INTRODUCTION
⌅The state has been considered the main cause and the principal agent of the spiral of political violence that destabilised and polarised the short life of the Spanish Second Republic. The premises of this interpretative frame were established by Manuel Ballbé (1985Ballbé, M. (1985) Orden público y militarismo en la España constitucional (1812-1983). Madrid: Alianza Editorial., pp. 318-320, 335-339), who argued that the governments did not demilitarise nor democratise the administration of public order, but rather implemented authoritarian and partisan policies that radicalised the opposition. Michael Mann (2004Mann, M. (2004) Fascists. Cambridge: Cambridge University Press., p. 314) adds that most of the killings were produced by state agencies, with leftist groups as the victims. Chris Ealham (2005Ealham, C. (2005) La lucha por Barcelona: Clase, cultura y conflicto 1898-1937. Madrid: Alianza Editorial., pp. 134-136), likewise, describes the perpetuation of repressive, anti-working class mechanisms of control that undermined civil liberties. Rafael Cruz (2006Cruz, R. (2006) En el nombre del pueblo: República, rebelión y guerra en la España de 1936. Madrid: Siglo XXI., pp. 34-42, 118, 166) underlines that police interventions to control protests caused the majority of the deadly clashes. According to him, restrictive exclusion policies forced protesters to resort to more violent forms of mobilisation. Eduardo González Calleja (2014González Calleja, E. (2014) En nombre de la autoridad: La defensa del orden público durante la Segunda República española (1931-1936). Granada: Comares., pp. 51-53, 322-326) sums up that the democratisation of the police was a failure because the Republican leaders continued prioritising the principle of authority over the citizens’ rights and policemen kept treating conflicts in an almost military manner instead of taking a more proportionate, flexible approach.
The postulates of this reading have been questioned from two different lines of argument. The first one decreases the state’s responsibility without denying the heavy-handed character of the policing methods. Instead, it emphasises the causal weight of the brutalisation of protest. According to Manuel Álvarez Tardío and Roberto Villa (2010Álvarez Tardío, M. and Villa García, R. (2010) El precio de la exclusión: La política durante la Segunda República. Madrid: Encuentro., pp. 205-209), deadly clashes with the police were usually started by radicalised groups that applied violent strategies to destroy the state. Fernando del Rey (2007Rey, F. del (2007) “Reflexiones sobre la violencia política en la II República española.” In M. Gutiérrez Sánchez and D. Palacios Cerezales, eds., Conflicto político, democracia y dictadura: Portugal y España en la década de 1930. Madrid: Centro de Estudios Políticos y Constitucionales, pp. 17-97., pp. 36-37) maintains that the security policies during the first biennium were often tolerant of left-wing actors and policemen were frequently forced to defend themselves against unprovoked aggressions. Stanley Payne (1993Payne, S. G. (1993) La primera democracia española: La Segunda República, 1931-1936. Barcelona: Paidós., pp. 403-406), from a distinctive viewpoint, maintains that, in 1936, the rulers made the mistake of not developing tougher police policies as a result of their alliance with the working-class movement. According to him, the Socialists were the primary source of violence.
The second interpretation argues that the demilitarising and professionalising police reforms enacted from 1931 to 1933 represented a qualitative breakthrough regarding the monarchist public order paradigm. Diego Palacios Cerezales (2011Palacios Cerezales, D. (2011) “Ansias de normalidad: La policía y la República.” In F. del Rey, ed., Palabras como puños: La intransigencia política en la Segunda República española. Madrid: Tecnos, pp. 596-646., pp. 598-599, 644-645) argues that the nature and procedures of the police were significantly altered. The problem was that the high level of conflict did not allow for the routinisation of the new non-lethal riot control techniques. Gerald Blaney (2012Blaney, G., Jr. (2012) “En defensa de la democracia: políticas de orden público en la España republicana, 1931-1936.” Ayer, 88, pp. 99-123. Available at: https://revistaayer.com/articulo/331 [Accessed 4 Jun. 2021]., pp. 104-105, 112-113, 118) adds that restrictive measures similar to those implemented by the Spanish Republican governments were adopted in other countries to counter extremist movements that threatened democracy. According to him, both the Civil Guard and the Assault Guard presented patterns similar to those of other European gendarmeries and riot police forces. The key factor causing state violence was the policemen’s sense of danger triggered by the rising social disorder.
Somehow or other, public order management and police action have become irreplaceable elements in the many explanations proposed to understand the violence that gripped the Republic until its last days. Therefore, the security system has been analysed addressing its different functions, organisations and agents, from disparate theoretical and hermeneutical approaches, and considering diverse territorial scales.1
The demilitarisation of the policing of protest and its correlation with the democratisation of institutions and society lie at the core of the historical debate on the police. Social scientists have defined two styles of policing protest. The “escalated force” repertoire, typical of authoritarian or predemocratic regimes, is described as brutal, repressive, diffused, illegal, reactive, confrontational and rigid. In contrast, the “negotiated management” style, developed in Western democracies since the 1960s, is recognisable for being soft, tolerant, selective, legal, preventive, consensual and flexible. The first type is normally associated with the continental security model, which is embodied by gendarmeries that are more prone to apply force and present low accountability. The second one is more related to the British system: a civil, communicative and lightly armed police, perfectly integrated into the community, whose uses of excessive force are punished when necessary. The transition between the two repertoires, however, is far from being definitive. It is a contingent process subject to cyclical setbacks, which shared a dynamic of reciprocal influence and adaptation with the institutionalisation and pacification of protest (Della Porta and Reiter, 1998Della Porta, D. and Reiter, H. (1998) “Introduction: The Policing of Protest in Western Democracies.” In: D. della Porta and H. Reiter, eds., Policing Protest: The Control of Mass Demonstrations in Western Democracies. Minneapolis: University of Minnesota Press, pp. 1-32.; Della Porta and Fillieule, 2004Della Porta, D. and Fillieule, O. (2004) “Policing Social Protest.” In: D. A. Snow, S. A. Soule and H. Kriesi, eds., The Blackwell Companion to Social Movements. Oxford: Blackwell Publishing, pp. 217-241.; Della Porta and Reiter, 2006Della Porta, D. and Reiter, H. (2006) “The Policing of Global Protest: The G8 at Genoa and its Aftermath.” In: D. della Porta, A. Peterson and H. Reiter, eds., The Policing of Transnational Protest. Aldershot: Ashgate, pp. 13-42.).
Nevertheless, the decontextualised application of these theoretical models has been called into question by specialised historians, because it may lead to a teleological, monosemic idea of “democratic police” that hinders properly comprehending societies undergoing the democratisation process (Johansen, 2017Johansen, A. (2017) “Future Trends in Historical Research on Policing: Towards Global and Interdisciplinary Perspectives.” Crime, History & Societies, 21(2), pp. 113-121. doi: 10.4000/chs.1826). The dichotomy between the British and continental systems has been discussed by Clive Emsley (1991Emsley, C. (1991) “Police Forces and Public Order in England and France During the Interwar Years.” In: C. Emsley and B. Weinberger, eds., Policing Western Europe: Politics, Professionalism, and Public Order, 1850–1940. Nueva York: Greenwood Press, pp. 159-186., pp. 160-162), who pointed out that discipline, officers and part of the rank-and-file of the Metropolitan Police came from the military. Furthermore, during the interwar period, soldiers were deployed as an auxiliary contingent during major strikes without necessarily causing a step backwards in the modernisation of policing. Anja Johansen (2005Johansen, A. (2005) Soldiers as Police: The French and Prussian Armies and the Policing of Popular Protest, 1889-1914. Aldershot: Ashgate., pp. 275-282) demonstrates how the French Republican authorities increasingly called on the military to gather enough men to implement the non-lethal, intimidating tactics of crowd control invented by Louis Lépine, prefect of the Parisian police. Diego Palacios Cerezales (2016Palacios Cerezales, D. (2016) “Policing, non-lethal weapons and the costs of repression in historical perspective.” In: M. J. Funes, ed., Regarding Tilly: Conflict, Power, and Collective Action. Lanham: University Press of America, pp. 229-248., pp. 235-236), for his part, points out that, despite not appearing in their regulations, gendarmes normally used their weapons in a dissuasive, non-lethal way (rifle butt strikes, sword attacks with the flat side, shooting into the air, cavalry charges…) before shooting into the crowd. Other researchers also explain that some paramilitary attributes of the current riot police units enable them to scatter demonstrators without firearms. According to P. A. J. Waddington (1991Waddington, P. A. J. (1991) The Strong Arm of the Law: Armed and Public Order Policing. Oxford: Oxford University Press., pp. 136-137), thanks to military discipline, officers can ensure that policemen hold a compact formation, correctly execute manoeuvres and apply force proportionately.
This article intends to rethink this very issue through an exhaustive analysis of the policing of popular protest and political violence during the Spanish Second Republic. The chronological framework begins with its foundation, on 14 April 1931, and concludes with the coup d’état of 17-18 July 1936. The spatial context is the province of Madrid. Despite of its unique character, this case study is particularly interesting because it challenges the dominant thesis on this topic. On the one hand, as the capital of Spain, every police innovation was always tested in Madrid before being extended to the whole country, because the government directly managed public order in this territory. This fact allows me to analyse the immediate effects of national police policies on a local scale. On the other hand, Madrid was the province with the second highest number of political killings in this period, which makes it an excellent scenario to dissect the relation between policing and political violence.3
The empirical material for this research is provided by a database of approximately 450 policing events. For each event, certain data points have been gathered: date, location, police forces involved, type of intervention (surveillance, negotiation, dissuasion, repression), weapons, number of dead and wounded people, arrests, protesters’ affiliation, forms of action and protest nature (prescribed, legal, transgressive, violent). The quantitative dimension of this study is limited to episodes in which there were deaths for two reasons. Firstly, the documentation of the Ministry of the Interior that has been consulted at the Archivo Histórico Nacional, the Archivo General de la Administración and the Centro Documental de la Memoria Histórica is mainly the correspondence between the minister, the civil governors and the mayors. These files have significant gaps that have been filled by looking at newspapers, police journals and specialised monographs, but the voluminous quantity of these sources has made a more selective approach necessary. Secondly, the varying duration and number of police actions per intervention have complicated the delimitation and overall count of the episodes. To minimise this problem, the rule has been to concentrate on one event the policing actions motivated by a particular conflict that happened successively on one or more days.
TOWARDS A REPUBLICAN REPERTOIRE OF COERCION
⌅From April to October 1931, with the purpose of ensuring its supporters’ hegemony on the streets, the Provisional Government placed them at the top of the order of precedence regarding the occupation of the public space. The demonstrations, strikes and meetings performed by Socialists, Radical-Socialists, left-wing Republicans and Radicals were authorised or directly facilitated. On the contrary, even though their meetings were generally allowed, the protests carried out by anarchists, syndicalists, Communists, Catholics, traditionalists and monarchists were usually prohibited or even repressed. The second element that conditioned the security policies was the rulers’ mistrust towards police forces, who were still perceived as henchmen of the overthrown dictatorship. This prejudice imposed three major guidelines on the policing of protest: discrete police presence, tolerance of pro-government voters’ riots and institutionalisation of a Republican-Socialist civic guard. On the evening of 14 April, precisely, many young militants protected the Royal Palace wearing red armbands, while policemen remained on “mandatory holiday” (Maura, 1962Maura, M. (1962) Así cayó Alfonso XIII…Barcelona: Ariel., pp. 177-178; Juliá, 1984Juliá, S. (1984) Madrid, 1931-1934: De la fiesta popular a la lucha de clases. Madrid: Siglo XXI., pp. 12-13).
This trusting approach was abandoned after certain disturbances that showed the inertia of the traditional repertoires of both police action and collective protest, as well as the paralysis of the policemen, who feared to be sanctioned if they charged against the Republican demonstrators. On 10 May, some monarchists played the “Marcha Real” and thereby provoked a protest by supporters of the Republic in Alcalá Street. The minister of the interior, Miguel Maura, attempted to calm the protesters, although it was necessary to call the Civil and Security Guards, whose men, in a sign of the times, sheathed their sabres in response to people’s whistles, and three detainees were beaten by the crowd. Then, the rioters decided to assault the newspaper ABC’s building but were intercepted by some civil guards who, after being stoned, finally shot at them, thus killing two people. In protest against these killings, thousands of demonstrators gathered in Puerta del Sol. An Athenaeum delegation delivered its complaints to Manuel Azaña, the minister of war, requiring Maura’s resignation and the disarmament of the Civil Guard. Maura gave the order to disperse the crowd, but his colleagues refused to employ any “three-cornered hat” against the pueblo and, eventually, some gunmen triggered another riot during which two men were murdered. These riots also left twelve people injured, including two policemen, and 40 people were detained (Maura, 1962Maura, M. (1962) Así cayó Alfonso XIII…Barcelona: Ariel., pp. 240-245; Azaña, 2000Azaña, M. (2000) Diarios completos: Monarquía, República, Guerra Civil. Barcelona: Crítica., pp. 433-435; Bravo Morata, 2001Bravo Morata, F. (2001) Historia de Madrid, vol. IV: De la Dictadura al Madrid de la República. San Fernando de Henares: Trigo Ediciones., IV, pp. 155-156).
The following day several extremist rioters maintained the revolutionary pulse during the disgraceful episode of the “burning of the convents.” Keeping their non-interventionist attitude, policemen and civic guards limited themselves to guaranteeing the safety of priests and nuns. Twelve religious buildings were evacuated, sacked and burned while their furniture and paintings were reduced to ashes in massive bonfires. There were no more deaths, although one person was stabbed. In order not to resort to the Civil Guard, the government called in the army, which was surprisingly well received by the arsonists. In fact, it did not need to exercise violence to control the situation. Everything concluded with a classic monarchist ritual: Captain General Queipo de Llano read a declaration of a state of war, flanked by a company of the Regiment of León and a military band, which at least played the Republican “Himno de Riego” by popular request (Pla, 2003Pla, J. (2003) Madrid: El advenimiento de la República. Madrid: El País., pp. 75-77; Gutiérrez-Ravé, 1932Gutiérrez-Ravé, J. (1932) España en 1931: Anuario. Madrid: Imp. Sáez Hermanos., pp. 151-162).4
These disturbances discredited the authorities because they involved significant and contrasting political costs as a consequence of both the disproportionate repression and the arsonists’ impunity.5
This innovative paradigm of coercion aspired to offer a more proportionate response to the proliferation of demonstrations, strikes, rallies and other forms of the modern repertoire of protest derived from the new context of political opportunities.8
These difficulties became tragically evident during the strike called by the Confederación Nacional del Trabajo (CNT) at the Telephone Company. On 6 July 1931, the anarcho-syndicalists established pickets to intimidate scabs, whose exit from the building had to be protected by the Security Guard. In the afternoon, the assault guards performed their first action: they cleared the Gran Vía by executing a non-lethal, choreographed baton charge that deserved the public’s applause. During the following weeks, demonstrations, beatings, sabotages of phone lines, explosions and attacks against shops took place, motivating an escalating number of detentions. The violence reached its boiling point on 7 August. That night, after a frightening explosion, four policemen attempted to search a vehicle in San Jerónimo Street and its occupants began a shootout in which Agent Conrado Álvarez died and one gunman was seriously injured (Juliá, 1984Juliá, S. (1984) Madrid, 1931-1934: De la fiesta popular a la lucha de clases. Madrid: Siglo XXI., pp. 198-206; Herrerín López, 2019Herrerín López, A. (2019) Camino a la anarquía: La CNT en tiempos de la Segunda República. Madrid: Siglo XXI., pp. 94-95).9
During the summer, other actors carried out transgressive mobilisations with some violence, although their consequences were much less serious. On 6 August, a Communist rally at the Wonders Theatre was followed by multiple unauthorised demonstrations. The security guards attempted to scatter one peacefully, but they were shot at, so the cavalry guards charged against the crowd. Shortly after, assault and security policemen attacked with their truncheons and swords, respectively. One of the latter was surrounded by the protesters and fired at them, seriously wounding one activist.10
RIOT POLICE ARMED AGAINST INSURGENCY
⌅Since December 1931, police reform increased its pace and scale because of the formation of a new government by Azaña with only leftist Republicans and Socialists. For two years, these leaders fostered the republicanisation, professionalisation and demilitarisation of the security system along the following axes. The first was the expansion of the assault guards to other major cities and provincial capitals, and the acquisition of war weaponry to make them able to crush anarchist uprisings without the aid of the army, which naturally challenged their raison d’être as non-lethal riot police, but did not entirely remove this original function. The second was the attempted limitation of the Civil Guard’s range of operation to the countryside and the increase in their accountability to civilian authorities, which was achieved by dismantling its General Directorate and that of the Carabineers at the Ministry of War (Palacios Cerezales, 2011Palacios Cerezales, D. (2011) “Ansias de normalidad: La policía y la República.” In F. del Rey, ed., Palabras como puños: La intransigencia política en la Segunda República española. Madrid: Tecnos, pp. 596-646., pp. 616-620; Blaney, 2007bBlaney, G., Jr. (2007b) The Civil Guard and the Spanish Second Republic, 1931-1936. PhD thesis, University of London., pp. 153-154). The last axis was the sanction of emergency legal instruments that enabled the security forces to implement a graduated response to demonstrations that did not rely upon the military, although the purpose was also to narrow the opposition’s constitutional rights of protest. These instruments were the October 1931 Law for the Defence of the Republic and the July 1933 Public Order Law.12
The first anarcho-syndicalist insurrection, which took place in January 1932, was still expeditiously repressed by the army. Hundreds of workers were arrested and deported to Villa Cisneros, in Western Sahara, but there were no fatalities (Casanova, 2010Casanova, J. (2010) De la calle al frente: El anarcosindicalismo en España. Barcelona: Crítica., pp. 102-106). In Madrid, in fact, this revolt had almost no repercussions in comparison with the next disturbance on 29 May. On that day, the police posted squads at strategic locations while soldiers patrolled some streets and stood at the barracks. In Angel Square, a sudden firefight broke out between two agents who were conducting pat downs and some suspects; one policeman and a gunman were severely wounded. Afterwards, two assault groups charged at both ends of a Communist demonstration without leaving the mandatory escape route. This compelled the participants to run away through the side streets towards the flea market, where a second shooting took place. Meanwhile, several anarcho-syndicalists provoked yet another gunfight in Magdalena Street and killed Sargent Enrique Mateos, who was not even on duty. Furthermore, in Antonio Zozaya Square, other guards accidentally fired on two civilians who were watching the spectacle from their balcony. Such a deployment of soldiers to stamp out a revolt was certainly not a new tactic. However, it must be stressed that the soldiers’ role was essentially dissuasive since there were no direct confrontations. The first line was always formed by assault men, who only abandoned their truncheons when they were shot at.13
The intrinsic politicisation of the democratising process had a deep impact on the youth, causing a powerful escalation of student protests. On 5 April 1932, 200 Carlists destroyed some posters of the Federación Universitaria Escolar (FUE) and triggered a brawl; in response, the police made 58 arrests and closed the Carlist headquarters, where they seized many clubs and pistols (González Calleja, 2009González Calleja, E. (2009) Rebelión en las aulas: Movilización y protesta estudiantil en la España contemporánea, 1865-2008. Madrid: Alianza Editorial., pp. 146-147).15
The foremost challenge for the Assault Guard and the process of police modernisation that it represented was presented by the military, the monarchist old elites and some extreme right forces. The pronunciamiento headed by the director general of the Carabineers Corps, José Sanjurjo, who had held a similar position in the Civil Guard, started on 10 August 1932. Some officers guided a squad from the Remonta Barracks to the Hippodrome, where they confiscated numerous vehicles. From here, the military rebels marched toward Cibeles Square, where they fought a bloody battle against the assault policemen. Another detachment appeared at the Ministry of War. Disobeying the orders of Arturo Menéndez, the director general of security, they refused to retreat and started a fatal shooting. Shortly after, four officers tried to take the Communications Palace, but one civil guard managed to stop them until the arrival of reinforcements, which led to a last gunfight. Defeated, the insurgents collected their fallen comrades and returned to their barracks on a sequestered bus. They had suffered ten fatalities, nine military men and one Carlist, 18 wounded and 90 arrests; in contrast, five guards, one watchman and a civilian had been injured by them (Gutiérrez-Ravé, 1932Gutiérrez-Ravé, J. (1932) España en 1931: Anuario. Madrid: Imp. Sáez Hermanos., pp. 372-375; Viqueira Hinojosa, 1993Viqueira Hinojosa, A. (1993-1999) “Historia de la Policía Española (1931-1936).” Policía, 86-140.-1999, pp. 113-119).18
The failure of this military rebellion consecrated the Assault Guard’s new role as a counter-insurgency force, since it verified their effectiveness versus enemies with war training and weaponry. Actually, the progressive government’s fear of future coups fostered a deeper militarisation of their discipline, protocols and tools. The inherent dangers of this orientation were denounced during an extended, hostile campaign against the cabinet motivated by the massacre perpetrated by assault guards in Casas Viejas during the second anarchist uprising, in January 1933. Fortunately, the outcome of this insurrection was less dramatic in Madrid, where there were no deaths despite the insurgents’ obstinate attempts to assault the main barracks of the capital. On the 8th day, the Civil Guard aborted an attack against the military railway station of Cuatro Vientos, which compelled the soldiers to fight the anarchists back. In addition, other guards had a fierce firefight with 400 anarchists, gravely injuring one rebel, and some surveillance and security policemen were involved in other shootouts by the Montaña and María Cristina Barracks (González Calleja, 2018González Calleja, E. (2018) “¿Qué fue de ‘la ciudad alegre y confiada’?: La violencia política en Madrid durante la Segunda República.” In: J. A. Martínez Martín and L. E. Otero Carvajal, eds., La sociedad urbana en el Madrid contemporáneo. Madrid: Los Libros de la Catarata, pp. 142-175., p. 147).19
Once again, the next anarchist disturbance on 8 May produced much bloodier results in the capital. During a 48-hour general strike, numerous affiliates of the CNT were arrested mostly for coercing other workers and planting explosives in railroads and electrical installations. However, some of them also shot a truck and threw a bomb into a crowd, injuring two civilians. In the working-class neighbourhood of Cuatro Caminos, the assault men charged at the strikers and ended up firing into the air to disperse them; in Francos Rodríguez Street, the anarchists shot someone else. As a preventive measure, the police had been equipped with short-barreled rifles, but their deployment was calibrated to the violence applied by the strikers; as a rule, they used their truncheons in front of unarmed workers. Nonetheless, in Manuel Becerra Square, some policemen attempted to frisk certain strikers, and a young woman dropped a bomb that instantly killed Agent Francisco Juarros and triggered a gunfight. As a consequence, two gunmen passed away, four more were wounded and four policemen were also injured (Viqueira Hinojosa, 1993Viqueira Hinojosa, A. (1993-1999) “Historia de la Policía Española (1931-1936).” Policía, 86-140.-1999, pp. 169-170; Hernández Quero and Cruz Salanova, 2019Hernández Quero, C. and Cruz Salanova, L. de la (2019) “Cuando los suburbios dejaron de ser periferia: la lucha por el control de la calle en el Madrid de los años 1930.” Rubrica Contemporanea, 8(16), pp. 67-86. doi: 10.5565/rev/rubrica.178, p. 71).20
KEEPING THE FIREARMS DRAWN AND LOADED
⌅Since autumn of 1933, the coalition governments controlled by the Radical Party, a Republican centre-right formation, and pressured by the Catholic parliamentarians of the Confederación Española de Derechas Autónomas (CEDA), promoted a gradual rollback of the policies of police modernisation implemented up to that point. Determined to re-establish law and order, these authorities developed a counter-reformist programme to reorganise the administration of public order in a militarising, centralising and heavy-handed direction. The renamed General Inspectorate of the Civil Guard recovered its organisational autonomy with respect to the political authorities because the recently founded Special Section and the also new Technical Secretariat at the Ministry of the Interior were neutralised (Blaney, 2007aBlaney, G., Jr. (2007a) “Keeping Order in Republican Spain, 1931-1936.” In: G. Blaney, Jr., ed., Policing Interwar Europe: Continuity, Change and Crisis, 1918–40. Basingstoke: Palgrave Macmillan, pp. 31-68., pp. 49-50). Most importantly, the Assault Guard experienced intense structural and operational militarisation. Contradicting their restrained philosophy on firearms, Lieutenant-Colonel Muñoz Grandes ordered the policemen to carry their weapons drawn and loaded even inside their trucks to be prepared to repel any possible attack (Palacios Cerezales, 2011Palacios Cerezales, D. (2011) “Ansias de normalidad: La policía y la República.” In F. del Rey, ed., Palabras como puños: La intransigencia política en la Segunda República española. Madrid: Tecnos, pp. 596-646., 627-628). In addition, from now on constitutional liberties would be almost permanently suspended as a result of the repeated proclamations of the states of exception as authorised by the Public Order Law (Ballbé, 1985Ballbé, M. (1985) Orden público y militarismo en la España constitucional (1812-1983). Madrid: Alianza Editorial., p. 363).
Likewise, these governments applied different tolerance standards regarding the occupation of the street to favour their supporters and punish their rivals. On the one hand, they encouraged and protected the political events and mobilisations of Radicals, Catholics, Agrarians and other conservative groups, while Falangists were still pursued but benefited from the complicity of certain police and judicial sectors. On the other hand, the growing demonstrations and strikes of anarchists, syndicalists, Communists and Socialists were always banned and frequently supressed. At the same time, police agencies had to operate in a much more hazardous environment due to the labour movement’s strategic radicalisation and the hostile counteroffensive of the Falange.
The combination of these variables began a long-term escalation of violence, whose first victims in Madrid were not caused by police militarisation itself, but by union rivalries. In October 1933, the CNT called a massive construction strike that became the first industrial general strike in those years. It was particularly dangerous for the anarcho-syndicalists’ adversaries of the Socialist Unión General de Trabajadores (UGT). From the 23rd to the 25th, the strikers shot some scabs and exchanged gunfire with the police, wounding five civilians, while the assault guards charged violently in Cuatro Caminos and injured one striker in the head. The movement ended with more than 120 detentions, but the only deaths were caused by the anarchists, who shot dead two workers and fatally stabbed another (Juliá, 1984Juliá, S. (1984) Madrid, 1931-1934: De la fiesta popular a la lucha de clases. Madrid: Siglo XXI., pp. 232-257; Souto Kustrín, 2004bSouto Kustrín, S. (2004b) “Y ¿Madrid? ¿Qué hace Madrid?”: Movimiento revolucionario y acción colectiva (1933-1936). Madrid: Siglo XXI., p. 102; González Calleja, 2015González Calleja, E. (2015) Cifras cruentas: Las víctimas mortales de la violencia sociopolítica en la Segunda República española (1931-1936). Granada: Comares., p. 348).21
The intensification of ideological maximalism and the unbridled proliferation of armed skirmishes between antagonistic groups further poisoned the university atmosphere. Nonetheless, the delegitimation of the government that resulted from the repression of students caused the restoration of a tougher policing style to be less intense on university campuses. On 24 October at the Faculty of Health, some students occupied the street, sabotaged the tramways and threw objects out of the windows. The assault guards charged with batons, cordoned off the street and inexplicably fired into the building, which compelled Muñoz Grandes to command them personally to sheathe their guns.22
The triumph of the centre-right parties in the general election on 19 November 1933 was answered by anarchists and syndicalists through a third uprising on 8 December.23
The next year the incessant confrontations and struggles between the radicalised youth organisations of certain parties became a major catalyser of the political violence. Most of these deadly conflicts were triggered by sellers of newspapers, as were the skirmishes on 11 January and 9 February in which the Falangist students Francisco de Paula Sampol and Matías Montero were killed.25
The government’s attitude towards the meeting of the youth branch of Acción Popular on 22 April in El Escorial demonstrated a new determination to protect Catholic mobilisations from the Socialists and Communists’ counter-protests, who called the first political strike of that period. A few days before, the strikers fired on Acción Popular’s headquarters and killed one militant; in another shooting, the police killed one Communist. On the 22th day, an impressive security deployment was established in El Escorial with the assistance of the young Catholic militants, who controlled the entrances wearing red armbands. In Madrid, investigation agents lit the streetlights while security guards delivered bread supplied by the soldiers, who were again performing policing tasks. In Puerta del Sol, some assault guards that had been previously attacked shot the strikers, killing one of them. Another protester passed away because of the explosion of a bomb he probably planted himself. Lastly, in Puente de Vallecas and the Pacífico neighbourhood, many protesters stoned the vehicles that returned from El Escorial and exchanged gunfire with various right-wing militants and some guards (Souto Kustrín, 2004bSouto Kustrín, S. (2004b) “Y ¿Madrid? ¿Qué hace Madrid?”: Movimiento revolucionario y acción colectiva (1933-1936). Madrid: Siglo XXI., pp. 133-139; González Calleja, 2015González Calleja, E. (2015) Cifras cruentas: Las víctimas mortales de la violencia sociopolítica en la Segunda República española (1931-1936). Granada: Comares., p. 358).28
This episode reveals some important changes regarding protest control. On the one hand, there was the already mentioned abandonment of the Republican-Socialist governments’ policy on Catholic rallies, which cancelled them when some leftist groups decided to sabotage them instead of sending police protection (Álvarez Tardío and Villa García, 2010Álvarez Tardío, M. and Villa García, R. (2010) El precio de la exclusión: La política durante la Segunda República. Madrid: Encuentro., p. 180). On the other hand, there were a greater disposition to shoot by policemen, who caused most of the 27 wounded in the previous incident, and a higher governmental resoluteness concerning the maintenance of essential services during strikes. Nevertheless, the most remarkable novelty was that Socialists were now being repressed as well, which logically fuelled the brutalisation of their characteristic repertoire of action.
In the countryside, the militarisation of the coercive approach promoted the revival of traditional, more aggressive mobilisation practices (thefts, arsons, sabotages, hunger riots…).29
Despite the dominant counter-reformist trend, police officers kept testing technologically more sophisticated crowd control resources. On the night of 21-22 July 1934, during a strike in the pits of the Bank of Spain, assault guards launched tear gas grenades after an unsuccessful negotiation. Consequently, the workers fled the building with symptoms of asphyxiation and one of them suffered a cardiopulmonary arrest. This was the second occasion that the police had deployed this device and, though their inexperience might have proved fatal, there were no further consequences.32
POLICEMEN FIGHTING IN A STATE OF WAR
⌅From September to October 1934, the convergence between the Socialist militias’ actions and the counterattack unleashed by the state raised the violence to its highest point up to that time. The leftists’ first test was presented on 8 September, when Socialists and Communists called another general strike to disrupt the scheduled assembly of the Instituto Agrícola Catalán de San Isidro at the Monumental Cinema. The authorities made a new effort to guarantee primary services: soldiers and guards distributed bread, drove tramways and coaches, and protected the young Catholics that sold the newspaper El Debate. Aside from the usual sackings, threats and physical attacks, there was a sharp increase in the shootouts with the policemen, which resulted in a security officer being severely injured. However, all the deaths were caused by the state agencies. In Bravo Murillo Street, an assault policeman shot a man dead; in Santa Isabel Street, the strikers fired on some guards and their reaction left three dead people, including a woman hit by a stray bullet; in Atocha Street, another policeman took down a shooter; and in Ángel Square, civil guards killed a striker just for throwing some stones at them. Additionally, in Carabanchel Bajo, another young protester that was sabotaging the tram lines died by gunfire. These bloody results (seven deaths, 40 wounded and 200 detentions) and the well-founded suspicions of brutal behaviour by the assault guards revealed an authoritarian leap in public order policing, as well as a growing use of weapons by the Socialists (Souto Kustrín, 2004bSouto Kustrín, S. (2004b) “Y ¿Madrid? ¿Qué hace Madrid?”: Movimiento revolucionario y acción colectiva (1933-1936). Madrid: Siglo XXI., pp. 147-162).34
On 4 October, as promised, the Socialist leaders called for the violent seizure of power in response to the appointment to the government of three ministers from the CEDA. Despite its intended national scope, the rebellion only broke out in a few provinces in northern Spain, mainly in Asturias and Catalonia, and was mercilessly “pacified” with brutal violence by the army, the Civil Guard, the Assault Guard and the colonial troops. The rebel militias’ ferocious assaults, the ruthless state counteroffensive and the repression of the working-class population resulted in a blood bath with more than 1,300 deaths.35
Madrid was the third major location of this armed movement. During the simultaneous general strike, the longest in the city’s history, soldiers and assault and municipal policemen took charge of the basic services (bakeries, public lighting, tramways, the slaughterhouse...) with the enthusiastic support of the right-wing parties’ young affiliates, who performed auxiliary tasks. The strikers were arrested for distributing propaganda, gathering clandestinely, damaging means of transport and threatening other employees. The armed insurrection was exclusively carried out by the Socialist militiamen, who were involved in innumerable shootouts with the state forces and attacked several barracks, police stations and institutional buildings. The first night, during a massive gunfight in the Prosperidad neighbourhood, an assault guard and a militiaman died, and another insurgent passed away during a failed assault on the Moret Barracks. The next day, in Atocha, some assault policemen backed by soldiers got involved in another firefight and killed one rebel, while the civil guards killed another rebel while repelling an attack on the Montaña Barracks. On Saturday the 6th, during a confrontation next to the Negro Hotel, three revolutionaries were killed; furthermore, two more civilians were killed by a Civil Guard’s stampede from the Guzmán el Bueno Barracks (Souto Kustrín, 2004bSouto Kustrín, S. (2004b) “Y ¿Madrid? ¿Qué hace Madrid?”: Movimiento revolucionario y acción colectiva (1933-1936). Madrid: Siglo XXI., pp. 238-265; González Calleja, 2018González Calleja, E. (2018) “¿Qué fue de ‘la ciudad alegre y confiada’?: La violencia política en Madrid durante la Segunda República.” In: J. A. Martínez Martín and L. E. Otero Carvajal, eds., La sociedad urbana en el Madrid contemporáneo. Madrid: Los Libros de la Catarata, pp. 142-175., pp. 154-156).
From the 7th onwards, the coordinated assaults were replaced with isolated violent actions: bomb-throwing, shooting from rooftops—the so-called paqueo—and sporadic gunfire. That night, in Puerta del Sol, General Cabanellas read a declaration of a state of war escorted by two infantry sections. The next morning military planes flew over Madrid in search of snipers and large reflectors were installed at the top of emblematic buildings such as the Telephone Company’s headquarters to blind them. Simultaneously, in the surrounding towns, there were bloody incidents as well, despite the majority of strikers were behaving peacefully because they had orders to do so. On Friday the 5th, in Carabanchel Bajo, an exchange of fire left one young Socialist dead and many others wounded. The following day, in Colmenar Viejo, after refusing to negotiate, the Civil Guard fired at hundreds of demonstrators who had stoned and fired on the town hall, causing five deaths. Lastly, on Wednesday the 10th, in a disastrous attempt to take the Carabanchel military base, four insurgents were killed. The results could not be more dramatic: 44 people dead (38 civilians, six policemen and soldiers), 50 injured and almost 2,000 detainees (Souto Kustrín, 2004bSouto Kustrín, S. (2004b) “Y ¿Madrid? ¿Qué hace Madrid?”: Movimiento revolucionario y acción colectiva (1933-1936). Madrid: Siglo XXI., pp. 254, 270-278, 282-283; González Calleja, 2015González Calleja, E. (2015) Cifras cruentas: Las víctimas mortales de la violencia sociopolítica en la Segunda República española (1931-1936). Granada: Comares., pp. 373-375).36
After the revolution, the government decided to re-establish state control of the streets by applying unprecedented strong-arm policies that brought the militarisation of the public order system to its peak under the umbrella of martial law. During the uprising, the authorities had already militarised the municipal guards and authorised soldiers and retired officers’ cooperation. Soon after, Muñoz Grandes ordered the assault guards to use Mauser rifles on their deployments and parades. Since December, the minister of the interior, Eloy Vaquero, sponsored two projects that restructured both the Investigation and Surveillance Corps and the Security and Assault Guard with an openly militaristic aim, although the pressure of the police’s civilian branch obstructed their ratification (Palacios Cerezales, 2011Palacios Cerezales, D. (2011) “Ansias de normalidad: La policía y la República.” In F. del Rey, ed., Palabras como puños: La intransigencia política en la Segunda República española. Madrid: Tecnos, pp. 596-646., pp. 633-634; Vaquero Martínez, 2017Vaquero Martínez, S. (2017) “Entre la republicanización y la militarización: Las transformaciones de las fuerzas policiales en la Segunda República española, 1931-1936.” Ler História, 70, pp. 79-92. doi: 10.4000/lerhistoria.2778, pp. 87-88).37
This militarised strengthening enlarged the costs of protest to such an extent that popular mobilisation was reduced to a minimum, which is why police work focused on the search for weapons depots and on the investigation of social killings. These murders were generally workers’ revenge for the numerous firings employers made to punish those employees who had gone on strike in October.39
From the beginning of 1936, Madrid’s political life and associative participation recovered their natural effervescence as a consequence of the calling of general elections on 16 February. The electoral campaign was disrupted by a violent atmosphere of discursive polarisation, labour confrontation and political murders. During this campaign at least five people died in individual attacks or violent brawls: two Socialists, two Falangists and one anarcho-syndicalist (Álvarez Tardío, 2013Álvarez Tardío, M. (2013) “The Impact of Political Violence During the Spanish General Election of 1936.” Journal of Contemporary History, 48 (3), pp. 463-485. doi: 10.1177/0022009413481823, p. 471; González Calleja, 2015González Calleja, E. (2015) Cifras cruentas: Las víctimas mortales de la violencia sociopolítica en la Segunda República española (1931-1936). Granada: Comares., pp. 392-395).42
THE EPHEMERAL REVIVAL OF THE TRUNCHEON
⌅The Popular Front coalition’s victory enabled Azaña to construct a new cabinet exclusively composed of centre-left Republicans but reliant on the Socialist and Communist parliamentary groups, who constantly lobbied for the application of tougher measures against fascism. Right after the amnesty of the October prisoners, the modernising police agenda of the first biennium was re-established by developing its three core axes: demilitarisation of the public order apparatus, decentralisation of security management in Catalonia and republicanisation of the police personnel through the invention of the “mandatory leave” status, which was conceived to neutralise possible collaborators with coup-plotting officers (Blaney, 2007bBlaney, G., Jr. (2007b) The Civil Guard and the Spanish Second Republic, 1931-1936. PhD thesis, University of London., pp. 233-236; Vaquero Martínez, 2019Vaquero Martínez, S. (2019) “La autoridad, el pánico y la beligerancia: Políticas de orden público y violencia política en la España del Frente Popular.” Historia y Política, 41, pp. 63-92. doi: 10.18042/hp.41.03, pp. 69-73).43
From February to July 1936, due to this reformist policy, the victims of state repression decreased drastically but the space resulting from police absence was occupied by extremist armed elements, whose attacks and clashes generated a chronic escalation in the number of political murders. On 10 March, in Alberto Aguilera Street, some Communists shot dead the Falangist students Juan José Olano and Enrique Belsoleil after attempting to pat them down.45
The “discretet instruction” transmitted to the security forces by the authorities and the radicalisation of protest caused an extraordinary revival of the anticlerical riots (Muñiz, 2009Muñiz, A. (2009) Días de horca y cuchillo: Diario 16 de febrero-15 de julio de 1936. Sevilla: Espuela de Plata., p. 71). On 13 March, the funeral of Jesús Gisbert, the agent murdered in the aforementioned attack on Jiménez de Asúa, led to an antifascist demonstration that finished with the burning of two churches and the newspaper La Nación’s office, during which two firemen died and a security guard was killed (Ruiz, 2012Ruiz, J. (2012) El terror rojo: Madrid, 1936. Madrid: Espasa., pp. 39-40).47
The Popular Front stage differed from earlier phases in that contentious politics took place in the context of a new scenario characterised by the convergence of four interrelated phenomena: the politicisation of funerals, far-right mobilisation, military insubordination and shootings triggered by agents provocateurs. These factors made their appearance together on 16 April 1936 during the service of the Civil Guard officer Anastasio de los Reyes. He had been murdered two days earlier, during a parade for the anniversary of the Republic. The perpetrators were left-wing militants, who also wounded four civilians and three civil guards, one of whom would die. Several officers who had stolen the deceased’s body transformed his funeral into a dangerous demonstration against the government.49
During the following month, the new minister of the interior, Santiago Casares Quiroga, intended to reinstate public order by intensifying the pursuit of Falangists and military plotters, controlling labour mobilisations with greater firmness and issuing instructions against some workers’ guards that usurped the police’s tasks in certain towns (Vaquero Martínez, 2019Vaquero Martínez, S. (2019) “La autoridad, el pánico y la beligerancia: Políticas de orden público y violencia política en la España del Frente Popular.” Historia y Política, 41, pp. 63-92. doi: 10.18042/hp.41.03, pp. 82-84). The major reason for this alternative policing strategy was the belligerent cycle of strikes promoted by the CNT since the end of May, which interrupted the activity of vital sectors of the city economy (gastronomy, textile, wood, construction…). Once more, tens of thousands of anarcho-syndicalists organised massive assemblies, planted explosives, attacked scabs and shot their antagonists from the Socialist UGT, whose directors were more willing to accept the government’s mediation (Juliá, 1991Juliá, S. (1991) “¿Feudo de la UGT o capital confederal?: La última huelga de la construcción en el Madrid de la República.” Historia Contemporánea, 6, pp. 207-220. doi: 10.1387/hc.19499, pp. 210-213; Sánchez Pérez, 1991Sánchez Pérez, F. (1991) “Clase obrera y conflictividad social en el Madrid del Frente Popular (febrero-julio de 1936).” Cuadernos de Historia Contemporánea, 13, pp. 47-71. Available at: https://revistas.ucm.es/index.php/CHCO/article/view/CHCO9191110047A [Accessed 4 Jun. 2021]., pp. 64-69; 2013Sánchez Pérez, F. (2013) “Las huelgas del 36: ¿por qué Madrid?” Bulletin d’Histoire Contemporaine de l’Espagne, 48, pp. 27-42., pp. 41-42). On 15 June, the assault force dispersed several female textile employees who were harassing their co-workers and performed 80 arrests.52
This level of disorder and violence delegitimised the public order policies and fostered the politicisation of the police forces, which emerged as the foremost challenge to the government before the coup d’état. On the evening of 12 July, enraged by the killing of Lieutenant Castillo, also a former instructor of the Socialist militias, several assault officers started a turbulent mutiny at the Pontejos Barracks. One of the several groups of guards that left to arrest illegally certain right-wing politicians targeted as “fascists” kidnapped the authoritarian monarchist leader José Calvo Sotelo, and Luis Cuenca, one of the Socialist gunmen that accompanied the guards, shot him twice in the nape of the neck. His funeral, celebrated two days later, was attended by the chiefs of all rightist parties and became another massive protest against the government. In order not to aggravate the audience, the Civil Guard monitored the Almudena Cemetery while the Assault Guard watched the road to the capital. As was expected, the attendees started a demonstration and the policemen managed to scatter them near Madrid’s bullring without violence. However, the protesters regathered in Manuel Becerra Square and fought some workers who responded to their lifted arms by raising their fists. In Alcalá Street, finally, unknown agents provocateurs took some shots from a vehicle and the assault guards fired on the crowd, killing three demonstrators and wounding other two (Gibson, 1982Gibson, I. (1982) La noche en que mataron a Calvo Sotelo. Barcelona: Argos Vergara., pp. 86-214; González Calleja, 2015González Calleja, E. (2015) Cifras cruentas: Las víctimas mortales de la violencia sociopolítica en la Segunda República española (1931-1936). Granada: Comares., p. 423).55
THE QUANTITATIVE DIMENSION OF POLICE VIOLENCE
⌅The quantitative analysis of the sociopolitical killings perpetrated in the province of Madrid allows to draw interesting conclusions on the policing of protest in the Republican period (Table 1). Far from provoking the majority of the 185 overall casualties, the state forces were responsible for 63 deaths (34.05%).56
Source: Database compiled by the author.
Contrary to the claim made by the predominant interpretation of protest control in Republican Spain, most of the lethal policing interventions did not happen during peaceful actions (Table 2).59
Source: Database compiled by the author.
It should be stressed that this analytical picture is exclusive to Madrid and perhaps other urban areas. In rural Spain, which represented the majority of the national territory, both the persistence of military policing protocols and the coercive agencies’ responsibility in the escalation of hostilities were clearly dominant. This clarification is important because most of the murders happened in small and medium-sized towns. However, the reason was not that mortality was higher in the countryside but that rural provinces were higher in number. In fact, the deaths per person were generally more elevated in urban provinces (González Calleja, 2015González Calleja, E. (2015) Cifras cruentas: Las víctimas mortales de la violencia sociopolítica en la Segunda República española (1931-1936). Granada: Comares., pp. 110-111). This could be explained by certain factors that were specific to larger cities such as Madrid, which promoted a more extensive application of violent strategies among society (higher concentration of sociopolitical associations, bigger competition to control the public space, greater access to firearms…).
On the authorship of the murders (Table 3), there is a notorious difference between the 72 deaths provoked by left-wing elements and the 24 killings executed by rightist actors.60
Source: Database compiled by the author.
CONCLUSION
⌅The evolution of the policing of protest in Madrid during the Republican years demonstrates the coexistence of two styles: a traditional, military and deadly repertoire that required the utilisation of sabres, rifles and machine guns; and a modern, civil and non-lethal paradigm characterised by the deployment of batons, tear gas and water tanks. This archetypical dichotomy, however, should not be accepted without nuances. On one hand, the brutal repertoire was not only restored. In fact, the counter-revolutionary reaction added to it a much greater destructive capacity that increased its lethality to unprecedented levels. On the other hand, applying a normative approach, the softer style cannot be categorised as entirely “democratic” because the right to protest was not effectively guaranteed, especially when it was exercised by the opposition. Nevertheless, this must not lead one to underestimate the breakthrough that, from a historically contextualised perspective, this non-lethal repertoire represented compared to the monarchist period. Additionally, it must be remembered that the Second Republic was a new democracy born in the polarising interwar years, when not even the most advanced democracies met today’s parameters of “democratic policing.”
Therefore, the relation between the policing of protest and the escalation of political violence is only understandable by recognising the alternation between both repertoires and their respective operative problems and outcomes. The implementation of a proportionate and more tolerant style by the left-wing Republicans reduced the number of casualties caused by state actors, although the persistent repression of peaceful mobilisations against the government led some protesters to develop more aggressive methods. On other occasions, the absence of police offered the extremist sectors a space to exercise violence with a certain impunity. Likewise, although the right-wing governments’ purpose was to contain the popular protests through a dissuasive, frightening exhibition of force that made it unnecessary to repress them, the re-emergence of military manoeuvres and firearms, instead of re-establishing law and order and decreasing the number of victims, dramatically increased the killings by the state and fostered the strategic brutalisation of both working-class organisations and far-right parties, which elevated the rate of deadly violence to an unprecedented level.
In conclusion, the deficiencies, dysfunctions and collateral effects of the policing of protest were caused by both the restoration of a heavy-handed style and, more indirectly, the development of a non-lethal repertoire. The inconsistent implementation of each one according to the protesters’ political affiliation generated rising violence that obstructed the transition to a softer policing paradigm, a process that was definitely interrupted by the military uprising. In the end, the Republic was unable to accomplish both the protection of the citizens’ right to protest and the maintenance of control over the streets, something necessary in every process of police democratisation, and this failure was utilised by the Republic’s most dangerous enemy to destroy it.