3
Labor Politics and Regime Change
Internal Trajectories versus External Influences

Ruth Berins Collier

Social analysts have long been fascinated with the idea of critical junctures or historic watersheds—particular "moments" or transitions of fundamental political reorientation. It is this fascination that has inspired the present volume, the guiding hypothesis of which is that the 1940s represents precisely such a moment of reorientation. In this chapter I examine the politics of labor and regime change in Latin America during the 1940s from two analytical perspectives, each related to a different critical juncture. The first perspective emphasizes the influence of international events: World War II and the international realignments that accompanied the advent of the Cold War.^[1] The second perspective views the development of Latin America in the 1940s as part of a longer evolution within each country: the unfolding of an internal trajectory of change, the parameters of which derive from the initial incorporation of labor.^[2] This incorporation was part of a major and multifaceted socioeconomic transformation, which brought with it a set of political changes: the shift from a laissez-faire to an activist state; the appearance of a formal industrial sector regulated by the state; the change in forms of social control from clientelism to corporatism; the advent of mass society and mass politics; and the emergence of capital-labor relations as a major social cleavage accompanied by new forms of social protest. The analysis of these changes focuses on three phases: the incorporation period itself, the initial reaction in the "aftermath" period, and the longer-term legacy that is labeled the "heritage" of incorporation. These two perspectives represent quite different approaches to the politics of labor and regime change in Latin

America in the 1940s. Instead of arguing the merits of one perspective at the expense of the other, I shall attempt to combine them through a comparative analysis of four Latin American countries: Brazil, Chile, Mexico, and Venezuela.

Figure 1 schematically represents the intersection of the two perspectives in the four countries. The skeleton of the internal trajectories argument is laid out horizontally. The period of the initial incorporation of labor occurred differently in the different countries, setting in motion distinct trajectories of change that unfolded during the subsequent aftermath and the heritage periods.^[3] The alternative perspective that emphasizes the role of international forces is shown by the long box in the middle of the figure.

As figure 1 shows, the two perspectives differ in their treatment of time. First, the international forces perspective is synchronic and cross-sectional; the internal paths approach is diachronic and longitudinal: while the former examines a slice of time, the latter employs a historical, over-time approach. Second, while the international perspective uses real time (i.e., the 1940s), the internal paths perspective is based on analytical time, so that the analytically defined periods of incorporation, aftermath, and heritage vary in historical timing as well as duration.

A word might be added about the types of causes that are the focus of each of the perspectives. Although generated within each country, the various elements of the internal forces are more generic and structural than unique or idiographic, since they constitute a common social transformation based on the institutionalization of labor-capital relations and the integration of the working class in the political system. The transformations are therefore parallel or similar in the various countries.

An international perspective may focus on a variety of external causes. One is the diffusion or contagion of models or ideas from abroad: for example, the international appeal of fascism, communism, or democracy. A second is common or repeated pressures from abroad: for example, American or Soviet foreign policy toward Latin American countries. A third issue refers to the impact in Latin America of fundamental shifts in the international order that had an impact on the world at large: a bipolar international system, world hegemonic leadership of a hemispheric power, the salience of the Cold War with its competing ideologies, and the dominance of a world capitalist order characterized by a Keynesian compromise between labor and capital.

Figure 1.  Intersection of Two Analytic Perspectives: Internal Trajectories (incorporation, aftermath, and heritage periods) and External
Events

We have, then, two quite different perspectives. They are not mutually exclusive; each may be rounded out by reference to the other. For example, it is important whether or not labor incorporation occurred before or after the rise of fascism or the adoption of a popular front policy by the Comintern, since these were the types of external conditions that affected internal development. But equally important, the impact of these external influences was affected by a nation's position along its internal trajectory. Thus the two perspectives are not rival but complementary approaches that can be usefully combined. Before assessing the nature of this complementarity, however, I begin by treating each view separately.

The International Conjuncture of the 1940s

Though the external forces unleashed during the 1940s may have had wide-ranging political and economic impacts, I focus here on their impact on two related dimensions of domestic change. The first concerns coalitions and patterns of political and class collaboration or conflict. Specifically, were labor unions and leftist parties collaborating with other political and economic actors, or were they engaging in strikes and other forms of working-class protest and confrontation? The second dimension concerns patterns of democratization, political opening and closing, and reformist initiatives. That is, to what extent does regime change in Latin America during the 1940s reflect or match international events or trends? Those external events most likely to influence these features of Latin American politics are presented in table 11.

A first consideration is the possible links between these international events and the patterns of political and class collaboration in Latin America. The popular front policy of the Comintern was clearly important. Originally adopted in the mid-1930s, this strategy was abandoned temporarily following the German-Soviet pact of August 1939 and then renewed in a more radical form after the German invasion of the Soviet Union in June 1941. The subsequent popular front policy advocated the formation of broad anti-Fascist alliances, including cooperation with bourgeois parties, anti-Fascist governments, and non-Communist groups within the labor movement. It meant class collaboration, the moderation of labor demands, and in its most extreme forms, the dissolution of Communist parties and the abandonment of strikes for the duration of the war. In 1943, consistent with the priority

given to a broader anti-Fascist alliance, the Comintern was disbanded, signifying the dissolution of the international revolutionary alliance. Because Communist parties were influential in labor movements throughout Latin America, we would expect to see reverberations of Comintern policy throughout the region.

The spirit of multiclass solidarity and the support for a new coalition against fascism were not limited to the Communists. The same sentiments flourished in the United States and through this route too reverberated in Latin America. The same year that the German invasion brought the Soviet Union into the war, the Japanese attack on Pearl Harbor provoked the United States to join the Allies. U.S. sentiment toward collaboration with the Communists changed accordingly. This flip-flop was reflected in Time magazine's two descriptions of Vicente Lombardo Toledano, the Marxist leader of Mexican labor. The first, published in 1940 before the United States entered the war, referred to the "large-eared, hot-eyed, Communistic little Vicente Lombardo Toledano" and ridiculed his "screaming" and "unblushing" shift of policy following the change in Comintern line after the German-Soviet nonaggression treaty.^[4] In 1942, just a year and a half later, however, Time described Lombardo as "the brilliant, aggressive and fluid leader of Mexican labor, ... a slight, gentle little man with big ears and dreamy eyes, ... a

Puritan in his personal life, abstemious, logical in argument, part-Indian, part-Italian, philosopher, archaeologist, scientist, scholar."^[5]

We might expect this change in the attitude of the United States to be reflected in pressure on non-Communist factions within the labor movement to pursue a more moderate course, as well as on various other groups to recognize that the struggle against fascism was of paramount importance. International influences created a context in which many Latin American governments and bourgeois parties might be receptive to the line of collaboration proposed by Communist parties. These changes in international influence might thus lead us to expect a brief period of multiclass collaboration starting in 1942.

By 1945 the conclusion of the war brought an end to the rationale for political and class collaboration. External alliances were quickly reordered, and those changes had implications for internal alliances as well. At this point we would expect the reemergence of a postponed reformist or radical political agenda, ideological polarization, and a renewal of class conflict and labor protest.

The second issue concerning the impact of international forces relates to regime change. Here the expectation would be that the victory of the Allies and the triumph of democracy over fascism led to the diffusion of democratic and reformist values throughout the world, a process reinforced by the United Nations Charter. Moreover, from 1943 the United States began to press for political liberalization and democratization in Latin America. By 1945 and the beginning of Cold War hostilities over the issue of democracy in Eastern Europe, "it became even more imperative that the allies of the United States in Latin America were seen to be democratic."^[6] If international influences were important, we would expect at this point a pattern of democratization and political opening in Latin America. We might also expect a period of new reformist initiatives, both because popular demands could be expressed more openly and because the models of social democracy, the welfare state, and a Keynesian class compromise were taking shape in the United States and Europe.

International pressures for reformist democracy in Latin America were short-lived. New international forces unleashed by the Cold War tended to support a political closing, as democracy-versus-fascism became superseded by the new cleavage of capitalism-versus-communism. If during the war the Soviet Communists subordinated class struggle abroad for the anti-Fascist

fight, after the war the Western democracies, particularly the United States, subordinated the prodemocracy struggle to the struggle against communism. Though it is hard to date the onset of this period, any ambiguities were resolved by 1947. In that year the Soviet Union established the Cominform to replace the dissolved Comintern and reinstate Moscow discipline, and the United States proclaimed the Truman Doctrine to contain communism and aid other governments in the anti-Communist struggle. Nineteen forty-seven was also the year of the Rio Treaty, a pact of hemispheric solidarity and mutual assistance that brought the Cold War to the Western hemisphere. Finally, it was in 1947 that the battle against communism penetrated the labor arena. In the United States the Taft-Hartley Act barred Communists from union leadership; in Latin America the anti-Communist labor confederation, Organización Regional Interamericana del Trabajo, was formed under the auspices of the American Federation of Labor.

The impact of international influences on the two dimensions of multiclass collaboration and democratization might thus be expected to produce four partially overlapping periods. These periods, which are presented in the upper half of figure 2, are summarized as follows:

1941–1945: After the greater level of class conflict and strikes during the period of the German-Soviet pact, renewal of popular front strategies, class cooperation, decline in strikes, and labor peace.

1944–1946: Democratization, political opening, and reformist initiatives.

1945–1946 (or later): Ideological polarization, greater class conflict, labor protest, and political opposition.

From 1946–1947 on: Political closing, restored labor discipline, and a retreat from reformist politics.

To what extent did these hypothesized phases actually occur in the four Latin American countries under examination?

1. Multiclass Collaboration and Political Cooperation by Left and Labor (1941–1945) . The popular front policy of the Comintern had an important effect throughout Latin America, and all four countries underwent this phase. In each of these countries the Communist party was influential among organized labor, pursued a policy of cooperation with governments, and encouraged the moderation of industrial conflict. For their part centrist groups were willing partners to collaboration.

The influence of the popular front policy was strongest in Chile. In the mid-1930s the Communist, Socialist, and Radical parties established a Popu-

Figure 2.  Timing of Expected Outcomes of International Events and the
Incorporation and Aftermath Periods

lar Front that unified large segments of the working and middle classes. A Popular Front government under Radical leadership was elected in 1938, and it held together despite the German-Soviet pact. Elections were held again in 1942, after the adoption of the Comintern's more extreme version of popular frontism. Although the Popular Front as a formal coalition was not renewed, the Communists and Socialists continued to collaborate with the Radical party and supported its more conservative presidential candidate, Juan Antonio Rios Morales, and subsequently his government.

In Brazil popular front collaboration occurred in the mid-1940s. At first, in 1941–1942 under the still heavily authoritarian government of Getúlio Vargas, the popular front policy had little impact. At this point there were no elections, labor was controlled through a corporatist system, and the Communist party was banned. Conditions began to change, however, in 1943 as the expectation grew that after the war the authoritarian regime would yield to a democratic regime. By 1945 a coalition of the Vargas forces, the working class, and the Communist party began to form. The pro-Vargas forces now took the initiative by founding the Partido Trabalhista Brasileira (PTB), aiming to mobilize labor support. Vargas and the Communists then struck a deal, which

led to a political amnesty for imprisoned Communists and the legalization of the Brazilian Communist party. Both the PTB and the Communist party participated in the queremista movement that urged Vargas to become a presidential candidate, and in 1945 they organized the Movimento Unificador dos Trabalhadores as a political vehicle to enlist working-class support.

Mexico was another example of popular front collaboration. In the second half of the 1930s the Mexican Communist party (PCM) shifted from an anticollaborationist stance to support for the Cárdenas government. This posture continued for the duration of the war despite the election of the more conservative government of Manuel Ávila Camacho in 1940. Under Cárdenas the Communists promoted the reorganization of the governing party to resemble a popular front coalition. In the 1940 election the PCM withheld its support from a reformist successor to Cárdenas in favor of Ávila Camacho, for whom it pledged to act as a "shock brigade."^[7] Following the German invasion of the Soviet Union, the Confederación de Trabajadores de México, the major national labor confederation, entered into pacts with business to promote collaboration and in 1942 renounced the use of the strike for the duration of the war.

Venezuela deviated only partially from the general pattern. During the popular front period of the 1930s the Communists collaborated with the reformist opposition in an antigovernment front, which collapsed at the time of the German-Soviet pact. With the renewal of the popular front policy in 1941 the Venezuelan Communist party began to collaborate with the government in a "marriage of convenience"^[8] that emphasized class harmony and political stability. The following year the party supported congressional candidates loyal to President Isaias Medina Angarita. Until 1944 the Communist party remained the dominant influence within the working class and brought the bulk of the urban labor movement into the coalition. For his part Medina established a more open atmosphere, allowing political freedom, party activities, and the formation of new unions. During this period, however, the reformist opposition under Acción Democrática rejected collaboration and remained in opposition. Acción Democrática had considerable popular support: it was the dominant influence among the peasantry, and after 1944 it became the dominant force in the working class as well. Because of the continued opposition to the government by Acción Democrática and its affiliated popular groups, Venezuela constitutes a partial exception to the hypothesized pattern of broad anti-Fascist collaboration.

2. Political Opening and Reformist Initiatives (1944–1946 ). Evidence of this trend is more mixed. During this brief period new democratic regimes were created in both Brazil and Venezuela, but Mexico and Chile did not experience reform or opening. In Brazil a combination of international diffusion and direct pressure from the United States contributed to the downfall of the Vargas government in 1945 and the inauguration of a new, electoral regime. Besides adopting a more open political system, Brazil enacted a number of labor reforms beginning in 1943 when Vargas himself began attempting to construct a new constituency among the urban working class. These reforms, which continued until around 1946, included a greater toleration of strikes, wage increases, the cancellation of a loyalty test for labor union leaders, the introduction of union elections, the legalization of the Brazilian Communist party, and the political amnesty of 1945. Similarly, despite a seemingly improbable beginning in which Acción Democrática achieved power through a coup, in 1945 Venezuela witnessed the advent of a more democratic government. Although a reformist opening had already begun under Medina, the year marked a deepening of the commitment to reform. A constitutional convention wrote a new democratic charter, which was quickly implemented in subsequent elections. Reformist changes were also made in labor policy, including a more favorable labor law, some labor participation in government, rising wages, and the spread of collective bargaining.

These trends were absent in Chile and Mexico. Even though democratic rule had already prevailed in Chile for more than a decade, the mid-1940s were an antilabor period. The interim Duhalde government of 1945–1946 was hostile to labor and repressed strikes and demonstrations led by the Communists. In Mexico the antireformist trend was even more marked, as this period brought greater control of labor and the tightening of authoritarian one-party dominance. Real wages fell steeply, especially in the unionized industrial sector, even as the economy grew. Unionization stagnated, and a change in labor law restricted the right to strike and made it easier to dismiss workers. The government used a newly enacted "crime of social dissolution" to persecute dissident union leaders, while changes in the structure of the ruling party reduced the relative weight and influence of labor within it. These steps culminated in January 1946 with the reorganization of the ruling party into the Partido Revolucionario Institucional (PRI), a measure that further subordinated labor and the other organized "sectors" of the party and concentrated power in central party organs. Finally, again in early 1946,

changes in the electoral law made it more difficult for dissident groups to register as political parties and therefore helped to insulate the PRI from challenges by the opposition.

3. The New Combative Posture of Labor and the Left (immediate postwar years ). All four countries underwent heightened class conflict and growing political opposition during the immediate postwar period coinciding with the end of the popular front policy and the onset of the Cold War. This pattern was particularly striking in Brazil and Mexico. In Brazil the opening years of the Dutra government saw an intensified process of the repoliticization and reactivation of the working class begun during the last years of the Vargas government. The trend became visible in the party system and in the sphere of industrial relations. In the newly opened political arena, parties with a base in the working classes achieved unprecedented electoral success. In the presidential and congressional elections of December 1945 the Communists won 10 percent of the national vote and achieved pluralities in major industrial cities and a number of state capitals. The Partido Trabalhista Brasileira won another 10 percent, giving the two parties based in the union movement approximately one-fifth of the vote. This electoral mobilization was accompanied by a more militant political posture on the part of both the Communists and the PTB.

As John French has argued, during this period the working class attained a new level of consciousness that marked a radical break with the past. Even in this early period, shortly after its founding, the PTB could not be seen merely as an instrument created from above by Vargas. Instead, as competition from the Communist party put the PTB under pressure to be responsive to the rank and file, the PTB came to represent an independent working-class voice.^[9] In addition to the electoral sphere the new militancy of the Brazilian working class was also evident in the labor unions, which became more democratic and activist. Major changes in the leadership of the unions occurred as Communist party and PTB militants began to take control. In 1945 an attempt by the government to curb these trends by instituting plural unionism failed in the face of working-class opposition. In 1946 Communist leaders organized a major wave of strikes affecting many rural as well as urban areas.

Mexico experienced similar heightened political and class conflict during this period. When the war ended, the Confederación de Trabajadores (CTM) split on the issue of continued collaboration with the government. Dissidents

under Communist influence now portrayed collaboration as a wartime expedient and declared their determination to return to a more militant and aggressive posture. Accordingly, a number of important national unions left the CTM and formed the rival Central Única de Trabajadores. Somewhat later Lombardo was expelled from the CTM and formed his own movement, the Alianza de Obreros y Campesinos de México—later the Unión General de Obreros y Campesinos de México (UGOCM). The two dissident confederations supported a more combative and independent stance, and at their height they represented around 40 percent of the organized labor movement. This split became reflected in politics, as Lombardo founded the Partido Popular Socialista (PPS) as an opposition party based on working-class support. In the long run the PPS did not fare particularly well, but these developments illustrate the greater intensity of labor militancy during the immediate postwar period.

Chile and Venezuela present a somewhat different picture in that political collaboration continued even with the expected rise in class conflict. In Chile the Communist party became more aggressive, and under its leadership labor conflict, including rural strikes, rose substantially. The growth of the Communist left was also reflected in the party's increased share of the vote in the 1947 municipal elections. Yet at the very time they were becoming more active and combative the Communists supported the Radical party candidate, Gabriel González Videla, in the presidential elections of 1946 and even joined González Videla's new government despite its rejection of their formal participation in Radical-led governments throughout the period of popular front policy. In Venezuela the particular twist was that the growing radicalization and militancy of the labor unions took place in the context of expanded collaboration between the labor movement and the government. During the trienio of 1945–1948 Acción Democrática was in power, and affiliated unions strongly supported and collaborated with the government. The Communist party also commanded influence in the labor movement, and it too decided to support what it considered the progressive policies of the government, although a dissident Communist group, the Machamiques, rejected this position.

4. Collapse of Reformist Initiatives and Political Closing ((from 1946–1947 on ). The onset of the Cold War undoubtedly had a major impact in Latin America. In all four countries the Communist party was banned, strong antilabor measures were adopted, the earlier reformist tide was reversed, and in some

cases democratic regimes were overthrown. Although the democratic regime in Chile remained intact, the opportunities for reformist initiatives rapidly dwindled. These trends appeared in response to rising labor militancy and in conjunction with the anti-Communist atmosphere sparked by the Cold War. In 1947 legislation drastically restricted rural unionization and outlawed strikes in the rural sector. The González Videla government embarked on a vigorous anti-Communist campaign, denouncing all strikes as the products of Communist subversion. The Communists were ousted from the cabinet, and in 1948 the Communist party was banned.

In Brazil the transition to a new electoral regime continued on track, but beginning in mid-1946 the government reasserted control over the unions and oversaw a period of retrenchment with respect to labor reforms. The government intervened in union elections, placed many unions under the direct control of the Labor Ministry, and hardened its position on strikes, either restricting them by legislation or repressing them by force. In addition the government reinstated the loyalty test used under Vargas to prevent Communists from becoming union leaders and then banned the Movimiento Unificador dos Trabalhadores (the Communist labor front), the Confederação dos Trabalhadores do Brasil (a new labor central), and finally in 1947 the Communist party itself.

Mexico experienced a similar antireformist period in the late 1940s. In 1946, as we have seen, the government moved to limit political opposition and strengthen one-party dominance. The party sphere was further restricted in 1949, when the Communist party lost its registration. From 1947 to 1949 the government reacted to the growth of a dissident labor movement by intervening in union elections and denying recognition to the new labor central, the UGOCM, created by Lombardo. The result by 1950 was a heavily controlled labor movement from which leftists were excluded.

Venezuela became the most dramatic case of political closing during the second half of the 1940s. In 1948 the coup led by Marcos Pérez Jiménez and his co-conspirators overthrew the electoral regime and brought an abrupt end to the period of reform under Acción Democrática. The military regime, which remained in power until 1958, became synonymous with extreme political and labor repression. Although the first two years of the regime under Carlos Delgado Chalbaud were milder than the dictatorship established by Pérez Jiménez in 1950, the first junta nevertheless quickly moved against Acción Democrática and thereby also the labor movement.

Table 12 summarizes the impact of international forces on domestic trends and indicates that the relationship between them was very close. At this broad correlational level the general hypothesis about the causal importance of the international conjuncture seems to hold up quite well, and events in Latin America demonstrate a substantial fit with the cross-sectional explanation that focuses on external causes. We can now turn to the second perspective.

The Initial Incorporation of Labor as a Critical Juncture

Like the perspective that focuses on international forces, the perspective based on the incorporation of labor offers a set of predictions about labor protest, state-labor relations, political coalitions, governmental reformist initiatives, and political openings and closings. Yet unlike the first perspective it examines the unfolding of internal trajectories of change over time.

According to this argument, the type of initial incorporation of labor marks the critical transition that sets each country along a particular trajectory. The argument centers on change in the historical development of the relations between the state and the working class. It focuses on the politics of the emergence of an organized labor movement, specifically on the domestic political conjuncture that prevailed at the time the labor movement first achieved legal recognition and became incorporated as a legitimate social actor. According to this view, labor incorporation contains some important common features, but different patterns can be distinguished. These differences had lasting impacts on trade union politics and, more broadly, on the

general trajectory of political change and the shaping of the political arena in each country. The argument presents three sequential analytical phases: incorporation, aftermath, and heritage. In the first phase different types of incorporation experiences are identified with their own specific dynamics and contradictions. The contradictions are worked out or resolved during the subsequent aftermath period. These two phases together establish a particular political heritage.

Incorporation Period

Throughout Latin America the state undertook the incorporation of labor to address the "social question": the rising level of working-class protest triggered by dismal labor and social conditions. The response to this question was the creation of institutionalized channels for the resolution of labor-capital conflict. The period of incorporation is identified by the shift in the relationship between the state and labor from one based on repression to one in which state control was exercised through a legalized state-sanctioned labor movement. In this period the state moved to legitimate, support, and shape an institutionalized labor movement.

This change—from repression to legitimation, from exclusion to incorporation—occurred in most of Latin America during the first half of the twentieth century. In Brazil the period of incorporation came during the first Vargas presidency, beginning in 1930, and reached its climax with the creation of the Estado Nôvo in 1937. In Chile the same process occurred in the Alessandri-Ibáñez period, starting with the election of Alessandri in 1920 and reaching its height after Ibáñez formally assumed power in 1927. Mexico had the most extended period of incorporation, beginning in 1917 and culminating during the Cárdenas presidency in the 1930s. Finally, in Venezuela incorporation occurred in the period following the death of Juan Vicente Gómez: it began in 1935 and culminated during the trienio of 1945–1948, when Acción Democrática came to power. (The time lines in figure 2 show the incorporation periods.)

These incorporation periods were the political outcome of economic growth and social change. The rapid expansion of Latin American economies that began during the late nineteenth century led to urbanization and the development of a broad range of new economic activities in commerce and manufacturing. The new economic sectors spawned two new social groups:

the working class and a broad range of middle-sector groups, which included owners, managers, professionals, and other intermediate sectors. These emerging social actors put new items on the political agenda, including the resolution of industrial conflict that arose with growing worker protest and organization, and the transformation or reform of the oligarchic state demanded by the middle sectors.

Incorporation did not occur until the oligarchy had lost control of the state and representatives of middle-sector reformers came to power. This change represented a transition from a laissez-faire state controlled by the traditional oligarchy to a more activist and interventionist state, one more responsive to the urban middle sectors. Even though the oligarchy itself had in some countries put forth an incorporation proposal of its own, no such project was carried out while the oligarchy retained power. Rather it was introduced only as part of a larger transformation of the state—from a laissez-faire state to a more activist state. The new state upheld traditional liberal property rights but took on new social, welfare, and economic responsibilities. Along with these new responsibilities the state adopted a paternalistic stance toward labor, manifested in legislation on such issues as working conditions, the minimum wage, and social security. At the same time, the state attempted to create an institutionalized system of labor relations, casting itself in the role of mediator of class conflicts and arbiter of labor-management relations. These measures as a whole signified the advent of state corporatism, new structures that vertically integrated the emerging urban, industrial society.

Such, then, is the sociological commonality of labor incorporation in Latin America. In the course of economic growth two new emerging classes, the middle sectors and the working class, were integrated into the polity in dominant and subordinate positions, respectively. However, important differences existed among the countries. Though nowhere did the oligarchic state remain intact, oligarchic interests remained powerful to varying degrees in the new, "postoligarchic" state. Hence, a major issue was the relationship—or cleavage—between the traditional oligarchy and the middle sectors seeking to reform the state. The nature of this relationship is reflected in the different types of incorporation projects that were adopted.

Although the middle sectors generally succeeded in challenging oligarchic hegemony, they often found it difficult to establish their own political dominance and to consolidate a more interventionist state. In all four cases under

consideration here, political stalemate between the older and newer dominant classes prevented or stymied reform. Two solutions to the deadlock emerged. In Brazil and Chile the military played a key role in breaking the political impasse and intervened to oversee the introduction of the new state. In Brazil this stage occurred in 1937 when Vargas exploited his military support to abandon the electoral regime and install the authoritarian Estado Nôvo. In Chile the same process began in 1924 when army officers, including Carlos Ibáñez del Campo, intervened in the political process, and it took a more definitive form in 1927 when Ibáñez formally assumed power. In these two cases the authoritarian state enforced a modus vivendi between the traditional oligarchy and the reformist middle sectors. The middle sectors were able to effect the transformation to an activist state while the traditional oligarchy, despite its loss of political control, was able to protect its interests.

In Mexico and Venezuela the oligarchies were relatively weak, and traditional clientelist relations were eroding in the rural societies of these two countries. An alternative strategy to overcome the deadlock became available through the mobilization of labor and the peasantry. In Mexico the protracted stalemate after the revolution of 1910–1917 ended by virtue of the continuous mobilization of popular-sector support by successive governments during the 1920s and 1930s. The height of the incorporation period in Mexico occurred during the government of Cárdenas between 1934 and 1940. in Venezuela, following the death of Gómez in 1935, successive governments embarked on indecisive reforms, failing to satisfy the reformist middle-sector groups, which remained in opposition. During the decade 1935–1945 the reformers mobilized popular support, and finally Acción Democrática, the leading opposition party, gained power in 1945.

The political mobilization of the popular sectors, particularly the working class, represents the critical difference between these two types of incorporation, which may be called state and party incorporation, respectively. In both types of incorporation political leaders sought to respond to the growth of labor unions and class tensions through control over the working class. However, in the two cases of party incorporation, Mexico and Venezuela, political leaders sought not only to control labor and the peasantry but also to win their popular support. State and party incorporation are thus distinguished by this difference in the balance between the control and mobilization of the popular sectors. The main features of the two types of incorporation are outlined in figure 3.

Figure 3. Party versus State Incorporation: Contrasting Patterns of Change

In cases of party incorporation sufficient benefits were offered to induce the dominant part of the labor movement to cooperate with the state, and labor became part of a multiclass alliance. In cases of state incorporation, in which such mobilization was not pursued, labor retained greater political autonomy. This difference in the political position of labor had important and enduring consequences.

Thus, although the two types of incorporation have certain common features, they produced quite different political alliances. State incorporation, as in Brazil and Chile, was based on an accommodationist alliance consisting of an uneasy truce between the reformers and the oligarchy. The emphasis lay on the depoliticization of the working class and the control of the union movement. A highly elaborated labor law defined the system of union representation and exercised control over unions in such spheres as union tactics (particularly the use of the strike), internal governance, and the selection of labor leaders. Independent and leftist unions were repressed and replaced by a state-sponsored and state-penetrated labor movement. With this strong emphasis on control little or no political mobilization of either the urban or rural popular sectors occurred, and little effort was made to incorporate the popular sectors into a populist political party.

Mexico and Venezuela, in contrast, became examples of the alternative model of party incorporation. There, incorporation was based on a populist alliance between elements of the new urban middle sectors, the working class, and, in these two cases, the peasantry.^[10] Unlike the attempt to depoliticize the labor movement under state incorporation, the mobilization strategy central to party incorporation entailed the politicization of the working class. In this way incorporation involved as a first priority not only the integration of labor as a functional group but also its integration as a political movement. The result was the creation of a broad multiclass coalition that found expression in a political party—Acción Democrática in Venezuela and what became the PRI (Partido Revolucionario Institucional) in Mexico. These parties institutionalized the populist alliance and channeled working-class political activity into support for the government. In addition to attracting the working-class vote these parties established organizational links with the labor unions.

The dynamics or logic of support mobilization meant that, compared with state incorporation, party incorporation implied more concessions and a stronger political position for the labor movement. Instead of leftist and

independent unions being repressed, they were tolerated or even became part of the governing coalition. Corporatist labor codes were promulgated, but they imposed fewer constraints on unions and union activity. The same kind of state-penetrated union movement was not established, although mobilization induced the labor movement to support the government and, by virtue of the benefits it received from it, to become dependent on the state. In general the adoption of a mobilization strategy entailed an increase in the political power of labor since its very utility to the political leadership as a political resource depended on its strength.

Party incorporation and the political mobilization of labor did not threaten the basic capitalist orientation of the state but rather did much to co-opt the working class. Nevertheless, working-class mobilization threatened important sectors of society. Thus the dynamic of party incorporation was political polarization: a progressive coalition in power was opposed by the dominant economic sectors, which formed a counterrevolutionary or counterreform alliance.

Aftermath

Each type of incorporation established a distinctive political agenda for the following period—the aftermath—ultimately producing different types of regimes as a political heritage. Here we are concerned with the aftermath period alone, since the heritage period in all four countries occurred after the 1940s (see figure 1). Hence, issues relating to regime outcomes in the heritage period will be mentioned only briefly.

The aftermath of state incorporation unfolded in three steps: (1) conservative governments protected established interests while labor was becoming a participant in the reopened, democratic regime; (2) attempts were made to form populist coalitions both through the formation of a populist party and through multiclass alliances of working-class and middle-sector parties: and (3) these populist experiments collapsed.

The aftermath began when authoritarian regimes that had established state incorporation became discredited and were superseded by democratic openings. At this point labor, which had hitherto been denied a role in the earlier period of demobilization, became an important political actor. Under state incorporation institutional channels for the political participation of workers had not been established, partisan identities among workers had not

been consolidated, and coalitions had not been formed between labor and other classes or political actors. As a result the working class was politically autonomous from governing parties at the same time that it was tightly constrained under the prevailing system of industrial relations.

Labor's role in competitive party politics became a central issue in the aftermath period. With a new opportunity to enter the political arena the labor movement was swiftly revitalized and politicized. In both Brazil and Chile the Communist party quickly reestablished its influence among unionized workers. Meanwhile, middle-sector reformers sought to fill the void left by state incorporation and to shape labor's participation. In this effort the reformers made what might be called, from a comparative perspective, a "belated" attempt to establish a populist party in order to mobilize and channel working-class political participation and to enlist the support of labor for the electoral coalitions they were seeking to construct. These efforts were made by the original leaders of the middle-sector reform movements. In Brazil Vargas himself in 1945 took the initiative in creating the Partido Trabalhista Brasileira as the vehicle for labor representation and participation. In Chile Marmaduque Grove, Ibáñez's original coconspirator in 1924, along with other reformers sought to mobilize the support of the working class through a new populist Socialist party.

This attempt to establish populist parties, however, proved unsuccessful. The middle sectors failed to unite around these parties and instead for the most part supported center or center-right parties that lacked a large working-class base of support. As a result the new movements failed to achieve power and were therefore never in a position to offer sufficient concessions to satisfy their labor constituency. Instead, the aftermath of state incorporation brought a period of coalitional politics in which center or center-right parties, the Partido Social Democrático in Brazil and the Radical party in Chile, won power on the basis of electoral coalitions with the populist parties (and, in Chile, with the Communist party too). However, because oligarchic groups retained substantial influence in Congress, the centrist parties were as strongly drawn toward an accommodationist coalition with the conservatives as they were toward a populist alliance with the working class. As a result the populist parties remained only junior partners in the coalitions and from this position were not able to extract enough from their collaboration to satisfy the working class. In reaction an increasingly radical, anticollaborationist wing emerged within both the populist parties and the labor move-

ments. In both Chile and Brazil these tendencies were reinforced by relatively powerful Communist parties competing for working-class support. Disillusioned by coalition politics and pressured by the Communists, the populist parties developed strong left-wing factions. A process of polarization began, and the period ended with the abandonment of the now discredited pattern of coalition politics. By 1952 in Chile and by 1960 in Brazil the labor movement and the parties or party factions that had attracted working-class support abandoned collaboration with the political center. The aftermath period ended with the collapse and discrediting of populism and coalition politics.

The outcome of state incorporation and its aftermath was a failure to create a strong, stable political center. The weakness of the political center subsequently became one of the most striking features of the party systems of both Brazil and Chile. The long-term legacy of state incorporation during the heritage period may be labeled multiparty polarizing politics . Although Brazil and Chile were far from identical cases, the two countries had in common highly fragmented party systems with a built-in tendency toward polarization. In both countries, in the context of the political and economic pressures of the 1960s, polarization intensified until a broad coalition favoring a military coup emerged. In Brazil in 1964 and in Chile in 1973 the military intervened to establish the two most durable bureaucratic-authoritarian regimes in South America.

In Mexico and Venezuela the main feature at the end of the period of party incorporation was a conservative reaction as the dominant economic sectors resisted the mobilization of labor, the progressive reforms, and their own exclusion from the governing coalition. In Venezuela the conservative reaction culminated in a military coup in 1948 and a decade of repressive, counterreformist rule. In Mexico the same polarization and rightist opposition occurred, but the party managed to stay in power by electing a conservative successor to Cárdenas in 1940. In the effort to regain power in Venezuela or to retain power in Mexico, the populist parties attempted to construct a new centrist bloc to end the political polarization provoked by the populist coalition.

The aftermath period in Mexico and Venezuela thus formed a striking contrast with that of state incorporation in Brazil and Chile. In Mexico and Venezuela the conservative reaction signaled the political limits of reform in late capitalist development. It provided a big incentive to avoid the polariza-

tion of the immediate past, to include the bourgeoisie and middle sectors in the dominant political coalition, and to reconstitute the multiclass coalition based this time in the center right. This effort to reintegrate the right and to create a new governing coalition had four chief features: a programmatic turn to the right; the exclusion of the left from the alliance; the retention of the labor movement (urban and rural) in the alliance by maintaining union-party linkages and labor support; and finally the establishment of conflict-limiting mechanisms to avoid the kind of polarization that had led to the toppling of the Acción Democrática—dominated regime in 1948 and had threatened the dominance of the ruling party in Mexico in 1940. In Mexico the mechanism was the strengthening of one-party dominance and in Venezuela the use of party pacts.

The heritage of party incorporation in Mexico and Venezuela was very different from that of state incorporation in Brazil and Chile. It consisted of a party/political system that was integrative, rather than polarizing, and that institutionalized something approaching a coalition of the whole in contrast to the unstable coalitions in Brazil and Chile. These regimes developed sophisticated conflict-limiting mechanisms that facilitated the formulation of consistent, sustainable policies and avoided the zero-sum conflicts and policy vacillation and immobilization that became the chief features of policies in Brazil and Chile. Mexico and Venezuela also developed strong centrist parties supported by a substantial proportion of the working-class as opposed to a party system that relegated parties with substantial working-class support to a position of nearly permanent opposition. The heritage phase in Mexico and Venezuela was thus characterized by the emergence of a stable, hegemonic regime able to weather the economic crises and the political challenges that confronted the countries of Latin America throughout the 1960s and 1970s and that in Brazil and Chile led to prolonged and repressive military rule.

Combining Perspectives

The above discussion summarizes two quite different approaches to understanding labor politics and regime change. The first examines the course of Latin American politics during the 1940s in light of powerful international factors reflecting the changing relationships among the major world powers.

The second, which focuses on a formative internal transition that in most instances occurred earlier in the century, does not focus on the 1940s per se. Instead, it develops an account of different paths or trajectories of change followed by Latin American countries during periods that do not necessarily coincide and at rates of change that vary considerably among them. From this point of view the 1940s is a rather arbitrary decade that catches countries in different phases in the unfolding of their trajectories. How do the two explanatory schemes intersect? How may they be combined or juxtaposed?

We begin by noting which of the internal phases occur in the 1940s. This decade corresponds to the incorporation or aftermath periods, or both, in all four countries (see figure 1). Two countries, one an example of state incorporation and the other of party incorporation, underwent the transition from incorporation to aftermath. In the other two cases the entire decade corresponded to the aftermath phase. (Figure 2 presents a more detailed chart of the intersection of the two perspectives.)

In an attempt to link the two perspectives we might ask three questions. These questions, as well as an outline of the answers, are presented in figure 4.

The first question considers the two strands of analysis as rival hypotheses. It asks to what extent the outcomes are caused by international events as opposed to internal dynamics. In other words, is the internal explanation spurious? The answer seems to be negative because for each pair of cases (of party and state incorporation) similar or parallel internal dynamics and steps unfolded in different decades, quite independently of the international context.

The second question considers international factors as a complement to internal trajectories. Did the international events of the 1940s affect the timing, intensity , and variations of outcomes that formed part of the internal dynamic? In Brazil there appears little doubt that this was the case. The authoritarian regime of the incorporation period was certainly not going to last forever, and in Latin America the typical pattern is for authoritarian regimes to be replaced by electoral regimes. Nevertheless, according to standard accounts of the period, the international climate as well as direct pressure from the United States favored democratic opening at this time. International factors also appear to account for the particular intensity with which Brazil experienced certain phases of the aftermath period. The end of the popular front policy of the Communists abroad and their readoption of strategies of class confrontation help explain why Brazilian labor experi-

enced an unprecedented, almost explosive reactivation and repoliticization during the immediate aftermath period of 1945–1947. Equally, the international conjuncture often affected the intensity of the conservative orientation of the government in the early aftermath period. Conservative from the start, in 1947 the Dutra government cracked down on labor, reintroduced a number of Estado Nôvo controls, and restored the ban on the newly legalized Communist party. The conservative reaction in Brazil during the late 1940s was stronger than that in the analytically comparable period in Chile in the 1930s. While specific conditions in the two countries help to explain this difference, there is no doubt that the contrasting international context was important.

International factors also affected the distinct features of Chile's aftermath period, in particular, the timing, duration, and specific character of the

coalition governments. In Chile the coalitions of the aftermath period included the participation of the Communists and took the form of the Chilean Popular Front, a direct reflection of Comintern policy and the general international context of anti-Fascist collaboration. In Brazil, in contrast, the entire coalitional period corresponded to the Cold War so that Communist party participation was precluded from the beginning. The timing of the Cold War was also significant in Chilean politics. According to the internal trajectories thesis, the aftermath of state incorporation consisted of an abortive attempt to establish a viable, multiclass political center. Looking at this issue from the vantage point of the international conjuncture creates new perspectives on the degree of this populist failure. In Chile the onset of the Cold War and the regionwide move to proscribe the Communist party came at the end of the aftermath period, when the populist experiment and coalition politics were breaking down of their own weight, rather than at the beginning of the period as in Brazil. This contrast in timing helps explain why in Chile collaboration and coalition politics were discredited in the eyes of the labor movement and the left much more decisively than in Brazil. The discrediting of coalition politics in turn might have contributed to the more intense radicalization and polarization that subsequently occurred in Chile in comparison with Brazil.

In Mexico international factors might also explain some of the distinctive aspects of the aftermath period, in particular why the conservative reaction to party incorporation remained relatively mild, that is, why in Mexico, unlike in most countries, the incorporating party avoided a military coup. Perhaps these distinctive features reflected the timing of the initial aftermath period during the Comintern's second, more extreme, popular front line favoring multiclass collaboration. During this period, as we have seen, most sectors were willing to support broad anti-Fascist fronts while labor remained pledged to restraint in support of the war effort. Under these circumstances labor proved more willing to acquiesce in the conservatization typical of the aftermath period of party incorporation. Thus international factors facilitated the transition to conservatism and the avoidance of an institutional breakdown. In other countries, such as Venezuela, in contrast, the aftermath period coincided with the Cold War and the internationally influenced period of political closing during the late 1940s, when international trends may have encouraged a harsher, more repressive conservative reaction. In these countries a military coup overthrew the incorporating regime and banned the incorporating party.^[11]

The timing of the internally unfolding steps in relation to the international

events illuminates another aspect of Mexican politics in the 1940s. Because of Mexico's success in maintaining institutional continuity, a populist party rather than a counterreformist, anti-populist military government was in power during the internationally defined phase of political closing during the Cold War. The existence of a governing populist party may have contributed to the development of an alternative to conventional labor repression in the form of the distinctive Mexican institution known as charrismo : the informal coercive control exercised by the party over unions to eliminate leftist or independent leadership. In this way the transition to Cold War politics in the late 1940s might help to explain the distinctive character of state-labor relations in Mexico.

In Venezuela, despite the coincidence of timing, regime change did not seem to be strongly influenced by the international conjuncture. There the high point of party incorporation began in 1945. Though Venezuela's democratic, reformist period coincided with pressures pushing in the same direction as international factors, the latter did not appear very significant in determining either the timing or the type of labor incorporation. The timing itself seems more closely related to the internal situation, particularly the breakdown of negotiations over the selection of candidates for the forthcoming elections. This episode provoked the military coup that carried Acción Democrática to power and initiated the trienio . Nor did international factors influence the type of incorporation period. The mobilization of the popular sectors began in the mid-1930s at a time when the mobilizing parties remained in opposition. Similarly, it was in the 1930s that Acción Democrática adopted a democratic and reformist position. Thus when it gained power in 1945 its populist coalition and its commitment to democracy and reform were already in place.

The end of the incorporation period in Venezuela occurred in 1948 when a military coup ended the trienio . The timing of this event too appeared predictable in light of prevailing international conditions. However, accounts of Venezuelan politics rarely refer to the international context, and other considerations cast doubt on its importance. The impact of the international conjuncture in the late 1940s is usually understood in light of the turn against the Communists, but in Venezuela it was the populists and Acción Democrática, rather than the two competing Communist factions, that became the target of the conservative military. Indeed for several years afterward one of the Communist factions collaborated with the government.

Thus, international factors obviously had important consequences in

Latin America. Yet the international conjuncture was never more powerful in shaping political outcomes than the internal trajectory. In a number of instances international factors affected the timing and intensity of the stages following labor incorporation; they help to round out the picture and explain some of the variations within patterns of change for which the internal argument alone could not account.

Finally, the third question raises the issue of how the internal dynamics explain the differential impact of international events. As table 12 showed, sometimes the expected outcomes of the 1940s did not occur. The deviation can often be explained by internal trends. Thus, when the two logics contradicted one another and pointed toward different outcomes, the internal dynamic took precedence.

Among the four countries considered, Brazil appeared to follow most closely the periodization suggested by international forces; yet even in Brazil the deviations can be explained by the internally driven logic. An example is the political opening that occurred at a time when the Cold War might have suggested a political closing. In 1950 a reformist opening began that could be understood only as an integral part of the aftermath of state incorporation—as part of the attempt to establish acceptable channels for the political participation of labor. Mexico too generally conformed to the phases suggested by the 1940s conjuncture but again with an exception: the period of political opening and reform in 1944–1946 failed to occur. Instead Mexico was moving in the opposite direction toward political closing, the strengthening of one-party dominance, and the marginalization of labor—a pattern explained by the internal dynamics perspective. In Mexico this period marked the aftermath of party incorporation and hence constituted a period of conservative reaction to the prior reformist period. In this case, then, the logics of the two perspectives led in opposite directions, and the internal trajectory proved stronger than the international conjuncture.

In Chile international influences became most visible in the relationship between the policy of the Comintern and the formation of the Chilean Popular Front in the mid-1930s and in the subsequent coalitions among the Communist, Socialist, and Radical parties in the 1940s. Although the formation of the Popular Front cannot be understood without reference to international factors, these factors alone are insufficient to explain the persistence of the coalitions during the 1940s. Contrary to the internationalist hypothesis, the Chilean Popular Front survived the German-Soviet pact, and the

coalitions that followed the Popular Front continued for some time beyond the end of World War II, when international factors pointed to increased class conflict and protest. Although there was a wave of strikes in Chile after 1945, in other respects the shift to the politics of confrontation was limited despite strong ties between a relatively strong, class-conscious labor movement and Marxist political parties—conditions that might have suggested particular receptivity to the influence of international communism. In 1946, when Communist parties elsewhere in Latin America were returning to confrontation, the Chilean party not only joined the governing coalition but also for the first time formally joined and participated in the government—a step it had declined to take during the years of Comintern's popular front policy. The various deviations from the expected patterns of the 1940s reflected the playing out of the attempts at coalition politics typical of the aftermath phase. At the end of World War II the possibilities and drawbacks of collaboration were still being explored, the issues remained unresolved, and all parties were vacillating. Thus although ultimately discredited, at this point coalitional politics temporarily continued despite pressures from the international conjuncture toward renewed political confrontation.

Venezuela too conformed to the expected patterns of the 1940s except in one instance. Although labor protest increased dramatically in the immediate postwar years, this activity was less a result of the end of popular front collaboration on the part of the Communist-influenced labor movement than a reflection of the new activism of Acción Democrática unions undertaken with the support of the trienio government. This outcome reflected the general pattern of party incorporation and the mobilization of labor support that characterized it.

Conclusion

This analysis has juxtaposed and combined two different analytic perspectives on labor politics and regime change in Latin America. Both the international and the internal perspectives contribute to our understanding of Latin America during the 1940s and hence should be viewed not as rival but as complementary explanations. There is no doubt that the international events of the 1940s left a strong imprint on the political landscape of Latin America. Yet these events failed to deflect the unfolding of internal trajectories set in motion by the initial incorporation of labor. Overall, the internal dynamics

prove stronger causal factors. However, the international factors do help explain the distinctive features and variations of the internal patterns, filling in some of the details and helping to account for the timing and intensity of the steps as they unfolded in each country.