Democracy's breakdown and the rise of fascism: the ...

3 downloads 10 Views 231KB Size Report
May 2, 2001 - quently, Second Republic urban democratic political elites were left to attempt .... In 1931, the Socialist PSOE and the other parties of the San ...

Social History Vol. 26 No. 2 May 2001

Sara Schatz

Democracy’s breakdown and the rise of fascism: the case of the Spanish Second Republic, 1931–6 The social bases of the demise of democratic rule in Spain, Italy and Germany have been widely debated theoretically in the literature on interwar Europe (Moore, 1966; Stephens, 1989; Luebbert, 1987; Linz, 1976, 1978; Burton and Higley, 1989; Eley, 1983; Simon, 1978). Currently, a number of plausible, competing, mass-based accounts exist in the literature explaining the breakdown of democracy and the rise of fascism in Spain, Italy and Germany. Barrington Moore’s (1966) argument that the crucial condition for the rise of fascism was the development of a coalition of large landholders, the crown (the monarch, bureaucracy and military, i.e. the state), and a politically dependent bourgeoisie of medium strength has been advanced as an explanation for the rise of fascism in Germany and, to a lesser extent, in Italy and Spain (Stephens, 1989; also Gerschenkron, 1943, for the German case). Luebbert (1987), on the other hand, attributes the breakdown of democracy in Spain, Italy and Germany to the failure of the interwar Liberal-Labour {Lib-Lab} class alliance of peasants and urban Socialists to sustain itself. My objective in this article is to specify further empirically mass-based accounts of the social origins of democracy and modern authoritarianism in the Spanish Second Republic (1931–6). The Spanish case provides an interesting ‘test’ or speciŽ cation of mass theories of the breakdown of democracy and the rise of fascism. Spain can be understood to Ž t the general pattern of interwar social democracies based on a Lib-Lab social coalition (Luebbert, 1987: 469–72). The new democratic regime in Spain (1931) enabled forces of the moderate left to attack institutions and social groups privileged by the old regime through labour legislation, anti-clerical legislation and land reforms. At the same time, Spain did not experience a revolutionary breakthrough by peasants in the nineteenth century in combination with the urban strata. Consequently, Second Republic urban democratic political elites were left to attempt serious structural agrarian reform, including land redistribution, in an unstable parliamentary democracy without strong uniform peasant support (Malefakis, 1970). The absence of strong peasant support for land reform and the retention of an important share of political power by landed elites have been postulated as central factors explaining the rise of fascism in Spain (Moore, 1966: 438; ff. 4). In specifying mass-based general theories of the breakdown of democracy and the rise of fascism in interwar Europe, I contend that the polarizing dynamic of agrarian reform is a central factor undermining the Spanish Second Republic. SpeciŽ cally, I argue: 1. The emergence and consolidation of a social democratic alliance of urban workers and the rural proletariat provided Social History ISSN 0307-1022 print/ISSN 1470-1200 online © 2001 Taylor & Francis Ltd http://www.tandf.co.uk/journals DOI: 10.1080/03071020110039742

146

Social History

vol. 26 : no. 2

the base of support for the Republic’s agrarian reforms; 2. The latifundista effect contributed to the emergence of a ‘proto’-fascist coalition of conservative political power and large landowners; a social alliance that facilitated the counter-agrarian reform of the early Francoist regime. The Spanish Second Republic’s sequence of political development (weak parliamentary government attempting agrarian reforms, the rise of the latifundista effect and the ultimate failure of land reform) is also demonstrated comparatively in other authoritarian conservative cases in interwar Europe (Hungary, Italy, Greece, Poland and Japan). The latifundia is generally operated by the owner or his agent with hired workers (rural proletariat) whereas the large estate is operated primarily by share tenants (peasants) (Moore, 1945: 257). In Spain, the latifundista provinces and the provinces of large estates overlap in the southern and south-central regions. Those provinces in 1930 in which over 55 per cent of the landholdings were held by large landowners – those with 250 hectares or more – included Salamanca, Cáceres, Badajoz, Huelva, Sevilla, Cádiz and Jaén (Malefakis, 1970: 30). The provinces where 45–55 per cent of the landholdings in 1930 were held by large landowners included Córdoba, Ciudad Real, Albacete and Toledo. The southern and south-central provinces were also those in which most land was redistributed under the revived agrarian reform campaign of the leftist Popular Front goverment (February–March 1936). This article is thus justiŽ ed in conceptualizing a ‘latifundista effect’ as incorporating land tenure arrangements in both the southern and south-central provinces. THE SPANISH CASE Historians and scholars of the Spanish Second Republic typically divide the Republic into three periods (Jackson, 1972; Julia, 1984; Linz, 1978; Malefakis, 1970; Payne, 1993; Preston, 1978; Robinson, 1970; Thomas, 1961). From 1931 to 1933 the country was governed by a coalition of radicals, centre-left Republicans and bourgeois Socialists. In this Ž rst two-year period or bienio, the abandonment of support by the Radical Party led to the parliamentary victory and the governance of the nation by the centre-right, CEDA, and other rightist groups (September 1933). In the second bienio (1933–6), corruption in the Radical Party, the rebellion of the Catalan Generalitat, the rigidity of the right and the aborted Asturian revolution were events leading up to the election of February 1936. The collapse of the shaky coalition of the middle class sympathetic to the church but ambivalent about land reform, the restoration of Jesuit properties, and the maintenance of Catholic schools and divisions over regional autonomy meant the collapse of centre-right government and of middle-class consolidation (Luebbert, 1987: 469–72). At that time Socialists, leftist Republicans and communists ran a ‘Popular Front’ campaign and won the election (Linz, 1978: 165). This victory led to the short third period, which lasted only from February to June 1936. It ended abruptly when the army rebelled on 17 July 1936, Spain divided into pro-coup (Loyalist) and anti-coup (Nationalist) regions, and the civil war began. My argument is that the polarization dynamic that occurred over the short life of the Republic was importantly rooted in the question of agrarian con ict. Previous theories of the rise of polarization and the breakdown of democracy in the Spanish Second Republic point to global expressions of labour activism (total number of strikes, work days lost and striking workers) and to the high incidence of Anarchist- and Socialist-led strikes as explanations for

May 2001

Democracy and fascism in Spain

147

extremism, especially during the Ž rst bienio, as major factors contributing to democracy’s breakdown (Linz, 1978: 162; Thomas, 1961). I do not deny the importance of numerous strikes in creating an overall perception of leftist-originated social unrest during the Republic. Indeed, the number of strikes was one variable of labour unrest used in the statistical analysis in this article.1 Nevertheless, the measure of strikes alone is quite urban-focused and can overlook the centrality of the agrarian dynamic in the breakdown of the Republic and the rise of fascism (see also Stephens, 1989: 1063). One key point is the extent to which the rural and urban proletariat and the peasantry were uniŽ ed as politically efŽ cacious actors who lent coherent political support to urban democratic political elites undertaking land reform. This is why the salience of the problem of the agrarian proletariat and agrarian reform is central for mass-based theories of the Spanish Second Republic. The attempt of urban democratic reformist political elites to solve land inequalities is one central starting point for the analysis of the social origins of democracy’s breakdown (Moore, 1966: 436–7; Luebbert, 1987: 469–72). In 1931, the Socialist PSOE and the other parties of the San Sebastian coalition issued a series of emergency decrees: they forbade expulsion of small tenants; forbade owners from withdrawing their land from cultivation; established rules that favoured collectives in the renting of large properties; established an eight-hour day and a de facto wage increase; gave the Federación Nacional de Trabajadores de la Tierra (FNTT) (National Federation of Agrarian Workers) greatly enhanced control over wage negotiations; and established arbitration boards that generally enforced the leading position of the FNTT (Malefakis, 1970; Luebbert, 1987). The chief function of these arbitration boards or jurados mixtos was thus to supervise the labour legislation of the Republic and also to provide an institutional measure for collective bargaining (Malefakis, 1970: 167; Ministry of Labour decrees of 7 May 1931 and 11 July 1931). Each board was made up of labour and business leaders divided evenly, while the decisive power of the chair was delegated to a representative appointed by the government. Because of the supervisory and punitive faculties of the mixed juries to levy Ž nes against infractions of general labour legislation, their power was considerable (Malefakis, 1970: 167). The ‘mixed juries’ initially displayed a pro-agricultural worker bias as the new labour conciliation committees spread Ž rst in areas of Socialist strength (Carmona, 1989: 408). These efforts of Second Republic democratic elites to cope with the chronic structural problem of inequality in landownership through agrarian reform and better collective bargaining for peasants, in turn, created resistance among large farmers and business organizations to their attempts to redistribute lands and to implement increases in agricultural wages (Malefakis, 1970: 167; Ministry of Labour decrees of 7 May 1931 and 11 July 1931, Confederación Patronal Española {El Sol, 27/1/33}; Círculo de la Unión Mercantil e Industrial {El Sol, 19/7/33}; Jackson, 1965: 123). The rise in agricultural wages during the Ž rst bienio was reversed by the newly elected centre-right government of the second bienio only after protests by business groups (Jackson, 1965: 123; Confederación Patronal Española {El Sol, 27/1/33}; Círculo de la Unión Mercantil e Industrial {El Sol, 19/7/33}). In the brief third period (February–July 1936), the leftist Popular Front government managed to reinstate the pro-agricultural worker policies of the Ž rst bienio until the rightist coup aborted it in July 1936.

1 The total number of strikes was not found to be statistically signiŽ cant as an explanatory factor of

electoral support in the three elections (1931, 1933, 1936). These results are available upon request.

148

Social History

vol. 26 : no. 2

The failure of political elites to implement land reform has consistently been proposed as a central factor explaining the breakdown of democracy in Spain (Linz, 1978: 154; Malefakis, 1970). Linz argues that it was the failure of political elites to centre land reform on a small number of very large landowners in areas of the country with serious unemployment problems and social discontent that was a signiŽ cant contributing factor in democracy’s breakdown. The inconsistent, under-resourced, only partially implemented and sometimes inequitable nature of Spanish land reform (1931–6) – conŽ scating from a very large number of small- to middlesized farmers as well as from a few large ones – caused it to run into resistance not only from the latter but also from some of the former (Malefakis, 1970; Linz, 1978: 154). Mass-based accounts of the Ž nal breakdown of Spanish democracy stress the importance of the social structure of the rural sector and the particular features of the Spanish land tenure situation as important factors precipating democracy’s demise. In Spain, large estates dominated the countryside more than in Italy, Austria and Germany, and Spain was heavily dependent upon agriculture. Stephens argues that, even more than in Italy, the immediate reaction of the upper classes to the political organization of the working class and landless agricultural workers was ‘a central, if not the central, dynamic in the breakdown of the regime’. Thus, the agrarian question was a focal point of class con ict and the hostility of large landowners to state intervention in local agrarian markets and to agrarian reform were also important factors in the breakdown of the Second Spanish Republic (Stephens, 1989: 1063). Robledo (1996: 336) and Barciela (1996: 352) argue that there was a series of conservative counter-agrarian reform measures aimed at re-establishing pre-Republic order in the countryside. The Ž rst of these involved the actions of the 1934–5 second bienio government which halted land expropriation entirely by September 1934 (Robledo, 1996: 270). Under the centreright CEDA government’s new ‘Law of Reform of the {Agrarian} Reform’, only seasonal occupations were allowed: a process not necessarily detrimental to propertied interests. The expropriation of properties was halted. The second and Ž nal stage of the counter-agrarian reform came on 24 July 1936, seven days after the military uprising. On that day, the Francoist ‘Junta de Defensa Nacional’ adopted legal measures by which property owners were granted absolute freedom in labour-hiring decisions, the freedom to dismiss tenants from the land and the de-valorization of rural property (Robledo, 1996: 336). With these counter-reforms, the Junta realized its immediate objective: to paralyse the agrarian reforms of the Republic and to suffocate peasant resettlements (Barciela, 1996: 352). To foreshadow my argument in this article, I contend that there is a central regional effect of the agrarian question on the polarization and the breakdown of democracy. Clear regional effects are evident in the implementation of agrarian reform. A regional analysis of the Ž rst (31 December 1931) and the second wave of agrarian reform under the Popular Front government (March–July 1936) shows that the latifundista provinces were overwhelmingly represented in the seizure and redistribution of lands and in the resettlement of peasants (Malefakis, 1970: xi, 281; 377: Appendix). There was also a signiŽ cant latifundista effect in the March–July 1936 period when the Popular Front government vastly enlarged the scope of land redistribution. More land was redistributed during the third bienio than in the entire previous history of the Republic (Malefakis, 1970: 377) as Popular Front political elites targeted landowners of seignorial (hereditary) origin located importantly in the latifundista provinces. This acceleration, in turn, ampliŽ ed the circles of discontent among large landowners (Robledo, 1996: 318). The latifundista effect contributed to the emergence of a counter-reformist ‘proto’-fascist coalition of conservative

May 2001

Democracy and fascism in Spain

149

political power and large landowners, a social alliance that facilitated the counter-agrarian reform of the early Francoist regime. Furthermore, I argue that the labour consequences of the latifundista effect meant that early, urban, democratic political elites were left to attempt land redistribution without uniform peasant support. Regional differences within the Spanish peasantry and political divisions within the southern rural proletariat over the rate and timing of land reform in the Ž rst bienio (1931–3) inhibited the development of a politically united southern rural proletariat. Government intervention in the second bienio (1934–6) and the 17 July 1936 army rebellion effectively terminated the movement toward the political uniŽ cation of the rural and urban proletariat that had gained force by the third bienio (February–June 1936). DISTINCTIVE THEORETICAL ASSUMPTIONS One mass-based approach to the failure of land reform and the subsequent political polarization refers to the fact that Spain did not experience a revolutionary breakthrough by peasants in the nineteenth century in combination with the urban strata (Moore, 1966: 437–8). In Spain, as in the case of interwar Italy, Greece, Hungary and Poland, the retention of a very substantial share in political power by the landed elite and the political weakness of peasants were social structural factors that diminished the likelihood of successful agrarian reform policies altering the social structure. The absence of a previous peasant revolution meant that Second Republic (1931–6) urban democratic political elites were left to attempt serious structural agrarian reforms, including land redistribution, in an unstable parliamentary democracy without strong uniform peasant support. Moore (1966: 438) argued: As they proceeded with conservative modernization, these semi-parliamentary governments tried to preserve as much of the original social structure as they could, Ž tting large sections into the new building wherever possible. The results had some resemblance to present-day Victorian houses with modern electrical kitchens but insufŽ cient bathrooms and leaky pipes hidden decorously behind newly plastered walls. Ultimately the makeshifts collapsed. Spain, as well as other weak democracies such as those in Poland, Hungary, Greece and Japan, during the interwar period, went through a similar sequence of political development: the inability of reformist elites to ‘cope with the severe problems of the day’ and their reluctance or inability to bring about fundamental structural changes ‘opened the door to fascism’ (Moore, 1966: 438). The failure of reformist elites to ensure effective land reform is also the starting point for Luebbert’s (1987) mass-based analysis of the social bases of democracy’s breakdown in Spain (1931–6). Successful social democracy occurred when urban Socialists were able to organize a coalition with the Spanish countryside. The success of the ‘Lib-Lab’ social democracy in Spain depended upon the ability of urban Socialists to maintain the electoral cohesion of this alliance with the countryside. The success of this alliance, in turn, rested especially on the relative ability of social democratic elites to deliver beneŽ ts to the rural proletariat. Where Luebbert and others depart most signiŽ cantly from Moore’s (1966) agrarian-based structural analysis of the breakdown of democracy in interwar Europe is with respect to the political role of the landed classes. These scholars argue that Moore assigned an unrealistically

150

Social History

vol. 26 : no. 2

important political role to the landed upper classes in the breakdown of democracy in Spain. Luebbert (1987: 460) argues that large landed elites were ‘everywhere a spent force in the 1920s and 1930s; they were no longer in control of the middle peasantry and the rural proletariat’. Skocpol (1973: 23–4) faults Moore for exaggerating the political in uence of the large landed upper classes and with ‘systematically underestimat{ing} the degree to which bureaucratic and military elites are likely to act similarly across times and countries regardless of their class backgrounds’. Instead, Luebbert theorizes an alternative social actor to the large landed upper class when explaining the social bases of polarization in Spain: the middle peasantry. Accordingly, the continued political salience of the middle peasantry in the countryside meant that the land reform efforts of the early San Sebastian coalition ‘contained the seeds of its own failure’ (Luebbert 1987: 460). Middle peasants were particularly threatened by the campaign for land reform and redistribution as well as by the unionization of agrarian workers. When the Socialist union FNTT, driven by its growing support among the agrarian proletariat, radicalized the agricultural issue and exacerbated agrarian class con ict, it set the stage for further polarization in the 1933–6 period. According to Luebbert (1987: 472): In June 1934, the FNTT called a nationwide strike in response to the repeal of the 1931 Municipal Labour Act, the act that had created the arbitration boards on which so much of the FNTT’s in uence depended. The strike received only lukewarm support from the urban wing of the UGT and eventually failed, causing the loss of tens of thousands of union members. In its aftermath, and especially in the wake of the failed October Revolution, the gulf between the two wings of the Socialist movement, and between the movement and the middle peasants, became even greater. The polarization came to a head in February 1936, when middle peasants in the centre and north voted overwhelmingly for the CEDA and agrarian labourers in the centre and south-west voted overwhelmingly for the Socialists. This electoral outcome mirrored the alignments that formed on behalf of fascism and the Republic a few months later. DISTINCTIVE THEORETICAL PREDICTIONS The latifundista effect In demonstrating the regional latifundista effect in the agrarian question and class con ict, I begin with the following predictions derived from the mass-based accounts of the breakdown of democracy in Spain’s Second Republic (1931–6) in my empirical analysis of Spain’s Ž fty provinces. First, positive evidence for the Luebbert thesis in the Spanish case should show evidence of a social democratic alliance measurable by electoral alignments over the three elections of the Republic (1931, 1933, 1936). In addition, positive support for the Luebbert thesis should include evidence of an initial alliance between the urban working class and the family/middle peasantry in the Ž rst two elections of the Republic (1931 and 1933). Positive evidence for the Moore thesis would show strong electoral support by large landowners for conservative, pro-agrarian parties, especially in the 1933 election. Second, the case of Spain’s Second Republic (1931–6) also poses a cogent and stringent test of mass-based accounts, because the political geography of the 1930s showed signiŽ cant

May 2001

Democracy and fascism in Spain

151

variation in landed upper-class strength by region. Small farms (54.5 per cent to 78.7 per cent) were concentrated strongly in the northern region and, to a lesser extent, in the central region (33.4 per cent to 66.1 per cent). Medium-sized landholdings (14.1 per cent to 34.2 per cent) increased in the central area, but the great majority of large landholders were concentrated in southern Spain (35.8 per cent to 46.0 per cent) (Malefakis, 1970: 18). Southern and south-central Spain was thus the stronghold of the large estates: the latifundia of 250 hectares and above. If the Moore hypothesis that fascism relies on ‘labour-repressive’ patterns of agriculture has positive predictive value, one should also Ž nd evidence for the emergence of a regionally based ‘proto’-social coalition of forces characteristic of the labour-repressive regime existing in southern Spain, although not in northern and central Spain, in the breakdown of democracy and the initial rise of fascism. In other words, southern regional landholding patterns should be found that pave the way for the national rise of rightist authoritarianism, if the Moore thesis has general validity in the Spanish case. For Luebbert (1987), on the other hand, the strength of the middle peasantry’s resistance to land reform should produce the opposite regional effects: weak or no effects of large landholding patterns on the breakdown of democracy and signiŽ cant northern/north-central regional support of the middle peasantry for the fascist solution by 1936. Labour consequences of the latifundista effect The Spanish case also provides a good test of the labour consequences of the latifundista effect. In the southern and south-central latifundia provinces, the labour situation was markedly different from that obtaining in the north. The latifundia heritage prevented the spread of small family-sized farms, so the largest social class in southern and south-central Spain consisted of landless day labourers or the rural proletariat. Landless day labourers – those who had no access to the land but worked it for others, without permanent employment – made up as much as 43.3 per cent of the active male rural population as late as 1956, two-and-a-half times as numerous as in the rest of Spain (Malefakis, 1970: 94–6). The long history of its latifundian heritage created a property structure in southern Spain that allowed for greater owner abuse over small tenants and sharecroppers (the yunteros) than in the rest of Spain. High rents, short leases and frequent subleasing meant that the yunteros of south-central Spain, especially in the province of Badajoz, faced greater insecurity and conditions which led to their widespread resentment as compared with northern small tenants and sharecroppers. As a general rule, the middle- and lower-middle-class family operated as owners and tenants in the northern, non-latifundista Spain and fundamentally accepted the existing social order (Malefakis, 1970: 5). Landowning middle-class peasants and tenants (labour-employing entrepreneurs who owned or leased landholdings of 100 hectares or less) and landowning lowermiddle-class peasant proprietors and tenants (family operating with landholdings of 10 hectares or less) made up a large 76 per cent of the total active male population in northern Spain (Badell-García, 1960). The owning or leasing of even small plots of land had a conservative in uence even on northern lower-middle-class peasants and tenants whose otherwise economically precarious existence might have led them to adopt a more radical political position. Moore hypothesized that the absence of a previous revolutionary breakthrough by peasants in the nineteenth century, in combination with the urban strata, would leave urban democratic political elites to attempt serious structural agrarian reform, including land redistribution,

152

Social History

vol. 26 : no. 2

without strong, uniform peasant support (1966: 437–8). The lack of strong, uniform support for the Republic’s land reform campaign in the Ž rst bienio (1931–3) is evident even within the southern rural proletariat. The maximalist Anarchists demanded the immediate conŽ scation, without compensation, of all large properties and the total abolition of all taxes, rents and mortgages on small proprietors in the southern provinces of Cádiz, Córdoba, Huelva, Málaga and Sevilla, where they held greater political control. The Socialist rural unions were initially willing to follow the moderate, gradual land reform campaign of the Ž rst coalition government in the south-central provinces of Badajoz, Cáceres, Salamanca, Jaén, Ciudad Real, Albacete, Granada and Toledo, where their political control was greater. Rural Socialist leaders did, however, unofŽ cially encourage farm invasions by the yunteros in Badajoz (Malefakis, 1970: 111). These political divisions within the southern rural proletariat mirrored distinctive regional tendencies in the rural social structure of latifundia Spain. In the Anarchist-dominated provinces, landless day labourers accounted for 78.1 per cent of the rural proletariat. In the Socialist-dominated provinces, landless day labourers constituted a smaller 58.1 per cent of the rural proletariat, although the largest provincial concentration of yunteros was nevertheless located in the Socialist-controlled provinces of Badajoz, Cáceres and Salamanca (Malefakis, 1970: 111). In the second and third bienios (1934–6, February–May 1936), a new political uniŽ cation of the southern rural proletariat occurred. Socialist unions radicalized toward the Anarchist position, demanding immediate land reform in 1934. This radicalization occurred in the wake of a massive drop in rural wages brought about by the second bienio centre-right government’s lax enforcement of labour laws, its replacement of Socialist personnel on the mixed jury boards, and its reversal of the 1931 términos municipales law that had protected local southern Socialist rural unions from the in ux of cheaper migrant harvest labour from Portugal and Galicia (Malefakis, 1970: 325). Yet neither the massive uniŽ ed Socialist-Anarchist led peasant strike in 1563 villages in mainly latifundia provinces in June 1934, nor the massive farm invasions in March 1936 by the yunteros in Badajoz, Toledo, Salamanca, Murcia and Madrid, nor the most serious and effectively coordinated rural–urban strike wave of Anarchist, Socialist and Communist rural and urban unions (1 May 1936 to 18 July 1936) in the history of the Republic created a revolutionary breakthrough by the rural and urban proletariat. In the second bienio (1934–6), the centre-right government’s capacity to return the old rural oligarchy to positions of power, and its successful mixture of conciliatory tactics toward striking moderates and prompt government repression and press censorship on the day of the June 1934 strike, rendered the southern rural Socialist unions ‘an inconsequential revolutionary force ... and gave the rural oligarchy complete control of the Spanish countryside’ (Malefakis, 1970: 336). In the third bienio (February–May 1936), the Popular Front government managed to slow and channel the March 1936 farm invasions by the southern rural proletariat into legalized land transfers by May 1936. They did so by accelerating peasant resettlement, by massively increasing state occupation of latifundia land in the southern provinces and by doubling rural wages for harvest labour (Malefakis, 1970: 369). Ultimately, the army rebellion of 17 July 1936 effectively terminated the movement towards a revolutionary breakthrough by the rural and urban proletariat, at least in the Nationalist provinces. 2 2 Malefakis (1970: 386–7) could be read to support the counter-factual thesis that in the Loyalist

provinces, a revolutionary breakthrough by the rural and urban proletariat might have been possible

May 2001

Democracy and fascism in Spain

153

The Luebbert thesis assumes that splits in the electoral base of the Lib-Lab coalition should become evident as northern middle-peasants become increasingly alienated by the radicalization of the Socialist urban and rural proletariat in the second bienio (1934–6). This alienation of the northern middle-peasantry stems initially from its resistance to the ruedo and lease provisions of land reform of the Ž rst bienio (1931–3) which made lands near villages and lands leased by absentee landlords with as little as 20 hectares subject to expropriation (Malefakis, 1970: 213–17). These provisions increased the enemies of land reform by extending it to the small and medium owners, who made up the majority in the northern provinces. Even here, however, a southern latifundista effect is evident. In southern Spain, the rural population did not live scattered throughout the countryside, as in the north, but was concentrated in relatively few villages widely separated from each other, so the land near the village assumed a special importance. Distant lands could not be worked by small peasants unless new homes were built for them near the plot they were to till. The Socialists supported the ruedo provision of the early land reform because they hoped that if enough land were reserved to individual settlers near the villages, opposition to the collective cultivation of the large estates located at a distance would diminish (Malefakis, 1970: 207). Luebbert hypothesizes that the resistance of the middle-peasantry only increased over the life of the Republic, as the newly united calls for immediate land reform by Anarchist-Socialist rural unions in 1934 further threatened the interests of medium peasants in particular. Positive evidence for the Luebbert thesis should include a pro-rightist vote by middle peasants in the 1936 election. RESULTS The 1931 elections and the dawn of the Spanish Republic: the lack of a clear class base Democracy in Second Republic Spain (1931–6) came about through the victory of Republicans and Socialists in the larger cities in the municipal elections of 1931 and in the subsequent ‘proclamation of the Revolutionary Committee, composed of Republicans and Socialist conspirators. The army maintained a passive role and did nothing to stop the Republican conspirators against the Primo de Rivera dictatorship’ (Stephens, 1989: 1063). My results show the lack of either a working-class or a rural proletariat base to voter alignments in the 1931 elections. Although Lib-Labism was re ected in the party coalition of the San Sebastian pact (radicals, centre-left Republicans and bourgeois Socialists) (Luebbert, 1987), I did not Ž nd evidence of an alliance between urban workers and the rural proletariat underlying the Lib-Lab political alliance. Instead, my Ž ndings support the argument that the institution of the democratic Republic was brought about by developments which did not re ect the power balance in Spanish society as a whole. The lack of clear working-class support in the social base of the San Sebastián coalition re ected in Table 1 supports the contention that working-class forces were the beneŽ ciary of had the Loyalists not lost the war. By May 1937, for example, one-half to two-thirds of all land in Republican areas had been seized by the rural proletariat and working-class organizations. Ironically,

most of this was land from small and medium-sized peasants since, as aforenoted, most of the latifundio districts had fallen to the Nationalists almost immediately after the outbreak of hostilities.

Social History

154

vol. 26 : no. 2

Table 1. Regression of right–left party index provincial variables in the 1931 election Variables Constant Manual workers CNT members 1931 Land 1 hectare or less Land 10 hectares or less Land 100 hectares or less Land 250 hectares or less R2 Adjusted R2 N

Standardized coefŽ cient

P-value

2.197 2 .651 2 1.569 2 .073 2 1.639 1.532 2 .342

.033 2 .518 2 .124 2 .942 2 .108 .133 2 .734

0.148 0.030 50

the introduction of democracy more than the initiator of it (Stephens, 1989: 1063). The Socialists (PSOE) played a secondary role in the transition to democracy and only joined the Revolutionary Committee in its Ž nal stages. Furthermore, the lack of rural proletariat support for the San Sebastián parties further explains why the coalition parties were compelled to enact the series of emergency decrees meant to reform rural inequality, in order to cement the political support of rural workers, especially the landless day labourers (Malefakis, 1970; Luebbert, 1987). The rise of agrarian-based con ict: the 1933 election By 1933, however, the alliance of urban workers and the rural proletariat is clearly re ected in the electoral alignments of the 1933 election. The regression results reported in Table 2 reveal strong correlations between voting patterns and measures of labour social organization and measures of the rural proletariat. This is evident in the signiŽ cant positive predictors of voting for the Left parties of measures of the urban working class (PSOE and UGT membership) and measures of the rural proletariat (agricultural unemployment). The 1933 electoral alignments in Spain show the urban Socialist strategy of alliance with the rural proletariat (Luebbert, 1987)3 and are consistent with a general interwar pattern of labour support for liberal democracy (Brademas, 1974; Kelsey, 1991; Malefakis, 1970; Mintz, 1982; Peirats, 1971). Yet the results also show important splits in political support in the form of the positive support of lower-middle-class peasants and the negative support of the rural and urban proletariat for the Centre-Left parties. These schisms provide further evidence of the lack of uniŽ ed peasant support for agrarian reform in the Ž rst bienio (1931–3). The tendency of lower-middleclass peasants and tenants to support the Centre-Left, largely Republican parties in the 1933 election is consistent with the thesis that their traditional orientation ensured their opposition would be expressed through normal political channels rather than in association with revolutionary labour organizations. As Malefakis (1970: 96) notes, small peasant proprietors and even labour-employing middle-class tenants and sharecroppers ‘may sometimes have supported 3 The exceptions to this general pattern were poor peasants in Rumania and workers in Hungary,

who gave disproportionate support to fascists (Nagy-Talavera, 1970: 152–4;Vago, 1987: 308–15).

May 2001

Democracy and fascism in Spain

155

Table 2. Regression of 1933 ideological tendencies on provincial variables Variables

Right Standard coefŽ cient P-value

Constant Agricultural unemployed Unemployment UGT membership PSOE membership Land 1 hectare or less Land 10 hectares or less Land 100 hectares or less Land 250 hectares or less R2 Adjusted R2 N

Centre-Right Stand co. P-value

Centre-Left Stand co. P-value

5.537 2 .990 2 .838 1.183 2 .138 2 1.106 1.250 1.089 3.412

.000 2.463 .018 2 .328 .186 .853 2 .407 2 .622 .537 .244 2 1.793 2 .080 2 .891 .671 .506 2 .275 .346 .731 .218 1.168 .250 2 .223 2 .824 2 .282 .001* 2 .353 2 .726 2

1.509 2.295 .067 3.286 1.565 .882 3.235 1.159 .155

.293 .155 50

.113 -.060

.403 .287

.139 2 .027* 2 .947 2 .002* .125 .383 .002* 2 .253 2 2 .877 2

Left Stand co.

P-value

.670 2.923 .694 3.780 2.073 .417 .720 .003 .949

.507 .006* .491 .001* .045* .679 .475 2 .998 2 .348

.551 .488

* p > .05

Republican parties, but they did not play an important role in the upheavals that threatened southern Spain’. The Moore thesis also Ž nds support in Table 2 in so far as the continued in uence of latifundia agriculture in the Spanish Second Republic is illustrated by the very strong positive effect of large landholdings (250 hectares or more) on the 1933 electoral right. This Ž nding appears to re ect the large landholders’ hostility to moderate left (PSOE) state intervention in local agrarian labour markets and the strengthening of labour’s power (Stephens, 1989: 1063). It also points towards the importance of southern regional effects in shaping national developments during the period of the Republic. The link between large latifundia landholders in the south and the Agrarian Party is well known. It is evident in the numerous attempts by Agrarian Party deputies in the Ž rst bienio to stall the passage of the Socialists’ agrarian reform bill and to overturn the provisions most damaging to large proprietors (Malefakis, 1970: 190, 262). Indeed, political agitation by large landowners organized through the Agrarian Party led to a number of anti-government rallies protesting against this legislation during the Ž rst bienio. This agitation culminated in the gigantic rally held in Madrid on 18 September 1933, in which 100,000 landowners from all parts of Spain gathered in opposition to the agrarian reform bill (Malefakis, 1970: 275). Thus protest on the right also contributed to the fall of the early coalition government of the Ž rst bienio. The data highlight the social bases of the electoral patterning of party support in the 1933 election and point toward the regionally speciŽ c importance of the agrarian question as a focus of class con ict in the Second Republic. The February 1936 election: the urban worker–rural proletariat alliance continues Further evidence of the alliance of urban workers with the rural proletariat is also revealed in predictors of the Popular Front vote in the February 1936 election in Table 3. The social bases

Social History

156

vol. 26 : no. 2

Table 3. Regression of right–left party index provincial variables in the 1936 election Variables Constant Manual workers CNT members 1931 Land 1 hectare or less Land 10 hectares or less Land 100 hectares or less Land 250 hectares or less R2 Adjusted R2 N

Standardized coefŽ cient 1.116 3.071 2.260 2 .232 1.019 2 1.504 2 .644

P-value .271 .004* .029* 2 .818 .314 2 .140 .523

0.311 0.214 50

* p > .05

of electoral alignments in the 1936 elections continue to reveal core support for the Popular Front government in both the agricultural proletariat and in urban workers (manual workers, CNT membership, 1931). The continued alliance with the countryside at the level of the social bases of party support for the left in the February 1936 election further explains why Popular Front political elites thought it possible to revive the land reform initiative and to accelerate considerably the actual number of hectares redistributed. The results do not, however, show regional evidence of the political support of the central or northern middle peasantry for rightism (Luebbert, 1987: 470–2). Nor is any effect of large landownership on political support for rightist parties evident in the February 1936 election, as can be predicted from the Moore thesis (1966: 437–48). Did, then, electoral alignments in the 1936 election mirror the alignments that formed on behalf of fascism and the Republic a few months later (July 1936 onward) as hypothesized (Luebbert, 1987: 470–2)? Social structural correlates of the rise of fascism in the Nationalist zones (October 1936): the rise of a ‘proto’-fascist coalition To specify this question in greater clarity, I measure the social forces in Spain in October 1936. By this time, the ‘counter’-agrarian reform measure had already been adopted by the Francoist forces in the Nationalist provinces of early civil war Spain (Barciela, 1996: 352). Therefore, one can predict that the social bases of support in Nationalist provinces should provide evidence for the Moore thesis of a movement toward the classic fascist alliance between conservative political power, the landowning classes and labour-repressive agriculture. The results in Table 4 present multiple regression results on provincial measures of Nationalist support in October 1936. In fact, one sees large landownership positively associated with proxy measures of the rise of Spanish fascism in 1936. Because the great majority of large landholders were concentrated in southern and south-central Spain (Malefakis, 1970: 18), the results show a strong effect for the stronghold of the large estates: the latifundia of 250 hectares and above on provincial measures of Nationalist strength. Small and medium-sized landholding are

May 2001

Democracy and fascism in Spain

157

Table 4. Regression of Nationalist militias on provincial variables Variables Constant Land 1 hectare or less Land 10 hectares or less Land 100 hectares or less Land 250 hectares or less Occupation clergy Education level R2 Adjusted R2 N

Standardized coefŽ cient

P-value

5.205 2 5.287 2 4.764 .402 .001 .786 2 .441

0.000 2 0.000* 2 0.000* 0.690 0.013* 0.436 2 0.662

0.551 0.488 50

* p > .05

not strongly associated with Nationalist militia strength. As Malefakis (1970: 386) argues, most of the latifundia districts fell to the Nationalists immediately after the outbreak of the hostilities in mid-1936. Taken together, the data show a clear latifundista regional effect associated with proxy measures of Francoism in the early rise of Spanish fascism. Thus, the results suggest the emergence of a regionally based ‘proto’-social coalition of forces characteristic of the labourrepressive regime existing in latifundista regions of Spain. DISCUSSION The results suggest that the rise of a polarizing class dynamic over the short life of the Republic has an important agrarian dimension. The lack of a clear class basis for the Republic in the 1931 election soon gave way, by the 1933 election, to a polarizing dynamic with agrarian roots. In the 1931–3 period, the numerous attempts by Agrarian Party deputies in the Ž rst bienio to stall the passage of the Socialists’ agrarian reform bill and to overturn the provisions most damaging to large proprietors (Malefakis, 1970: 190, 262); the rise of anti-governmental political agitation by large landowners organized through the Agrarian Party protesting agrarian reform legislation during the Ž rst bienio; and the electoral support of large landowners for the Agrarian Party in the 1933 election are evidence of how agrarian con icts contributed to the fall of the early coalition government. The results of the 1933 election, in turn, set the stage for the subsequent right–left polarization in the 1933–6 period. The halting of land redistribution by the centre-right government (1934–6) was followed by a signiŽ cant upsurge in the redistribution of lands under the Popular Front government (February–July 1936). My analysis of the social bases of electoral alignments in the 1936 elections shows strong continuity in the alliance of urban workers with the rural proletariat and may help explain why Popular Front political elites perceived they had the social bases to re-accelerate agrarian reforms. Finally, the regional results also show that the rise of fascism was accompanied by the movement toward the classic fascist alliance in latifundista Spain between the landed upper classes

158

Social History

vol. 26 : no. 2

and the far right during the Republic (1933) and in the early rise of fascism in Nationalist zones (1936). My results thus identify the social bases of the Nationalist’s counter-agrarian reform measures as importantly located in latifundista provinces which further specify the important regional sources of support underlying agrarian reform and class con ict. This speciŽ cation, in turn, suggests movement toward the classic fascist alliance in which conservative political power and the large landowning classes ally in the legal formulation of labour-repressive agricultural policies. The centrality of the agrarian question, the cycle of agrarian reform and counter-agrarian reform in Spain can be generalized across a series of ‘authoritarian conservative’ cases of the interwar period. Reformist political elites in Italy, Poland, Hungary, Greece and interwar Japan were unable to implement fundamental agrarian reform, in part because of central resistance from landed interests. For example, in interwar Hungary, land distribution was characterized by a relatively small number of very large estates and a large number of extremely small plots; approximately less than four-tenths of 1 per cent of the total number of landowners owned 43 per cent of the total land area (Moore, 1945: 232). The short-lived revolutionary Bela Kun government (1919) contemplated agrarian reforms of an extensive nature. Yet the subsequent counter-revolutionary Horty (1920) and Bethlen governments (1921–31) proposed only a modest land reform aimed at transferring land from large latifundiary estates to peasant proprietors (Janos, 1982: 195–204). Even then, the National Association of Agrarian Proprietors largely emasculated the impending land reform bill and expropriation from large estates was limited to only a speciŽ ed fraction of them (Dovring, 1970: 25). In the Hungarian case, as in the case of the Francoist counter-agrarian reform, the dominance of the landowning nobility was maintained (Moore, 1945: 223). Not until Communist rule did Hungary experience signiŽ cant land reform when the 1945–7 land reform eliminated nearly all holdings over 100 kat. and converted many into state farms (Dovring, 1970: 25). In Poland and Greece after the First World War, agrarian reforms were implemented to abolish aspects of feudal tenures. Yet, in Poland, the agrarian reforms did not abolish the large estate: the actual amount of land redistributed fell quite short of the original government plan, and only rarely did expropriation occur (Moore, 1945: 224). In Greece, the interwar land reform removed feudal aspects of property relations but did not abolish tenancy and, at the termination of the reform, large estates still comprised over half of the total cultivated land (Moore, 1945: 254). Only in Rumania, where large estates were few in number, was the interwar land reform ‘radical’ in character (Moore, 1945: 242). The Japanese and Italian interwar cases also show a similar pattern of attempted agrarian reform and counter-reform. In Japan, government reforms under the Meiji regime (1868–1913) made peasants formerly bound to the land by feudal ties become independent proprietors but brought no beneŽ ts to cultivators of tenanted land (Dore, 1959: 14). Following peasant uprisings and peasant unionization, Japanese landlords had amalgamated into a national body by 1925 and had considerable dominance over the political parties. Subsequent interwar land reforms (1926, 1931, 1936) aimed at attacking the problem of tenancy did not seriously challenge the interests of landlords (Dore, 1959: 108). Serious land reform in Japan only occurred under the US occupation (1945) which, by the end of the Second World War, coincided with the fact that there was no great proŽ t to be derived from being a landlord (Dore, 1959: 114). Italy was the classic European home of both the latifundia and the large estate and the modern remnants of feudal tenures lasted well into the interwar period. The fascist government’s in uence

May 2001

Democracy and fascism in Spain

159

was towards straightforward entrenchment of existing concentration of ownership; the fascists did not appreciably change the distribution of property despite the great concentration of agricultural property in the interwar period (Moore, 1945: 259). A project of land reform affecting the Sicilian latifundistas immediately before the Second World War was not carried out. The Spanish Second Republic’s sequence of political development (weak parliamentary government attempting agrarian reforms, the rise of the latifundista effect and the ultimate failure of land reform) has been overlooked as re ecting a distinctive path of development in which fragile parliamentary democracies give way to authoritarian-conservative regimes not ‘fully’ fascist. Because scholars have relied heavily on the German case as the proto-typic case of fascism involving a ruling-class alliance of labour-repressive landlords, sections of the bourgeoisie and the state, Moore’s conception of comparative fascism as involving a ‘revolution from above’ has been argued to be weakest in the Spanish, and other cases of authoritarian conservatism (Skocpol, 1973: 23–4). SpeciŽ cally, it is contended that because an analogous modernizing project to its German counterpart did not materialize in late nineteenth-century Spain, the Moore thesis is weak (Stephens, 1989: 1062–3). Yet, over-reliance on the German case as proto-typical of fascist development may be somewhat misleading since it was characterized by a powerful landed aristocracy but also experienced rapid and self-sustaining industrialization in the nineteenth century. Authoritarian state-led modernization did eventually occur after the fall of the Second Spanish Republic under the Franco regime. In the Spanish countryside (1939–45), state-led agricultural modernization led to signiŽ cant improvements in rural infrastructure and land productivity as well as to limited peasant resettlement (Barciela, 1996: 386–7). More to the point, the early modernization of agriculture under Franco disproportionately favoured large proprietors as the government bought up land they wanted to sell. Expropriation was avoided at all costs (Ortega, 1979; Tamames, 1977). State-led industrialization projects were aimed at reducing potential agrarian con ict without changing the structure of property relations. Industrialization was particularly concentrated in those southern and south-central latifundista regions (Badajoz, Jaén) where expropriation and agrarian con ict had been most severe at the end of the Spanish Republic (Barciela, 1996: 387–8). The Moore thesis thus appears to ‘Ž t’ the Spanish case. The social origins of the alliance of conservative political power and the large landowning classes necessary for the emergence of Spain’s state-led modernizing ‘revolution from above’ are evident at the collapse of the Second Republic. Although rooted in a social democratic Lib-Lab alliance of urban workers and the rural proletariat, structural agrarian problems, the effective political patterning of latifundista interests and the absence of uniŽ ed, effective, peasant political action were factors that proved difŽ cult for reformist democratic political elites to surmount. University of Florida

APPENDIX Design and variables Spain in the 1930s is ideal for assessing the breakdown of democracy and the rise of fascism. Spanish society had recently undergone a democratization phase to emerge from a period of military authoritarianism (1923–31) following a long experiment with liberalism. The major

Social History

160

vol. 26 : no. 2

social divisions were, however, unresolved. Indeed, Luebbert (1991: 99–100) contends that Spain emerged from its earlier experiment with liberalism (1898–1923) suffering the burden of backwardness – widespread illiteracy and apathy, a state-dependent middle class, regions of latifundia agriculture, highly centralized but inefŽ cient state structures – while simultaneously experiencing many ethnic, regional, linguistic and religious con icts. Linz (1978: 142–3) argues that Spanish society at the dawn of the Second Republic still experienced deep regional, cultural, class, religious and linguistic divisions. Unfortunately, no individual-level data on socio-economic position, voting behaviour or party membership are available for the period of the Spanish Second Republic. Analyses must be based on aggregate electoral results, party membership rolls, censuses and government records gathered at the provincial level. To infer the individual-level relationship between any two variables from their covariation at the aggregate level, Brustein argues (1991: 658), one requires: several strong assumptions about the statistical properties of their joint distribution to avoid committing the so-called ecological fallacy. Nevertheless, {one can} . . . suggest individuallevel interpretations from group data because these data are the best available for studying political processes of this period. . . . In any case, ecologically-based inferences, although risky, are more likely to be correct than false. Similarly, I attempt here to note conditions that reduce concern about the ecological fallacy. My research results are based on the outcome variables of social forces regressed on pertinent independent variables for the 1931–6 period. The Spanish Second Republic lasted only a relatively short time, marked a democratizing transition from the military authoritarian rule of Primo de Rivera (1923–31), and was succeeded by a fascist authoritarian dictatorship (1939–77) following a long civil war (1936–9). Therefore I combine insights from theories of democratization with standard indices used in analyses of the democratic Spanish Second Republic and indices from theories of the social bases for the rise of fascism. Thus I take into account electoral outcomes, state policies, measures of workers’ activism, land tenancy and the role of the clergy and the church. Table A reports the coverage and the years measured of all variables used Table A. Descriptive statistics for variables used in the analyses4 Variable

Year variable measured

Education Index of land tenancy Unemployment rates Index of worker organization Index of political polarization Index of clerical density Index of rise of fascism

1930 1933 1933 (July) 1919, 1931, 1932, 1933 1931, 1933, 1936 1930 1936

4 The list of total control variables collected but discarded in the analysis includes the following additional provincial-level measures: total population, total mortality, infant mortality, natural

population growth, illiteracy,index of Radical Party support, total number of strikes and index of education. A complete list detailing the construction of each variable is available on request.

May 2001

Democracy and fascism in Spain

161

in this analysis. All variables are expressed as a proportion of the provincial population unless otherwise noted. I begin with seven indicators to measure class relations: education, unemployment rates in July 1933, Anarchist membership in 1919 and 1931 (CNT – Confederación Nacional de Trabajadores), average membership in the moderate Socialist union (UGT – Union General de Trabajadores) for 1932–3, PSOE membership in 1932 and land tenancy patterns in 1933. Education is measured by the percentage of secondary school-age children (16–20) per province in school in 1930. Unemployment is further subdivided to create an additional indicator that captures the agriculturally unemployed (also as of July 1933). Anarchist membership, membership in UGT and membership in the Socialist Party (PSOE) all measure how class interests were grounded organizationally and institutionally. As Rueschmeyer, Stephens and Stephens (1992: 51) argue in their analysis of democratization, measurements of workers’ autonomous historical organization are critical for understanding their relative in uence on other classes. The connection between labour-repressive agriculture, fascism and the propertied classes’ support of authoritarian rightism (Moore, 1966) is measured through an index of land tenancy. I calculate my provincial index of landed wealth so that I can assess the relative effect of very small-scale property owners (holding less than one hectare of land), small property owners (10 hectares or less), medium-sized proprietors (100 hectares or less) and large-scale property owners (250 hectares or less) on electoral outcomes – that is, as proxies of polarization. My property index builds on Badell-García’s (1960) work on Spanish landownership patterns in the 1930s. As aforementioned, in Spain in 1930, the regional distribution of landholdings between small, medium and large farms showed a marked pattern that affected rural development. Finally, the struggle over the role of the Catholic church in Spanish society was a central issue of the Republic (Stephens, 1989: 1063). Spanish historians have shown that the Catholic church strongly supported the conservatism of the old regime and looked to the state for positive moral values and active social policies (Montero, 1960). Where a church is deeply implicated in the old-regime state, one can hypothesize that its conservatism may support authoritarian tendencies. To assess the possibility as a factor in the breakdown of the Second Republic, I constructed an index of authoritarian rightism as measured by the percentage of clergy per province in 1930 (Jiménez, 1969: 198–200). Electoral and regional measures of state–civil society relations, 1931–6 My electoral measures refer to state–civil society relations during the Republic and are represented by the pattern of right–left ideological tendencies in the 1931, 1933 and 1936 elections. Measurements of right–left extremism in both the 1931 municipal election and the bipolar right–left 1936 election are used widely as indices of right–left polarization in the literature (Cibrián, 1978; García Andreu, 1985; Linz and Tusell, 1977; Mir, 1985). The 1933 election re ects a mid-point in the life of the Republic and evidence of Lib-Labism, pro-agrarian right and/or working-class leftist political alliances should be present by this election. Constructing a provisional index of rightism–leftism for the decisive 1933 election is more complicated because of the complex ideological spectrum of parties that participated in it. To cope with the fragmentation, I constructed a four-part index expressed as a percentage of the

162

Social History

vol. 26 : no. 2

population, using data from Irwin (1991, Appendix I: 290–340, xi–xx). This process involved a series of steps. First, I constructed a four-part generic right-to-left ideological scale (right, centre-right, centre-left, left) by province (I discuss this index in note 5).5 I classiŽ ed the following parties as ‘right’, ‘centre-right’, ‘centre-left’ and ‘left’, discussed in detail in note 6.6 Next, I took the average number of votes in each ideological category for each province. Third, I added the averages within all categories to form a new universe. Fourth, in most cases, I calculated the number of votes in each of the four categories as a percentage of the total per province. In cases where the overall number of votes in a party was very small, I took a fraction of the ideological category based on the proportion of the party membership to the overall number of votes in the province. I then added this fraction to the total for that ideological category. The 1936 election was the turning point in the short life of the Second Republic because it was the last free election before the outbreak of civil war (Linz and Tussell, 1977: 27). I follow Linz and Tussell’s (1977: 64) dichotomous right–left index of political polarization in that election, which expresses right–left ideological tendencies by the proportion of votes per region and per district for the PSOE and the CEDA. Other outcome measures seek to assess the balance of social forces in the latter months of 1936 immediately after the formal breakdown of parliamentary democracy and in the early rise of fascism (July–December 1936). Because the Republic effectively divided into two distinctive political entities with the dawn and early period of the civil war (July–December 1936) – Loyalist or governmental zones and Nationalist zones – it was necessary to construct measures which adequately represent regions in which fascist forces predominated in the early rise of fascism. Recalling that the Moore thesis assumes that fascism relies on ‘labour-repressive’ patterns of 5

In constructing this index, I classiŽ ed the orientation of the different parties according to their basic ideological platforms on a variety of issues such as class and labour–business relations, property rights, agrarian reform, Catholicism, centralism versus regionalism, the military, the monarchy and Republicanism. I relied on the accounts of Preston (1978); Manjón (1976); Bermejo Martín (1984); Payne (1962, 1970); Heywood (1990); Carmona (1989); Morodo (1985); Vilanova (1986); Julia (1977); and Linz (1978). 6 I classiŽ ed the following parties as ‘right’: Comunión Tradicionalista, Partido Republicano Conservador, Renovación Española, Confederación Española de Derechas Autónomas, Derecha Independiente, Agrario, Partit Agrari De Catalunya, Partida Republicano Radical, Acción Popular, Monárquico Independiente, Liberal Demócrata, Unión de Derechas, Derecha Regional Valenciana. The following parties were classiŽ ed as ‘centreright’: Partido Nacionalista Vasco, Partido Republicano Radical, Partido Republicano Conservador, Acción Republicana, Partido Progresista, Regionalista, Centrista, Lliga Catalana, Radical Independiente Valencianista, Republicano Agrario, Agrupación al

Servicio de la República, Autonomista, Bloque Campesino, Derecha Independiente, Partido Unión Republicana Autonomista. I classiŽ ed the following parties as ‘centre-left’: Acción Republicana, Liberal Independiente, Esquerra Regional de Catalunya, Partido Republicano Demócratico Federal, Unio Socialista de Catalunya, Accio Catalana Republicana, Partit Republicano Republica d’Esquerra, Federales ‘del Pacte’, Unio de Rabassaires, Federalista, Radical Socialista, Radical Independiente, Organización Republicano Gallego Autónomo, Republicano Guipuzcoano, Republicano Autonomista, Republicano Agrario, Galleguista, Castellano Agrario Republicano, Republicano Vigués, Partido Progresista, Republicano Católico Independiente, Social Ibérico, Radical Disidente, Partido Republico Democrático Federal, Acción Valencianista Republicana, Republicano Castellano Católico. The following parties were classiŽ ed as ‘left’: Partido Comunista de España, Partido Socialista Obrero Español, Bloque Obrero y Camperol, Partit Comunista de Catalunya, Extrema Izquierda Federal, Campesino, Radical Socialista Independiente, Socialista Disdenite.

May 2001

Democracy and fascism in Spain

163

agriculture (1966), an adequate evaluation must test for landholding patterns in areas that can be deŽ ned as exclusively pro-fascist in the breakdown of democracy and the rise of fascism in 1936. I use Nationalist militia membership as a sufŽ ciently uncontaminated proxy regional measure of support for fascism in the early rise of Franco (Soti, 1996).7 Because they fought exclusively on Franco’s side and against the Republican government, I use the total number of volunteers in the Nationalist militias as a regional indicator of support for fascism in Nationalist zones (Casas de la Vega, 1974). Consistent with all the other variables, my estimates of Nationalist militia members are expressed as a percentage of the provincial population taken in October 1936. Methods I used the following statistical methods: bivariate relationships between continuous outcomes were assessed by Pearson and Spearman correlation coefŽ cients. I compared means across groups via one-way analysis of variance (ANOVA) methods and post hoc t-tests. I modelled multivariate relations between a set of predictors and a continuous outcome, using stepwise linear regression methods. I employed knowledge of the historical subject matter in constructing a parsimonious model. In addition, I used forward and backward regressions with approximately the same result in constructing my model assumptions. My statistical approach is close to that of a ‘best possible subsets’ regression model. BIBLIOGRAPHY Badell-García, Gabriel (1960) ‘La Distribución de la Propiedad Agrícola de España en las diferentes categorías de Ž ncas’, Revista de Estudios Agro-Sociales, vol. 34. Barciela, Carlos (1996) ‘La Contrareforma Agraria y la Política de Colonización del Primer Franquismo’ in Reformas y Políticas Agrarias en la Historia de España (Madrid: Ministerio de Agricultura, Pesca y Alimentación), 351–98. Bermejo Martín, F. (1984) La IIa República en Logroño: Elecciones y Contexto Político (Logroño: Communidad Autónoma de la Rioja). Blinkhorn, M. (1987) ‘The Iberian states’ in D. Muhlberger (ed.), The Social Basis of European Fascist Movements (London: Croom Helm). Brademas, J. (1974) Anarchosindicalismo y revolución en Espana (1930–1937) (Barcelona: Ariel). Brustein, William (1991) ‘The “Red Menace” and the rise of Italian fascism’, American Sociological Review, lvi (October), 652–64. Burton, Michael and Higley, John (1989) ‘The elite variable in democratic breakdowns’, American Sociological Review, liv (February 1989), 17–32. Carmona, A. (1989) El Trabajo Industrial en la España Contemporanea (1874–1936) (Barcelona: Anthropos). Casas de la Vega, P. (1974) Las Militias Nacionales (Madrid: Editora Nacional).

7 The Nationalist militias began as the paramilitary wing of the right-wing Falange Party (Payne, 1962: 45, 63, 81–3). In the early period of the Republic, militia members were engaged mainly in street Ž ghting and street battles against militant leftist youths (Jackson, 1972: 180–1; Payne, 1993: 199–200) and were very few in number. By 1936, however, the violence of the Nationalist militias and the damage they in icted on the regime were

substantial and out of proportion to their membership, and contributed strongly to the fear that public order had broken down (Blinkhorn, 1987). Early in the civil war (June 1936–April 1937), the total number of volunteers for these militias grew enormously; more than 4000 volunteers served in the Aragon region alone (Payne, 1962: 212). By spring 1937 it is estimated that their membership totalled about 151,000 (Payne, 1962: 46).

164

Social History

vol. 26 : no. 2

Cibrián, Ramiro (1978) ‘Violencia política y crisis democrática: España en 1936’, REP, vi (November– December), 81–115. Círculo de la Unión Mercantil e Industrial (1933) El Sol, 7/19/33. Confederación Patronal Española (1933) El Sol, 27/1/33. Dore, R. P. (1959) Land Reform in Japan (Oxford: Oxford University Press). Dovring, Folke (1970) Land Reform in Hungary (Washington DC: Agency for International Development), Country Paper (June). Eley, Geoff (1983) ‘What produces fascism: preindustrial traditions of a crisis of a capitalist state’, Politics and Society, xii, 1, 53–82. García, Andreu M. (1985) Alicante En Las Elecciones Republicanas 1931–1936 (Alicante: Universidad de Alicante). Gerschenkron, Alexander (1943) Bread and Democracy in Germany (Berkeley: University of California Press). Heywood, P. (1990) Marxism and the Failure of Organized Socialism in Spain, 1879–1936 (Cambridge: Cambridge University Press). Irwin,William J. (1991) The 1933 Cortes Election: Origin of the Bienio Negro (New York and London: Garland Publishing). Jackson, G. (1965) The Spanish Republic and the Civil War 1931–1939 (Princeton, NJ: Princeton University Press). Jackson, G. (1972) The Spanish Civil War (Chicago: Quadrangle Books). Janos, Andrew C. (1982) The Politics of Backwardness in Hungary 1825–1945 (Princeton: Princeton University Press). Jiménez, Ramírez, Manuel (1969) Los grupos de presión en la segunda República Española (Madrid: Editorial Tecnos). Julia, S. (1977) La Izquierda Del PSOE (1935–1936) (Madrid: Siglo XXI). Julia, S. (1984) De la Ž esta popular a la lucha de clases (Madrid: Siglo XXI). Kelsey, G. (1991) Anarchosyndicalism, Libertarian Communism and the State:The CNT in Zaragoza and Aragon, 1930–1937 (Amsterdam: Kluwer). Linz, J. (1976) ‘Some notes toward a comparative study of fascism in sociological historical perspective’ in W. Laqueur (ed.), Fascism: A Reader’s Guide (Berkeley: University of California Press). Linz, J. (1978) ‘From great hopes to civil war: the breakdown of democracy in Spain’ in J. Linz and A. Stephan (eds), The Breakdown of Democratic Regimes: Europe (Baltimore: Johns Hopkins University Press). Linz, J. and Tussell, J. (1977) ‘Hacia un análisis regional de las elecciones de 1936’ in Revista Española de la Opinión Pública, xlviii (April–June), 27–68. Luebbert, Gregory M. (1987) ‘Social foundations of political order in interwar Europe’, World Politics, xxxix, 4, 449–78. Luebbert, Gregory M. (1991) Liberalism, Fascism or Social Democracy: Social Classes and the Political Origins of Regimes in Interwar Europe (New York and Oxford: Oxford University Press). Malefakis, E. (1970) Agrarian Reform and Peasant Revolution in Spain (New Haven: Yale University Press). Manjón, O. (1976) El Partido Republicano Radical 1908–1936 (Madrid: Tebas). Ministry of Labour (1931) Decrees of 7 May and 11 July. Mintz, J. (1982) The Anarchists of Casas Viejas (Chicago: University of Chicago Press). Mir, C. (1985) Lleida (1890–1936): Caciquisme Polític I Lluita Electoral (Montserrat: l’Abadia de Montserrat). Montero Moreno, A. M. (1960) Historía de la persecución en España (Madrid: Biblioteca de Autores Cristianos). Moore, Barrington Jnr (1966) Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World (New York: Beacon). Moore,Wilbert E. (1945) Economic Demography of Eastern and Southern Europe (Geneva: League of Nations). Morodo, R. (1985) Los origines ideológicos del franquismo: Acción Española, 2nd edn (Madrid: Alianza). Nagy-Talavera, N. (1970) The Green Shirts and the Others: A History of Fascism in Hungary and Romania (Stanford: Hoover Institute Press). Ortego, Cantero N. (1979) Politíca Agraria y Dominación del Espacio. Origenes, Caracterización y resultados de la política de colonización planteada en la España posterior a la Guerra Civil (Madrid: Ayuso). Payne, S. G. (1962) Falange: A History of Spanish Fascism (London: Oxford University Press). Payne, S. G. (1970) The Spanish Revolution (New York: Norton).

May 2001

Democracy and fascism in Spain

165

Payne, S. G. (1993) Spain’s First Democracy: The Second Republic, 1931–36 (Madison, Wisc.: University of Wisconsin Press). Peirats, J. (1971) La CNT en la Revolución Española, 2nd edn, 3 vols (Paris: Ediciones CNT). Preston, P. (1978) The Coming of the Spanish Civil War. Reform, Reaction and Revolution in the Second Republic (New York: Barnes & Noble). Robinson R. (1970) The Origins of Franco’s Spain.The Right, the Republic and Revolution, 1931–1936 (Newton Abbot: David & Charles). Robledo, Ricardo (1996) ‘Politica y Reforma Agraria: De la Restauración a la IIa República (1868/74–1939) in Reformas y Políticas Agrarias en la Historia de España (Madrid: Ministerio de Agricultura, Pesca y Alimentación), 247–350. Rueschmeyer, Dietrich, Stephens, Evelyn and Stephens, John D. (1992) Capitalist Development and Democracy (Chicago: University of Chicago Press). Simon, Walter (1978) ‘Democracy in the shadow of imposed sovereignty: the First Republic of Austria’ in J. Linz and A. Stephan (eds), The Breakdown of Democratic Regimes: Europe (Baltimore: Johns Hopkins University Press), 80–121. Skocpol,Theda (1973) ‘A critical review of Barrington Moore’s Social Origins of Dictatorship and Democracy’, Politics and Society (Fall, 1993), 1–34. Soti, J. (1996) ‘La Represión’ in J. Tussell and S. Payne, La Guerra que dividió a España: una nueva visión (Madrid: Temas de Hoy). Stephens, John D. (1989) ‘Democratic transition and breakdown in western Europe, 1870–1939: a test of the Moore thesis’, American Journal of Sociology, xciv, 5, 1019–77. Tamames, R. (1977) Estructura Económica de España, vol. 1 (Madrid: Biblioteca Universitaria Guadiana). Thomas, H. (1961) The Spanish Civil War (London: Eyre & Spottiswoode). Vago, R. (1987) ‘Eastern Europe’ in D. Muhlberger (ed.), The Social Basis of European Fascist Movements (London: Croom Helm). Vilanova, M. (1986) Atlas Electoral de Catalunya durant la Segona República (Barcelona: Magrana).

Copyright of Social History is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.