997; Ramírez Calzadilla, 1997; Guanche Pérez, 1997). Moreover, the sample was relatively well educated. Forty percent had pre-university training, 22% had technical knowhow, and 19% had university or postgraduate degrees (Roberts et al., 1999, p. 13; to compare to the general population, see Espina Prieto, 1997). Again, as in other would-be immigrant populations we see an educational selectivity operating here as well.
2. The assumptions of OLS multivariate regression are met. There is very little multicollinearity impacting the results. The average VIF is 1.23, and all VIFs are below 1.50. All tolerances with the exception of three (in the upper .60s) are above .80. There are 39 cases—or 3.8 percent of the 1021 respondents—with standardized residuals greater than 2 standard deviations, well within the 5 percent of cases expected to have residuals outside two standard deviations. All have Cook’s distances below .05. The OLS model’s average leverage is .025. Only one standardized residual (case no. 851) has a leverage value twice as large (.0593). The Durbin-Watson statistic is 1.60, supporting the assumption that the residuals are independent. The plot of standardized residuals against standardized predicted values show that the assumption of homoscedasticity is also met. Moreover, the residuals are normally distributed. For a model with 25 predictors and 1021 respondents, Stein’s adjusted R square is .38. It is very close to the present model’s overall adjusted R square of .39 (see full model in Table 1) and provides validation of the accuracy of the findings across different samples (Field, 2000, pp. 103-162). Numerous attempts to identify statistically significant interaction terms failed (not shown), thus providing support for the assumption of additivity of OLS.