The Party or the Grassroots: A Comparative Analysis of Urban Political Participation in the Caribbean Basin by Alejandro Portes and Jos Itzigsohn The Johns Hopkins University October 1993 The purpose of this study is to examine the patterns of political participation and attitudes toward different forms of participation of urban low-income groups in five capital cities of the Caribbean Basin. Contemporary forms of popular participation in the Caribbean and, more generally, in Latin America represent major departures from the political experiences of the past in this region. These patterns are also at variance with the ways in which popular political participation is conceptualized in the advanced countries. The analysis of these novel trends and their determinants can cast light on processes so far unexplored in the contemporary literature on social movements. The emergence of what are known in Latin America as the "new social actors" took established political parties in the region as well as political analysts by surprise. These "new actors" comprise a wide variety of community and grassroots popular organization that have emerged spontaneously out of the needs of their respective constituencies and that, for the most part, have avoided entanglement with the traditional political organizations. Despite their dislike of "politics as usual," these grassroots organizations have come to play an increasingly important role in both local and national affairs through their organized demand-making, mass protests, and electoral support of like-minded figures (Jelin 1985). In Peru, for example, community organizations have eclipsed traditional political parties during the last two national elections and, in Colombia, a number of grassroots human rights organizations have sprung to provide a measure of protection for human rights and against widespread violence (Fals Borda 1992). This article will examine the extent of actual community grassroots participation among the population of five Caribbean capital cities as well as the attitudes held toward them. We will explicitly contrast this mode of participation with traditional political parties and will analyze the factors leading to one or another form of participation. The analysis will highlight the significance of different political systems and state policies in promoting or deactivating popular mobilization, a factor commonly neglected in studies of the topic in the First World. In the advanced democracies, the overarching institutional framework within which participation takes place is generally taken for granted. Since Lipset's Political Man (1963), the analytic focus of studies of popular participation has concentrated on individual-level variables such as class membership, level of education, gender, and family status which are theorized as key determinants of participation. Under the influence of rational choice theory, recent empirical work in this area has focused on the "free- rider" problem of collective movements and more generally on the individualistic calculus of rewards and costs that is assumed to underlie participatory decisions (Hechter 1987; Przeworski 1985; Nielsen 1986). Quite different is the situation of Third World countries where a range of political regimes--from relatively stable democracies to repressive dictatorships--plays a decisive role in the character and extent of popular political participation. The constraints imposed by the polity, allied to the hardships of economies of scarcity, have often led to innovative and vigorous forms of grassroots organization where the "problem" of free-riding and others of a similar bent, prominent in the research literature of the advanced democracies are moot (Portes and Johns 1989). The principal contribution of the following analysis is to bring into play both individualistic and national- level (political system) variables as potential determinants of the observed differences in actual participation and sympathy for community-based organizations. In order to accomplish this, it is necessary to examine such processes in a comparative framework involving a number of political systems. I. Theoretical Background The literature on Latin American urban movements has undergone a remarkable transformation since the days in which most writings on the topic emphasized the "urban explosion" and the widespread revolutionary fervor of the impoverished masses (Beyer 1967; Ward 1964). These theories were prompted less by direct observation of the behavior of the alleged radical groups than by the expectations of middle-class observers as to the political behavior appropriate for people living under such abysmal economic conditions. The vicarious imputation of seething fury to the "marginal mass," as urban popular groups were then labelled, found ample resonance in the printed media but precious few instances of empirical support. To the contrary, study after study weighed in with a very different portrayal of political behavior among urban low-income sectors. According to these studies, shantytown dwellers in Latin America were not about to lay siege to the cities that they surrounded in part because they were so preoccupied with their own survival. Instead of the dramatic instances of popular radicalism anticipated by journalistic and scholarly publications, researchers in the field found diverse and creative adaptations to the existing social order. These adaptations did include popular mobilizations and demand-making, but always within the bounds of established political parameters and always focused on specific goals (Leeds 1969; Mangin 1967; Eckstein 1977). This syndrome of "rational adaptation" to existing social and political conditions (Portes and Walton 1976: 109-110), included participation in traditional parties and effective manipulation of electoral politics for community advantage. The research literature that put to rest the myth of mass radicalism did not anticipate, however, that the adaptability of popular groups would go beyond taking part in existing political organizations. The shift toward new innovative forms of participation started in the 1970s and gathered momentum in the 1980s, in response to the rise of highly repressive authoritarian regimes, followed by a severe economic downturn. In Latin America, the debt-induced economic crisis led to a halt and then a reversal of economic development, as country after country registered negative growth rates during the early eighties. This decline, the worst since the Great Depression, was experienced with particular force among the low-income sectors (Iglesias 1985; Lagos and Tokman 1983; CEPAL 1990; PREALC 1990). During the seventies and early eighties, the situation was made worse by military dictatorships in many countries that blocked all institutionalized channels for the expression of grievances (Lehman 1990; Tironi 1986). In response, shantytown dwellers and other working- class groups began to create their own community-based organizations for survival and for the gradual expansion of alternative spaces for demand-making. To avoid repression, they resolutely shunned contact with the existing political parties and rallied instead around unimpeachable values: child health and education, mothers' need for care, access to basic food staples, shelter, and protection from crime (Cardoso 1983). Unlike trade unions, community organizations did not coalesce around workplaces as many of their members were jobless, but rather organized on the basis of common places of residence (Razeto 1985; Schkolnik 1986; Friedmann 1989). These neighborhood-based groups--mother's associations, youth centers, self-help housing cooperatives, communal kitchens, and so forth--proved so successful that they did not only endure the military dictatorships of the 1970s, but contributed effectively to their demise and then proceeded to expand to city-wide and national levels (Campero 1987; Hardy 1987; Matos Mar 1985). At this point, however, the literature on the "new social actors" became bifurcated into those authors who saw their growth as a sign of a qualitative shift in the relationship between civil society and the state and those who perceived them as no more than an innovative popular response to unusually difficult political conditions. Castells (1983), Slater (1985), and Friedmann and Salguero (1988) echoed the optimism of local social scientists, particularly in Chile, who regarded these new self-reliant community groups as the building blocks for the emergence of a truly democratic society. Others like Eckstein (1989), Portes and Johns (1989), and Cardoso (1992) were less sanguine about the transformative capacity of grassroots movements and viewed their growth as dependent on the spaces of opportunity created by the dominant political system. Contrary to the most enthusiastic exponents of the first position, who portrayed community-based movements as a new global wave transcending national differences, this second school regarded popular participation in these new movements as contingent on both individual-level variables and the national polity. The Latin American literature on the new social actors and the theoretical controversies surrounding it provide the context for our analysis of popular political participation. Instead of formal hypotheses, we opt for summarizing the key points of the above discussion in the following four questions: 1. What is the real extent of participation of urban popular sectors (defined below) in grassroots organizations versus traditional political parties? 2. What are the attitudes of these popular sectors toward both forms of participation? 3. To what extent are political attitudes and participation influenced by individual-level variables, such as those emphasized by the collective movement literature in the advanced countries? 4. To what extent are political attitudes and participation influenced by cross-national differences in political systems and, in particular, the character of the state? II. Data and Method The data for this study come from surveys conducted simultaneously in the capital cities of five Caribbean Basin countries during 1992. The countries selected were Costa Rica, Guatemala, the Dominican Republic, Haiti, and Jamaica. They comprise the largest island-nations of the Caribbean (with the exception of Cuba), the largest Central American nation, and encompass the most diverse political systems in both sub-regions. The five countries share a number of characteristics as part of the same broadly defined geographical area. These include their uniformly small size, the high concentration of the urban population in a single city, their proximity to the United States, and the profound dependence of their economies on the U.S. market. They are all members and beneficiaries of the Caribbean Basin Initiative (CBI). The Caribbean Basin was targeted for this study because of its geopolitical and economic importance as part of the immediate southern periphery of the United States and because of the extraordinary variety of political and economical systems encompassed in a relatively small area. The region includes the poorest and most politically unstable country in the hemisphere (Haiti) and the most stable democracy south of the U.S. border (Costa Rica). It also includes democratic parliamentary systems in the British model (exemplified by Jamaica) and military-dominated autocracies in the traditional Latin American mold (represented by Guatemala). These vast differences are directly relevant to the topic of study because they will allow a direct examination of the extent to which popular political participation in its alternative forms is affected by the national political system, controlling for individual- level variables. Table 1 presents a brief socio-political profile of the five countries included in the analysis. Although space precludes a detailed discussion of their histories, a brief outline of their respective political systems is necessary to clarify the character of their potential influence on popular participation: -- Costa Rica ranks as one of the most democratic countries in the continent where the state guarantees human and civil rights and encourages participation of the citizenry in electoral politics through the established parties. Although not actively encouraged by the state, community-based organizations are flourishing and, as legal groups, are granted ready access to the authorities (Lungo, Perez, and Piedra 1991; Vargas 1992). -- Guatemala is at the other extreme with a history of repression of popular organizations dating back to the CIA-sponsored overthrow of the elected Arbenz government in 1954. Since that time, political life has been dominated by an alliance between the armed forces and an entrenched economic elite. Paramilitary death squads have been used liberally to intimidate and eliminate opponents of the existing order. Although two successively elected presidents provided a semblance of democracy during the 1980s, the country is still politically unstable and continues to be dominated by the same elite/military alliance. Violations of human and civil rights are still widespread (Jonas 1991; Prez-S inz 1991). -- The Dominican Republic is an incipient democracy with a strong presidentialist regime. Since the U.S.-led invasion of the country to thwart a leftist takeover in 1964, the country has been governed by democratically elected presidents. Dominican political life has been dominated during the last two decades by the figure of Joaquin Balaguer, elected repeatedly as president. Under his administrations, political parties have been allowed to compete openly and the influence of the military on the government has waned. Although civil rights violations are still common, the situation has improved significantly. Community-based organization are legal and are seldom interfered with (Lozano and Duarte 1991; Dore-Cabral 1985). -- Neighboring Haiti continues its turbulent political life with a brief hiatus of democracy under president Jean Bertrand Aristide. As in Guatemala, an alliance of the armed forces with an entrenched economic elite has made use of any means to preserve its power. Death squads, known locally as Tonton Macoutes, have been used to systematically intimidate and do away with opponents. The Aristide electoral campaign gained widespread popular support and his election further stimulated popular mobilizations in hopes of transforming the old political order. After the coup that ousted Aristide, there was a sustained campaign by the military to deactivate community organizations and repress opposition parties (Manigat 1992; Trouillot 1990). At the time of this writing, international pressure has forced the military and traditional elites to acquiesce to Aristide's return, but the future of his U.N.- backed government is quite uncertain. -- The Jamaican parliamentary system is a legacy of the British colonial period. There has been an uninterrupted succession of democratically elected administrations, but elections are frequently marred by fraud and widespread violence. The two parties that alternate in power--the Jamaican Labor Party (JLP) and the People's National Party (PNP)--have large popular followings. However, armed gangs belonging to the two parties frequently stage violent confrontations in Kingston's shantytowns and other urban areas leading to a climate of widespread fear and instability (Gordon and Dixon 1991; Pinkow 1993; Palmer 1989). ------------------------------------------------------ Table 1 about here ------------------------------------------------------ Data collection for this study was conducted by local research teams under senior social scientists in each of the five selected cities. "Popular sectors" were operationally defined as those living in low-income areas of each city. To provide enough socioeconomic variation in the sample, sampled neighborhoods ranged from government housing projects and other areas with a population composed primarily of lower middle-income government employees and formal workers to peripheral shantytowns, where most inhabitants survive through work in the informal economy. Budgetary limitations prevented the organization of city-wide surveys; instead, neighborhoods deemed representative of the lower-middle to poor sectors in each city were selected for study. Within each neighborhood, dwellings were selected with a probability proportional to total size and in each dwelling the household head was interviewed. Four hundred interviews in three or four popular neighborhoods were conducted in San Jos (Costa Rica), Santo Domingo (Dominican Republic), and Guatemala City. In Kingston, the final sample reached 800 because it was necessary to cover the low-income population of the city itself and of the vast suburbs in the adjacent St. Catherine's Plain. In Port-au-Prince, a planned survey of three low-income areas was cut short by the coup that ousted President Aristide. The local research team managed, however, to complete a total of 300 interviews in these areas, mostly before the coup. The full sample thus consists of 2300 cases in five cities. The same questionnaire--translated into Spanish, English, and Haitian Creole--was applied in all city surveys to insure the comparability of results. The questionnaire included items on sex, marital status, migratory status, education, occupation, income, and class self- identification. Also included were membership in political parties, participation in community-based organizations, and attitudes toward both forms of political action. The last set of items comprises the dependent variables in the following analysis; the first set comprises the individual-level predictors whose effects vis--vis national political systems will be evaluated below. III. Preliminary Results Preliminary tabulations of our survey results reveal an interesting pattern of variation both across dependent variables and across countries. The emphasis placed by the recent Latin American literature on grassroots organizations as the favored form of popular political expression is not borne out by these results. As Table 2 shows, reported membership in any kind of community organization represents a tiny minority, hovering around 10 percent in most countries and in the full sample. On the other hand, participation in established political parties reaches almost half (47%) of the combined sample. This figure is inflated by results for Kingston, Jamaica where almost all respondents are members of one of the two main parties. However, in the other two democracies-- Costa Rica and the Dominican Republic--participation in party politics far exceeds that in community-based associations. There are notable differences between these countries and the remaining two. In Guatemala and Haiti, both long governed by military dictatorships or military- controlled civilian regimes, party membership among popular groups is minimal. This absence of participation is not compensated by grassroots organizations since participation in them is also very low. This general pattern of popular demobilization and apathy is also apparent in the attitudinal questions, especially in the Guatemalan sample. ------------------------------------------------------ Table 2 about here ------------------------------------------------------ There is a significant reversal between political behavior and attitudes toward parties and grassroots organizations. As shown in the next rows of table 2, majorities in all countries support community organizations as "true representatives of the popular will" and believe that more power should be vested in them. In four of the five countries and in the total sample, these proportions are significantly higher than those supporting party politics. The single exception is Haiti, but in this case the difference between the two figures is not statistically significant. The greater sympathy for community organizations among our respondents is still more evident in the last rows of Table 2. They present the distribution of responses to a forced-choice item about how hypothetical inhabitants of a popular neighborhood spend their free time. With the exception of Guatemala, majorities in all countries endorsed the option of joining a community organization to "improve the neighborhood." In the full sample, two-thirds supported this alternative, while less than 10 percent opted for joining a party to "resolve the problems of this city." The Haitian and Guatemalan sample had the highest proportion of respondents opting for the do-nothing option ("Remain at home"). The apathy toward popular participation in these countries is again evident in the distribution of responses to another item inquiring how much community organizations can accomplish for their members. About 70 percent of Guatemalans and Haitians did not know or believed that community organizations could accomplish nothing; the corresponding figure among respondents in the three democratic countries was less than 50 percent. These preliminary results show, first, a remarkable gap between theoretical expectations about the extent of popular involvement with the "new social actors" and actual participation in this type of organizations. Second, they indicate an equally notable gap between verbal endorsements of these grassroots movements and involvement in them. Third, they demonstrate major disparities across countries in a pattern supportive of past theorizing on the effects of national political systems on popular mobilization (Leeds 1969; Portes and Walton 1976). However, this last conclusion must be tested against the real possibility that differences among national samples are a spurious result due to their different age, sex, educational, and occupational profiles. The following section examines this alternative. IV. Determinants of Participation Despite the observed gap between attitudes and behavior, the two are not uncorrelated. Table 3 presents coefficients of association between party and community group participation and indices of attitudes toward the two forms of activism in the entire sample and in each individual country. Results show positive and significant associations between each attitudinal index and the respective behavior and, with the partial exception of Haiti, higher correlations of attitudes and behavior within each type of political participation than between them. We interpret these results as supportive of the internal consistency and validity of the data rather than as indicative of causal direction. In a cross-sectional survey such as this, it is not possible to infer that statements of opinion "cause" behavior rather than viceversa. Instead, the following analysis treats political attitudes and reported participation as separate dependent variables and cross-tabulates them against the individual-level predictors available for the sample. ------------------------------------------------------ Table 3 about here ------------------------------------------------------ Table 4 reveals wide variations by individual characteristics, especially in the attitudinal dimensions. Endorsement of community organizations is stronger among the better-educated and better-paid. It is also higher among workers (both formal and informal) than among the unemployed and self-employed and higher among individuals who identify themselves as members of the working-class. Interestingly, participation in political parties is also supported by the more educated and affluent although, in this instance, there are no statistically significant differences by type of occupation or class identification. Instead, ascriptive characteristics come into play with men being significantly more likely to endorse political party activism than women and the young more than the old. The data also reveal much less variation in actual participation in community groups which turns out to be quasi-constant across the entire sample. Neither gender nor age, occupation, income or class identification lead to differential rates of community grassroots participation. There is only a slight positive relationship with education, but it is statistically non- significant. Political party membership is strongly affected, on the other hand, by several individual characteristics. Better educated persons are much more likely to take part in this type of activity; this statistical association is the strongest found in the table. Workers (formal and informal) and individuals who identify themselves as workers are significantly more likely to engage in party politics than the self- employed, the unemployed, and those who define themselves as part of the middle-class. There is also a marginally significant tendency for younger respondents to be more active on party politics. These bivariate results are unsurprising from the standpoint of classical theories of political participation. Education has been repeatedly found to be a significant determinant of political activism (Almond and Verba 1963; Banfield and Wilson 1970). So has occupation, insofar as wage workers who repeatedly interact with each other are more likely to develop class consciousness and participate on that basis than others, such as the self-employed and unemployed, who are cut off from the wage bond (Lipset 1963; Nielsen 1986; Wright 1985). This pattern of empirical associations is much less supportive of the literature on the "new social actors" in Latin American since it reveals no sharp "disjuncture" in the individual correlates of this type of participation as contrasted with more traditional party activism: In both cases, significant effects tend to be associated with the same individual predictors--mainly education, occupation and class identification--and the direction of results is generally the same. It remains to be seen, however, whether these individual differences can account for the observed variation across countries in the four dependent variables or whether, on the contrary, the contrast among national political system overwhelms the effect of these individual characteristics. If the latter, we would have persuasive evidence against theories of popular mobilization that confine their focus to individual-level determinants. ------------------------------------------------------ Table 4 about here ------------------------------------------------------ The relevant data are presented in Table 5. They consist of a series of logistic regressions of the four dependent variables on all predictors including both individual characteristics and nationality, the latter variable serving as an indicator of differences in political systems. Regression coefficients indicate the net increase (or decrease) in the logarithm of the odds of endorsing or participating in either community or party politics associated with a unit increase in each predictor, controlling for others. To provide a clearer sense of the meaning of these results, the columns labelled ~p present the net increase (decrease) in the probability of endorsement or participation for each significant effect. Significant predictors are defined as those exceeding at least twice their standard errors. The independent variables in these equations are somewhat different than those presented in Table 4. Education and income are entered as continuous variables with education measured in years completed and income as the logarithm of monthly income in 1992 dollar equivalents. Occupation is also entered as a continuous variable measured in Treiman's international occupational prestige scores (Treiman 1977). Type of occupation is also included as a categorical variable. Coefficients associated with this and all other categorical variables have a somewhat different interpretation than the others. They represent effects relative either to the total effect of the variable or to its omitted category. In the case of occupational type, the omitted category is "employers"; for class identification, it is "middle class"; and for country it is Guatemala. In the first two cases, individual effects are estimated relative to the total variable effect; for country, we chose to make effects relative to those for Guatemala. ------------------------------------------------------ Table 5 about here ------------------------------------------------------ Table 5 reveals a clear pattern of results in which inter-country differences are vastly more powerful as predictors of attitudes and behaviors than individual- level variables. Once national differences are introduced in the equation, no other predictor retains a significant effect either in grassroots organization or party membership. Attitudes are influenced by some individual characteristics, but there is no compelling pattern of effects: married individuals and those who identify themselves as members of the working-class are significantly more likely to endorse community organizations; male, better-educated, and higher-income respondents are more likely to do so for political parties. The combined effect of these three individual variables is quite strong. For example a man with a high school (12 year) education, and a monthly income of U.S$400 is 47 percent more likely to favor party membership than a woman with grammar school education (6 years) and a U.S.$150 a month income. Nevertheless the strongest effects in these equations are those associated with nationality and they follow an unmistakable pattern. Relative to Guatemalans, all other members of the sample possess significantly higher levels of community grassroots participation, with the exception of Jamaicans, and all are significantly more likely to support such organizations, with the exception of Haitians. Haitians share with Guatemalans a level of party membership significantly inferior to the other two national groups for which data are available; but Guatemalans are in a class apart in their negative attitudes toward the party system. Compared to them, Dominicans and Costa Ricans are 32 percent more likely to endorse party membership, Haitians 27 percent, and Jamaicans 21 percent. Jamaicans are at the low end of attitudinal differences, but their probability of actual party membership exceed that of Guatemalans by an extraordinary 67 percent. These results clearly indicate that inter-country differences observed in the earlier bivariate tabulations are not explainable by differences in the socio- demographic profiles of the national samples. The opposite is the case, as most individual-level characteristics cease to have a significant association with political attitudes and behavior once nationality is taken into account. Hence, the obvious conclusion is that systemic factors associated with the polity play the decisive role in influencing patterns of popular participation. Within-country differences in political activism are still explainable through individual-level variables, but the "core" level of beliefs and behaviors of popular sectors is largely set by the characters of their historical interaction with the national political system. A final word must be said about the directions in which the political system affect political attitudes and participation. Table 2 indicated that Dominicans, Costa Ricans, and Jamaicans shared a similar pattern of higher participation in political parties than in community organizations, but higher support for participation in community organizations than in political parties. This can be related to the fact that in democratic systems people have access to public and private goods through the party system (and hence they take part in it) but they do not value participation per se. The most notorious example is the Jamaican case, where we found an extremely high percent of membership in political parties (98%) and an extreme low endorsement of this form of political participation (4.4%). This may be explained by the fundamental importance of clientelist relations in the Jamaican political system. Guatemala, an example of a non-democratic political system shows a very low degree of participation and support for political parties, which can be explained by the fact that, in an authoritarian political system, people do not find incentives to either participate or endorse political parties, leaving the field to the few politically committed. Conclusion In 1984, Luis Razeto, one of the most prominent students of Chilean popular grassroots organizations had this to say about their future: During a long time, these groups have served efficaciously to maintain and develop popular organization under difficult living conditions. Today, it is necessary that they do more than this. We know that these organizations were not created to protest or to reivindicate rights in front of the state, but many have known how to participate in protest rallies, in social demand-making, and in popular reivindications. Now that the people begins to express itself anew, these organizations, as integral part of the people, also take part in the struggle of all for a new society (Razeto 1984: 18-19). In a more cautious tone, John Friedmann and Mauricio Salguero also concluded their analysis of the Latin American "barrio economy" in this note: If civil society is to be mobilized for political ends, the factory cannot remain the primary locus. As the home terrain for the self-production of life, the barrio economy is the more inclusive alternative. It brings in the long-term unemployed, the young first-time job seekers, the workers active in the informal sector, older folks, and even small children, in short the entire working class sector of civil society... To be sure, the possibility of a counter-hegemonic mobilization in the popular barrios of Latin American cities is only a latent possibility at this time, though there is a growing evidence for it from a wide spread of cities, from S~o Paulo to Santiago to Lima to Medellin (Friedmann and Salguero 1988: 38; emphasis in original). Whatever this "growing evidence" was in the mid- 1980s, it was scarcely apparent in the capital cities of the Caribbean Basin in the early 1990s. Compared with the eloquent rhetoric about the counter-hegemonic and innovative potential of grassroots organizations, our figures on the actual extent of popular participation in these organizations look rather dismal. To be sure, there is widespread sympathy for their goals, but when the moment comes to translate sentiments into action, it is the traditional political parties rather than the "new" movements which have the upper hand. Still more important is the finding that individual- level influences on both sympathy for and participation in popular organizations do not form a distinct causal sequence. Instead, the same general set of characteristics that incite support for political party membership do so for grassroots organizations, suggesting the existence of a more general participatory syndrome. The more educated rather than the illiterate, the better off rather than the destitute, those who see themselves as members of the working-class rather than the poor lean more strongly toward both forms of participation. This pattern of relationships is, of course, at variance with the expectations of much of the earlier grassroots movement literature. The key finding of our analysis is, however, that when country differences in political systems are introduced, they do away with most of the observed individual influences on participation. This pattern of results indicates that what ultimately matters is the size, durability, and character of the political "space" opened by the dominant system for popular participation. Low-income groups are quick to seize such spaces when they emerge, but they lapse into indifference and apathy when these opportunities close due to systematic repression. In this sense, the best characterization of political action by low-income urban groups is rational adaptation to the existing structures of power rather than radical confrontation with them or the creation of an alternative civic order. Twenty-three years ago, a study of politics in the favelas of Rio de Janeiro concluded on the following note: Our evidence shows beyond the shadow of a doubt that squatment populations are shrewdly aware of [political] changes and make appropriate responses to them, shifting their form of behavior as is convenient, attempting to make the best use of the changing structure in their bad situation. Thus, there has always been a keen interplay between residents and politicians or candidates looking for votes in return for which gifts are given (Leeds 1969: 79) The findings of this comparative study agree in essence with that conclusion. They extend it by showing the key disjunction between endorsement of grassroots organizations, on the one hand, and membership in traditional parties, on the other, in the three democratic countries where both forms of participation are possible. Sympathy for grassroots organizations there is aplenty, but rational adaptation to reality dictates that actual investments of time and effort should go mostly toward those organizations capable of delivering individual favors and improvements in precarious living conditions. References Almond, Gabriel and Sidney Verba. 1963. The Civic Culture: Political Attitudes and Democracy in Five Nations. Princeton: Princeton University Press. Banfield, Edward and James Q. Wilson. 1970. The Unheavenly City: The Nature and Future of our Urban Crisis. Boston: Little, Brown. Campero, Guillermo. 1987. Entre la sobrevivencia y la acci"n pol!tica. Santiago de Chile: ILET. Cardoso, Ruth C.L. 1983. "Movimentos sociais no Brasil p"s-64. Balano cr!tico." Pp. 215-239 in Sociedade e politica no Brasil p"s-64, edited by B. Sorj and M.H. Tavares de Almeida. S~o Paulo: Brasiliense. Cardoso, Ruth C.L. 1992. "Popular Movements in the Context of the Consolidation of Democracy in Brazil." Pp. 291-302 in The Making of Social Movements in Latin America, edited by A. Escobar and S.E. Alvarez. Boulder: Westview Press. Castells, Manuel. 1983. The City and the Grassroots. London: Edward Arnold. CEPAL. 1990. Transformaci"n productiva con equidad. Santiago de Chile: CEPAL. Dore-Cabral, Carlos. 1985. "La distribuci"n espacial de los movimientos sociales de abril del 84." Impacto socialista 1(April-May):16-26. Eckstein, Susan. 1977. The Poverty of Revolution, the State and the Urban Poor in Mexico. Princeton: Princeton University Press. ________, ed. 1989. "Power and Popular Protest in Latin America." Pp. 1-60 in Power and Popular Protest: Latin American Social Movements. Berkeley: University of California Press. Fals Borda, Orlando. 1992. "Social Movements and Political Power in Latin America." Pp. 303-316 in The Making of Social Movements in Latin America, edited by A. Escobar and S.E. Alvarez. Boulder: Westview Press. Friedmann, John. 1989. "The Latin American Barrio Movement as a Social Movement: Contribution to a Debate." International Journal of Urban and Regional Research 13:501-510. Friedmann, John and Mauricio Salguero. 1988. "The Barrio Economy and Collective Self-Empowerment in Latin America." Pp. 3-37 in Power, Community and the City, edited by P.D. Smith. New Brunswick, N.J.: Transaction Books. Gordon, Derek and Cheryl Dixon. 1991. "La urbanizaci"n en Kingston, Jamaica." Pp. 99-212 in Urbanizaci"n en el Caribe, edited by A. Portes and M. Lungo. San Jos: FLACSO. Hardy, Clarisa. 1987. Organizarse para vivir. Pobreza urbana y organizaci"n popular. Santiago de Chile: Programa de Econom!a del Trabajo. Hechter, Michael. 1987. Principles of Group Solidarity. Berkeley: University of California Press. Iglesias, Enrique. 1985. "The Latin American Economy during 1984: A Preliminary Overview." CEPAL Review 35(April):7-44. Jelin, Elizabeth. 1985. Los nuevos movimientos sociales. 2 vols. Buenos Aires: Centro Editor de Amrica Latina. Jonas, Susanne. 1991. The Battle for Guatemala: Rebels, Death Squads, and U.S. Power. Boulder: Westview Press. Lagos, Ricardo and Victor Tokman. 1983. "Monetarismo global, empleo y estratificaci"n social." El Trimestre Economico 50(July-Sept.):1437-1474. Leeds, Anthony. 1969. "The Significant Variables Determining the Character of Squatter Settlements." America Latina 12(July-Sept.):44-86. Lehman, David. 1990. Democracy and Development in Latin America. Philadelphia: Temple University Press. Lipset, Seymour Martin. 1963. Political Man. Garden City, N.Y.: Doubleday. Lozano, Wilfredo and Isis Duarte. 1991. "Proceso de urbanizaci"n, modelos de desarrollo y clases sociales en Rep#blica Dominicana, 1960-1990." Pp. 213-347 in Urbanizaci"n en el Caribe, edited by A. Portes and M. Lungo. San Jos: FLACSO. Lungo Ucls, Mario, Marian Perez, and Nancy Piedra. 1991. "La urbanizaci"n en Costa Rica en los '80." Pp. 37-188 in Urbanizaci"n en Centroamerica, edited by A. Portes and M. Lungo. San Jos: FLACSO. Mangin, William. 1967. "Latin American Squatter Settlements: A Problem and a Solution." Latin American Research Review 2 (Summer):65-98. Manigat, Sabine. 1992. "L'urbanisation de Port-au-Prince dans les annees 1980: economie et conditions de vie des Port-au-Princiens." Paper presented at the conference on Caribbean Urbanization in the Years of the Crisis, sponsored by the Latin American Faculty of Social Sciences (FLACSO) and The Johns Hopkins University. Santo Domingo, July 29-August 2. Matos Mar, Jos. 1985. Desborde popular y crisis del estado. Lima: Instituto de Estudios Peruanos. Nielsen, Franois. 1986. "Structural Conduciveness and Ethnic Mobilization: The Flemish Movement in Belgium." Pp. 173-198 in Competitive Ethnic Relations, edited by S. Olzak and J. Nagel. Orlando, Fla.: Academic Press. Prez-S inz, Juan Pablo. 1991. "Ciudad de Guatemala en la decada de los ochenta: crisis y urbanizaci"n." Pp. 189-289 in Urbanizaci"n en Centroamerica, edited by A. Portes and M. Lungo. San Jos: FLACSO. Portes, Alejandro and Michael Johns. 1989. "Class Structure and Spatial Polarization: An Assessment of Recent Urban Trends in Latin America." Pp. 111- 137 in Lost Promises: Debt, Austerity, and Development in Latin America, edited by W.L. Canak. Boulder, Colo.: Westview Press. Portes, Alejandro and John Walton. 1976. Urban Latin America, the Political Condition from Above and Below. Austin: University of Texas Press. PREALC. 1990. Empleo y equidad: el desafio de los 90. Santiago de Chile: PREALC. Przeworski, Adam. 1985. Capitalism and Social Democracy. Cambridge: Cambridge University Press. Razeto, Luis. 1985. Las organizaciones populares. Mas all de la subsistencia. Santiago de Chile: Academia de Humanismo Cristiano. Schkolnik, Mariana. 1986. Sobrevivir en la poblaci"n Jose M. Caro y en lo Hermida. Santiago de Chile: Programa de Econom!a del Trabajo. Slater, David, ed. 1985. "Social Movements and a Recasting of the Political." Pp. 1-25 in New Social Movements and the State in Latin America. Amsterdam: CEDLA. Tironi, Eugenio. 1986. "El fantasma de los pobladores." Estudios Sociologicos 4(September-December):391- 397. Treiman, Donald J. 1977. Occupational Prestige in Comparative Perspective. New York: Academic Press. Trouillot, Michel-Rolph. 1990. Haiti, State Against Nation. New York: Monthly Review Press. Vargas, Jorge. 1992. "Los movimientos urbanos en Costa Rica durante los ochenta." Pp. 161-179 in El nuevo rostro de Costa Rica, edited by J. M. Villasuso. Heredia: Centro de Estudios Democr ticos de Amrica Latina. Wright, Erik O. 1985. Classes. London: Verso. Appendix English-version of Items Measuring Attitudes and Participation in Community Grassroots Organizations and Political Parties 1. Belongs to a Political Party: Do you belong to any political party today? 1. Yes _____ Which? _________________________ 2. No _____ 2. Belongs to a Community Organization: To what community organizations do you belong? (This question is part of a battery of questions about community organizations). 1. None _____ 2. One or More _____ Number: _____ Names: _________________________ _________________________ _________________________ 3. Supports Participation in Party Politics: Please, tell me with whom do you agree: James said: It is better to stay out of politics. After all, common people have very little say in how the country is run. George said: It is important to take part in politics. If common people get together they can have a say in how the country is run. Who is right? 1. James _____ 2. George _____ 3. None _____ Why? _________________________ 4. Don't know _____ 4. Supports Community Grassroots Organizations: Please, tell me with whom do you agree: Helen said: The city government should give more participation to neighborhood organizations because those are the true voices of the people. Laura said: The city government should not give more participation to neighborhood organizations because they represent only the interest of small groups. Who is right? 1. Helen _____ 2. Laura _____ 3. None _____ Why? _________________________ 4. Don't know _____ 5. Effectiveness of Community Organizations: How much would you say that community organizations can help improve the neighborhood? (Interviewer: If the respondent says some can help and others can't, try to obtain a general answer regarding these organizations). 1. A great deal _____ 2. Some _____ 3. A little _____ 4. Not at all _____ 6. Best Use of Free Time: I am going to read you a story and would like you to tell me with whom do you agree. John, Paul and George all live in the same neighborhood. Each uses his free time differently. John joined a political party. He says that only by participating in politics is it possible to solve the problems of the city; Paul joined a neighborhood committee that works to improve the conditions where he lives; George did not join any organization and spend his free time at home or with his friends. Who is making better use of his time? 1. John _____ 2. Paul _____ 3. George _____ 4. None of them _____ Why? _________________________ 5. Don't know _____ 6. No answer _____