Choosing from the repertoire of contention: Evidence from student protests in Latin America

Eligiendo del repertorio de contención: Evidencia de protestas estudiantiles en América Latina

ABSTRACT

Introduction:

Activists mobilize using various protest strategies to advance their demands but few studies analyze protest events to explain their choice of peaceful, disruptive, and violent methods. This paper argues that when movements have few supporters and allies - major resources for mobilization - they privilege violent tactics. Conversely, when mobilizations have more participants and partners, protests are more likely to use peaceful or disruptive strategies.

Materials and methods:

The paper uses a mixed methods research design. From a quantitative perspective, it uses a dataset of 4,700 college student protests in Latin America to test the theoretical claims with logistic regressions. A case study of protest tactics and resources during the December 2014-January 2015 youth protests in Lima, Peru is used to illustrate how human resources shape the repertoire of contention. This qualitative section uses evidence from interviews as well as secondary data.

Results:

The results of the regression analyses partially support the theory, showing that more participants tend to make protests more peaceful and disruptive and less violent. The effect of certain allies, however, goes against theoretical expectations. As the case study shows, after a relatively violent first march, young activists in Lima organized themselves on a territorial basis, which both decreased the influence of political parties in the movement and increased the number of activists and actors involved, making subsequent demonstrations less violent. However, the case study also shows that the presence of specific political actors is inherently associated with more violent strategies.

Discussion:

This paper shows how the presence of allies and the number of demonstrators, as major resources available to social movements, shape the choice of tactics used in protest events. While more participants and the presence of allies makes events more contained, the evidence also suggests that both very large numbers of participants and the addition of disparate actors into a mobilization may promote violence.

KEYWORDS:
Protest events; tactics; nonviolent movements; violent protests; repertoire of contention

RESUMEN

Introducción:

Los activistas se movilizan utilizando distintas estrategias de protesta para avanzar sus demandas, pero pocos estudios analizan los eventos de protesta para explicar su elección de métodos pacíficos, disruptivos o violentos. Este artículo argumenta que cuando los movimientos tienen pocos partidarios y aliados - principales recursos para la movilización - éstos privilegian tácticas violentas. Por otra parte, cuando las movilizaciones tienen más participantes y colaboradores, la probabilidad de que las protestas usen estrategias pacíficas o disruptivas aumenta.

Materiales y métodos:

Este artículo usa un diseño de investigación de métodos mixtos. Desde una perspectiva cuantitativa, se utiliza una base de datos de 4.700 eventos de protesta de estudiantes de educación superior en América Latina para evaluar los argumentos teóricos a través de regresiones logísticas. Se realiza un estudio de caso sobre las tácticas de protesta y recursos durante las movilizaciones de jóvenes en Lima, Perú de diciembre de 2014-enero de 2016 para ilustrar cómo los recursos humanos moldean el repertorio de contención. Esta sección cualitativa usa evidencia de entrevistas así como de fuentes secundarias.

Resultados:

Los resultados de los análisis de regresión apoyan parcialmente la teoría, demostrando que más participantes tienden a volver las protestas más pacíficas y disruptivas y menos violentas. El efecto de ciertos aliados, sin embargo, va en contra de las expectativas teóricas. Como muestra el estudio de caso, luego de una primera marcha relativamente violenta, los jóvenes activistas limeños se organizaron territorialmente, disminuyendo la influencia de los partidos políticos en el movimiento a la vez que aumentaron la cantidad de activistas y actores involucrados, volviendo a las manifestaciones posteriores menos violentas. El estudio de caso, sin embargo, también evidencia que la presencia de ciertos actores políticos está inherentemente asociada con estrategias más violentas.

Discusión:

Este artículo demuestra que la presencia de aliados y el número de participantes, como principales recursos disponibles para los movimientos sociales, moldean la elección de las tácticas usadas en los eventos de protesta. Aunque más participantes y la presencia de aliados vuelve a los eventos más contenidos, la evidencia también sugiere que una gran cantidad de participantes y la suma de actores dispares en una movilización pueden promover la violencia.

PALABRAS CLAVE:
Eventos de protesta; tácticas; movimientos no violentos; protestas violentas; repertorio de contención

I. Introduction¹

Between December 2014 and January 2015, students and young people in Lima organized five mass demonstrations against a law enacted on December 11, which limited social security benefits for young workers. On January 25 - the same day of the last demonstration - the Peruvian Congress repealed the “Ley Pulpín,” as it came to be known.² These mobilizations showed very high levels of organization,

The fives marches against the [Ley Pulpín] have symbolized a learning in citizens’ mobilization, and the emergence of young collectives that did not exist before the law. Demonstrators came from more than 30 districts in Lima and were divided into 12 zones, with people in charge of order and communications; in the fifth march there were even delegates in charge of first aid. Labor unions from both private and public companies also participated, and students from at least ten private and public universities and colleges in Lima took part even though the summer break had started in December (Fowks, 2015, author’s translation).

The five protests occurred in a short period of time but only two (the first and the fourth) experienced violent clashes. What factors determine the types of strategies used by social movements?

This study seeks to explain why movements choose peaceful, disruptive, and violent tactics out of their repertoire of contention.³ The main argument is that, as protests gather more participants and attract more allies, they tend to privilege peaceful and disruptive strategies; conversely, when mobilizations are smaller and have fewer allies, participants resort to violent tactics. This choice is not a minor issue, as social movement outcomes hinge to a great extent on tactical decisions (Barkan, 1979; Chenoweth & Stephan, 2013; Maney, 2012).

In recent years, the tactics of protests have received increased attention from scholars (Taylor and Van Dyke, 2007). Students of social movements have, for example, stressed the distinction between peaceful and transgressive actions (McAdam, 1983; McAdam et al., 2001; Van Dyke et al., 2004), or between violent and confrontational (Della Porta, 2013) strategies (Koopmans, 1993; Pearlman, 2012), while others have delved into the endogenous relationship between state repression and violent tactics (Della Porta, 2006; Koopmans, 1997). Few studies however, have considered all types of tactics and systematically assessed why movements choose one type of strategy over others. An important exception is Medel & Somma (2016), who analyze the determinants of restrained and transgressive tactics in Chile. However, since they use multinomial logistic regressions, they treat tactics as mutually exclusive, when in reality two or more kinds of strategies can be displayed within the same protest event.⁴

This study has four sections, after this introduction. The second section presents the argument and hypotheses, which emphasize allies and increased numbers as resources that define whether movements privilege peaceful, disruptive, or violent tactics. The third section uses a dataset of more than 4,700 college student protests in eighteen Latin American countries to test these claims using logistic regressions, showing that increased size and the presence of some social groups increase the likelihood of peaceful and disruptive tactics, while fewer demonstrators and the absence of certain actors are associated with violent strategies. The fourth section presents a case study of tactics, protest size, and allies during the December 2014 and January 2015 youth marches against the so-called “Pulpín” Law in Peru, Lima.⁵ The final section summarizes the results and presents the study’s implications.

II. Choosing protest tactics: numbers, allies and resources

The main argument of this study is that the amount of resources social movements possess plays a major role in defining its repertoire of contention. Resources as a driving force of protests have a longstanding pedigree in the study of social movements (McCarthy & Zald, 1977). Social movements can use five types of resources: “moral, cultural, social-organizational, human, and material resources” (Edwards & McCarthy, 2004, p. 117).

This research focuses on the effects of social-organizational and human resources on protest tactics. The former can include intentional resources (created specifically to advance the movements’ goals) and appropriable ones (which already exist, but must be seized somehow) (Coleman, 2000). Meanwhile, human resources refer to the “labor, experience, skills, and expertise” (Edwards & McCarthy, 2004, p. 127) that activists can provide.

The amount of resources, in turn, determines whether mobilizations follow what Della Porta & Diani (2006, pp.171-176) call the “logic of numbers” and the logic of damages” in social movements. The former logic argues that social movements have a greater social and political impact when their protests are larger. As Somma & Medel (2019, p. 233) claim,

Large demonstrations are more likely to be covered by the mass media, therefore getting more attention to movement demands. As more people support the movement, bystanders may be more inclined to think it is worthwhile. Activists marching by thousands feel empowered and reinforce their emotional connection to the movement. Organizers believe they did a good job and prepare optimistically for the next assault.

In a similar vein, protest size can indicate authorities how much support the movement has (DeNardo, 1985, p. 36). Strategies that embody the logic of numbers are those that allow movements to boast their numerical superiority: marches, demonstrations, petitions, strikes, and sit-ins, among others.

The logic of damages, by contrast, maintains that public opinion, the media, and politicians may pay more attention to movements when they are more violent. Indeed, the government may exhibit “additional responsiveness to violent protest”, which may “provide sufficient compensation for the movement’s smaller size” (DeNardo, 1985, p. 219). Thus, while not completely mutually exclusive, these two logics tend to operate separately, based on, for instance, the number of protest participants and the presence of allies.

Although social movement organizations may generate their own resources, the incorporation of additional assets is critical in determining their actions. As Edwards & McCarthy (2004, p. 116) claim, “[w]hen movement activists do attempt to create collective action (fielding protests, creating social movement organizations, and the like) through historical time and across geographical locations their successes are consistently related to the greater presence of available resources in their broader environments.” What are the resources that tip the balance towards peaceful, disruptive or violent strategies?

A major organizational and human resource is people themselves. In fact, activists tend to prioritize maximizing the number of participants over the number of protests (Popovic & Miller, 2015, p. 52; Tilly, 2015, p. 370). Protest size is clearly associated with the repertoire of contention, as “the tactics that activists decide to use in the context of a massive event will not have the same effectiveness as when the protest only summons a handful of people” (Medel & Somma, 2016, pp.174-175). As the logic of numbers suggests, numbers by themselves may draw media attention, influence public opinion, and eventually shape political and policy outcomes, making violence unnecessary to achieve social movements’ objectives.

The argument that increased protest size leads to peaceful demonstrators is not without its detractors, however. Some authors argue that increased numbers reduce the individual risks of repression, making violent actions less costly (Granovetter, 1978; Oberschall, 2017; Taylor & Van Dyke, 2007). More participants, by piquing the interest of the state, may also lead to more repression, which may cause activists to react and violence to escalate (Della Porta, 2006; Koopmans, 1997). In the case of self-determination groups, Cunningham (2013) finds that relatively larger groups are less likely to use non-violence and more likely to engage in civil war, vis-à-vis conventional politics. It is rarely the case, however, that most participants in a large protest event take part in violent actions - in general, aggression is a property of small groups.

• • Hypothesis 1a: increased numbers of protesters have a positive effect on the likelihood of peaceful tactics.

• • Hypothesis 1b: increased numbers of protesters have a positive effect on the likelihood of disruptive tactics.

• • Hypothesis 1c: increased numbers of protesters have a negative effect on the likelihood of violent tactics.

There are only so many resources that members from the same social movement organization or sector can harness by themselves and from their bases. Indeed, the involvement of potential allies from other organizations and sectors is a key dimension of resource availability. The evidence also suggests that movements in which different types of activists cooperate tend to be more successful (Gamson, 1990; Steedly & Foley, 1979). As Van Dyke (2003, pp.226-227) explains, “[c]reating broad coalitions that cut across movement boundaries is a central means by which movements are able to stage events with a very large number of participants.” Additional resources from other social actors, in turn, can shape the choice of protest tactics.

Resources from other groups must be coopted or appropriated in order to be used. As Edwards & McCarthy (2004, p. 134) explain, “cooptation refers to the transparent, permitted borrowing of resources that have already been aggregated by such groups. Appropriation refers to the surreptitious exploitation of the previously aggregated resources of other groups.” Recruiting professors for a college student demonstration, using a labor union’s headquarters to organize a march for pension reform, and feminists coordinating with students to stage sit-ins in high schools in favor of a feminist mobilization are all examples of these mechanisms being used to access additional resources.

Some authors have argued that the tactics employed by specific social groups are inherently contained or aggressive. For example, the labor movement tends to be disruptive because it can interrupt economic production (through strikes or slowdowns) with relative ease vis-à-vis other groups (McAdam, 1982; Schwartz, 1988). The unemployed in Argentina, for example, tend to use coercive tactics because of their lack of political resources (Freyre, 2014, p. 43). The literature also claims that some sectors are inherently more peaceful or violent depending on their level of political capital (Bernstein, 1997; Piven & Cloward, 1979; Van Dyke et al., 2004; Walker et al., 2008). For example, inhabitants from a poor neighborhood may use more transgressive tactics to voice their demands because they do not possess the same level of influence as their counterparts in more affluent districts. While this may be true of specific social movement sectors, the addition of virtually any group into a mobilization may add extra resources or capital and will cause its tactics to follow the logic of numbers rather than the logic of damages.⁶

Why would allies lend their resources to other actors? Movements often modify their collective action frame by including the demands of their allies (Benford & Snow, 2000). Thus, additional demands may cause mobilizations to become larger and more numerous, which may promote their resonance. Some actors may also decide to support protests targeting their adversaries. Scholars have previously argued, for instance, that linkages between social movements and opposition parties can promote large-scale protests and the achievement of policy outcomes (Almeida, 2010; Stearns & Almeida, 2004; Su, 2015). Strong political oppositions, for example, can promote mobilization to destabilize their adversaries in government (Morgenstern et al., 2008, p. 183).

Incorporating other types of participants requires formulating collective action frames that are appealing to different actors (Gerhards & Rucht, 1992). This means not only adding (often disparate) demands together, but also giving up ideological conformity and extreme demands, which are associated with more transgressive tactics (McAdam, 1986; Medel & Somma, 2016). As Tilly & Tarrow explain, social movements sometimes choose to substitute “more extreme goals and more robust tactics for more moderate ones in order to maintain the interest of their supporters and attract new ones” (Tilly & Tarrow, 2015, p.3). Thus, when a movement does not have allies, its tactics will be more violent, relative to movements with several types of protest participants, which tend to be more contained.

• • • Hypothesis 2a: The presence of other actors increases the likelihood of peaceful tactics

• • Hypothesis 2b: The presence of other actors increases the likelihood disruptive tactics.

• • Hypothesis 2c: The presence of other actors decreases the likelihood of violent tactics.

The repertoire of contention, therefore, is defined to a great extent by the resources available to activists. In particular, the number of demonstrators and the presence of allies from other social groups cause protests to follow the logic of numbers rather than the logic of damages, making them more peaceful or disruptive and less violent.

III. Statistical Analyses of Student Protest Tactics in Latin America

In this section, I use the Latin American Student Protest Dataset (LASPD).⁷ This dataset includes 4,717 different protest events with college student participants⁸ in eighteen Latin American countries, between the years 2000 and 2012. The events listed on the LASPD are based on the “Chronologies of Social Conflict,” which were monthly reports compiled by the Latin American Social Observatory (Observatorio Social de América Latina, OSAL, 2012). Although the LASPD focuses on a specific social group,⁹ it includes an unprecedented number of countries, events and variables - including protest size and the presence of other types of participants - so it is a useful resource to analyze the choice of protest tactics across time and space. For the purposes of this study, the unit of analysis is each protest event.

III.1 Variables

The dependent variables are the types of tactics used in the mobilizations. The LASPD covers twenty-one different protest tactics, which are aggregated in this study into three different types of strategies: peaceful, disruptive, and violent tactics. Peaceful and disruptive tactics are assumed to be more akin to the logic of numbers, while violent ones are better associated to the logic of damages. It is important to note that these categories are not mutually exclusive: for example, the same protest may involve a peaceful march, a disruptive roadblock, and isolated violent attacks between demonstrators and counter demonstrators.

Table 1 shows the frequency of these different tactics and types. Peaceful tactics are by far the most common, with almost 72% of all events using them. Disruptive tactics are also relatively frequent (about 37% of all events), while violent tactics are rare, with just 11% of all events using these strategies. This distribution is also reflected when protest tactics are disaggregated: the five most common strategies are marches, demonstrations, roadblocks, and sit-ins at schools; meanwhile, the five least common tactics are attacks on bystanders, funas, attacks on other protesters, use of weapons, and attacks on private property.

Table 1
Frequency of peaceful, disruptive, and violent tactics in Latin American college student protests

Strategy	N	%	Type	N	%
March	1,650	34.98	Peaceful	3,392	71.91
Demonstration	1,275	27.03
Rally	204	4.32
Public statement	418	8.86
Assembly	287	6.08
Cultural acts	225	4.77
Petition	146	3.10
Commemoration	207	4.39
Strike	267	5.66	Disruptive	1,749	37.08
Sit-in (school)	636	13.48
Sit-in (elsewhere)	168	3.56
Roadblock	702	14.88
Interruption of public activity	117	2.48
Funa	51	1.08
Self-destructive acts	137	2.90	Violent	528	11.19
Attacks on public property	100	2.12
Attacks on private property	89	1.89
Weapons	83	1.76
Attacks on security forces	174	3.69
Attacks on bystanders	15	0.32
Attacks on other protesters	73	1.55
Number of observations: 4.717

Source: author’s elaboration.

The independent variables are grouped into two dimensions of protest. These dimensions are the size of protests (in terms of number of participants), and the presence or absence of additional social actors in the mobilization. The size of protests is measured using an ordinal scale: a handful of participants (1.8% of all events); ten to 49 participants, (9.1%), 50 to 100 protesters (7.4%), hundreds (20.5%), thousands (16.6%), and tens of thousands or more (9.1%).¹⁰

In terms of additional actors, several groups are sometimes allied with college students: political party members (present in approximately 17.2% of all protests), high school students (6.7%), college faculty (13.4%), primary and secondary teachers (8.3%), labor (29%), peasants (7.2%), and indigenous and Afro-descendant people (4.1%). While the literature argues that some groups may decide to carry out more disruptive or violent actions based on their economic role (McAdam, 1982; Schwartz, 1988) and level of political capital (Bernstein, 1997; Medel & Somma, 2016; Piven & Cloward, 1979; Walker et al., 2008), I expect that the addition of any of these groups will increase student activists’ resources, turning their repertoire of contention in favor of the logic of numbers and against the logic of damages. A variable denoting when none of these other actors is present is also included. The absence of any allies is expected to have a negative effect on the likelihood of more contained forms of mobilization due to the relative absence of resources.

The analysis also incorporates several control variables. These variables correspond to three dimensions of protests events: targets, demands, and location. The variables related to targets are whether the protests focus on the government or colleges. The government is by far the most common target, with approximately 70% of mobilizations targeting political authorities. The college variable is also included because about 27% of college student mobilizations are aimed at their own institutions, making them the second most common target. The expectation is that activists are more likely to use violent tactics against targets that do not possess the capacity of the state to respond with repression (Walker et al., 2008). Events targeting the government will be, therefore, relatively contained in their strategies. Tactics against colleges, on the other hand, will tend to be disruptive or violent because college authorities, which are not subject to public opinion like political elites, are less likely to be swayed by peaceful demonstrations.

A second set of control variables is the types of demands advanced during the mobilizations. These include broader political (made in 15.8% of all events) and postmaterialist demands (feminist, environmental, sexual, ethical, and animal rights, etc., 5.1%). Also included are the more student-specific claims related to education costs (free education and student debt, among others, 13.4%), education support (scholarships, student bus fares, housing and meals, 6.2%), and autonomy (autonomy for public universities, student participation in college governance, etc., 6,3%). According to Medel and Somma (2016, p. 178), radical demands, which they define as “demands that require great institutional reforms to me bet, or outright antisystemic demands”, promote riskier, costlier and more violent demands. Relatively radical demands like antigovernment protests in Venezuela, or higher education policy reforms in Chile may be, therefore, associated with more disruptive or violent forms of mobilization.

The location of the events may also have an effect on the repertoire of contention. More specifically, the events staged in capital cities may be more visible (Taylor & Van Dyke, 2007; Van Dyke et al., 2004; Walker et al., 2008), making the logic of damages less appealing. Thus, capital cities (where 42.8% of events took place), may be more likely to experience more peaceful and disruptive demonstrations.

Finally, two more variables are included for methodological purposes. First, each strategy category is used as a regressor of the other two to make sure the effects of the other independent variables are not influenced by the other tactics present in the events. Additionally, since the three different coders worked on the LASPD, a coder variable is also included to make sure the results are not a function of who coded the events.¹¹

III.2 Results and robustness¹²

Data from the LASPD is regressed using two model specifications.¹³ Table 2 describes the results of six models: the first three are logistic regression models of peaceful, disruptive and violent protest tactics (I); the last three are multilevel logistic regression models (II). The first logistic models may be ignoring time- and country-specific dynamics of students’ repertoires of contention. Unfortunately, introducing countries as fixed effects into the models causes multicollinearity issues. Multilevel models with country-year and country random intercepts are, therefore, used as an alternative to account for these sources of non-independence. In the type II models, protest events are nested within country-years, which in turn are nested within countries. All models use robust standard errors. The effects of the independent variables in the type II regressions are similar to those in the type I models. The effects of all but two variables of interest are consistent in both model specifications, which suggests that the results are relatively robust to different model specifications.¹⁴

Table 2
Effects of protest size and actors on the strategies used in college student protests

	Logistic Models (I)				Multilevel Logistic Models (II)
	Peaceful	Disruptive	Violent		Peaceful	Disruptive	Violent
Size (reference: handful)
10-49	1.524	7.612***	0.354***		1.527	8.142***	0.323***
	(0.398)	(2.796)	(0.0859)		(0.612)	(3.736)	(0.119)
50-100	4.791***	21.85***	0.144***		4.924***	23.20***	0.133***
	(1.335)	(8.473)	(0.0439)		(2.112)	(10.81)	(0.0437)
Hundreds	7.426***	20.16***	0.250***		7.959***	22.28***	0.271***
	(1.905)	(7.451)	(0.0612)		(3.454)	(11.30)	(0.0823)
Thousands	20.47***	21.54***	0.285***		24.61***	24.28***	0.324***
	(5.991)	(8.226)	(0.0749)		(12.31)	(12.87)	(0.102)
Tens of Thousands	38.95***	54.08***	0.803		42.50***	61.36***	0.921
	(14.20)	(21.18)	(0.227)		(26.05)	(35.49)	(0.347)
Not enough information	4.402***	17.52***	0.229***		4.593***	19.89***	0.217***
	(1.090)	(6.430)	(0.0535)		(2.120)	(8.487)	(0.0595)
Additional Actors
None	0.551***	1.135	0.876		0.575***	1.203	0.899
	(0.0957)	(0.151)	(0.147)		(0.110)	(0.205)	(0.212)
Party members	1.439*	0.840	0.893		1.517**	0.837	1.051
	(0.296)	(0.110)	(0.151)		(0.298)	(0.123)	(0.262)
High school students	1.476	1.349*	1.849***		1.442*	1.290	1.619
	(0.379)	(0.218)	(0.347)		(0.293)	(0.231)	(0.572)
Faculty	0.858	0.909	0.512***		0.880	0.943	0.517***
	(0.162)	(0.124)	(0.109)		(0.119)	(0.192)	(0.126)
Teachers	2.035**	1.264	0.634**		2.256***	1.193	0.663**
	(0.586)	(0.184)	(0.141)		(0.619)	(0.279)	(0.115)
Labor	1.080	0.958	0.750*		1.137	0.978	0.783
	(0.182)	(0.112)	(0.115)		(0.243)	(0.138)	(0.195)
Peasants	0.563**	0.929	0.599*		0.523**	0.940	0.529**
	(0.163)	(0.165)	(0.167)		(0.136)	(0.161)	(0.170)
Indigenous and Afrodescendant people	1.899*	1.124	1.053		1.837**	1.125	0.990
	(0.719)	(0.252)	(0.315)		(0.493)	(0.234)	(0.190)
Targets
Government	1.567***	1.491***	0.893		1.572***	1.446***	0.858
	(0.201)	(0.168)	(0.115)		(0.226)	(0.161)	(0.108)
Colleges	0.790*	2.848***	0.743**		0.770**	2.907***	0.809
	(0.103)	(0.338)	(0.110)		(0.102)	(0.400)	(0.165)
Demands
Political	1.259	0.758**	0.889		1.313	0.790	0.781
	(0.213)	(0.0920)	(0.129)		(0.242)	(0.120)	(0.124)
Postmaterialist	2.426**	0.560***	0.577*		2.552	0.536**	0.564*
	(0.890)	(0.122)	(0.180)		(1.547)	(0.162)	(0.174)
Costs	1.145	0.900	0.757*		1.240	0.914	0.743*
	(0.161)	(0.104)	(0.120)		(0.167)	(0.139)	(0.129)
Support	1.171	1.081	1.157		1.230	1.023	1.054
	(0.207)	(0.162)	(0.245)		(0.195)	(0.194)	(0.240)
Autonomy	1.158	1.631***	1.341*		1.256	1.687*	1.145
	(0.212)	(0.278)	(0.238)		(0.235)	(0.494)	(0.197)
Other tactics
Peaceful		0.0295***	0.174***			0.0271***	0.178***
		(0.00332)	(0.0319)			(0.00612)	(0.0600)
Disruptive	0.0277***		0.388***		0.0240***		0.394***
	(0.00329)		(0.0723)		(0.00609)		(0.116)
Violent	0.161***	0.394***			0.152***	0.371***
	(0.0277)	(0.0717)			(0.0440)	(0.112)
Capital City	1.338***	0.973	0.852		1.363***	0.871	1.051
	(0.134)	(0.0815)	(0.0847)		(0.164)	(0.116)	(0.150)
Coder (reference: Coder 1)
Coder 2	0.333***	0.774**	5.499***		0.329***	0.787*	3.814***
	(0.0537)	(0.0955)	(1.763)		(0.0361)	(0.109)	(0.750)
Coder 3	0.336***	1.433**	6.005***		0.316***	1.444	3.878***
	(0.0618)	(0.205)	(2.013)		(0.0494)	(0.363)	(1.140)
Constant	8.523***	0.259***	0.596		8.570***	0.246***	0.547
	(2.868)	(0.106)	(0.257)		(4.446)	(0.113)	(0.234)
Observations	4.717	4.717	4.717		4.717	4.717	4.717
Number of country-years (random intercepts)					227	227	227
Number of countries (random intercepts)					18	18	18

Source: Author’s elaboration.Obs: Odds Ratios reported. Robust standard errors in parentheses.
• ^*** p < 0.01,
• ^** p < 0.05,
• ^* p < 0.1.

The effects of size on protests tactics confirm Hypotheses 1a, 1b and 1c. Indeed, increased protest size generally has large, positive and statistically significant effects on the likelihoods of peaceful and disruptive tactics. The number of participants also has significant but negative effect on the likelihood of violent ones. For example, compared to the smallest size category (a handful of demonstrators), the presence of thousands of participants increases the odds of peaceful tactics by about 20 times and that of disruptive tactics 22 times, while it decreases the odds of violent strategies almost 72%. Interestingly, compared to a handful, the effect on violent tactics of tens of thousands or more demonstrators is not significant, which suggests that their effects are similar. Overall, however, increased protest size is positively associated with the logic of numbers, while tactics associated with fewer participants tend to follow the logic of damages.

The evidence lends partial support to the Hypotheses dealing with the presence of other actors (2a, 2b, and 2c). Most of the effects are in the expected direction, but only some of them are statistically significant. The absence of any other actors has a negative effect on the use of peaceful strategies (confirming Hypothesis 1a), but its effect is not significant on disruptive or violent tactics. The effect of peasants also goes against expectations, as their presence decreases the likelihood of both peaceful and violent tactics, which may be explained by their low levels of political capital (Bernstein, 1997; Somma & Medel, 2019). Although workers would be expected to have a positive effect on the use of disruptive strategies (McAdam, 1982; Medel & Somma, 2016; Schwartz, 1988), their presence only reduces the likelihood of violent tactics. The effect of additional actors, therefore, while not completely consistent, suggests that they add resources that veer movements towards the logic of numbers, rather than the logic of damages.

The effect of some control variables is also statistically significant. In terms of the location of events, staging protests in the capital city is associated with a 34% increase in the odds of peaceful tactics. Since the media and political elites may pay more attention to events happening in, say, the capital city of Buenos Aires than in a provincial town in Patagonia, the evidence suggests that the “logic of damages” may be more appealing away from the centers of power. Among the demand variables, postmaterialist claims have a clear pacifying effect: these demands are associated with a 143% increase in the odds of peaceful tactics,¹⁵ and a 44 and 42% decrease in the odds disruptive and violent tactics, respectively. Other demands affect only some types of tactics: political demands decrease the odds of disruptive strategies, while demands for autonomy increase the likelihood of both disruptive and violent methods. Making claims about education costs decreases the likelihood of violent strategies - in line with the finding that these demands are associated with larger protests (Disi Pavlic, 2016). The use of other types of strategies in the same event is consistently negative and statistically significant across all models, indicating that, while different types of tactics may overlap in the same event, activists tend to choose specific types of tactics from their repertoire of contention. This negative effect is particularly strong on the co-occurrence of peaceful and disruptive tactics: the use of one strategy decreases the odds of the other by 97% in both cases.

III.3 Postestimation

Fitting the logistic models (I) on Table 2, I estimate the predicted probabilities of peaceful, disruptive, and violent tactics by the protest size variable and the statistically significant additional actor variables.¹⁶ Figure 1 shows the predicted probabilities of the three types of tactics by each category of the size variable.¹⁷ The probability of peaceful tactics grows as the number of participants increases, going from 51% with a handful of demonstrators to 98% at tens of thousands or more, in line with the logic of numbers, and supporting Hypothesis 1a. In the case of disruptive tactics, the number probability tends to increase with protest size, lending support to Hypothesis 1b. Meanwhile, the probabilities of violent tactics indicate that the relationship between protest size and these tactics is somewhat U-shaped: the two categories with the highest predicted probabilities are those with the smallest (23%) and largest (20%) numbers of participants. Violent tactics, therefore, are more common in small protests of likeminded, radical activists, but that they also tend to occur in the largest ones, perhaps due to increased state repression (Della Porta, 2006; Koopmans, 1997) or decreased risks for individual activists of performing violent actions (Granovetter, 1978; Oberschall, 2017; Taylor & Van Dyke, 2007). Larger numbers may, for example, increase anonymity, which decreases the danger of committing violent acts.

Figure 1
Predicted Probabilities of Tactics by Protest Size, with 95% CIs

Table 3 shows the predicted probabilities of protest tactics by the additional actor variables that were statistically significant across the first three models. The results show that the presence of allies increases the probability of peaceful tactics, while it decreases the predicted probabilities of violent strategies. When the only actor present is college students, the probability of using peaceful tactics is about 83%, while the presence of any additional actor increases that probability to approximately 90%, which supports Hypothesis 2a. The presence of peasants, however, make peaceful tactics less likely (but they also decrease the probability of violent actions). The predicted probabilities of disruptive actions increase from 34% without high schoolers to approximately 42% when they are present. These results indicate that the effect of additional actors on tactics in college student demonstrations is generally pacifying, but that it by varies significantly by the type of actor. The probability of violent actions decreases with the presence of all actors - with the exception of high schoolers.

Table 3
Predicted Probabilities of Tactics by Additional Actors, with 95% CIs

Additional actors		Tactic Type
		Peaceful	Disruptive	Violent
Students only	No	0.901***	Nonsignificant	Nonsignificant
		(0.00919)
	Yes	0.834***
		(0.0156)
Party Members	No	0.867***	Nonsignificant	Nonsignificant
		(0.00795)
	Yes	0.904***
		(0.0165)
High school students	No	Nonsignificant	0.344***	0.0755***
			(0.00937)	(0.00507)
	Yes		0.415***	0.131***
			(0.0373)	(0.0206)
College faculty	No	Nonsignificant	Nonsignificant	0.0854***
				(0.00557)
	Yes			0.0456***
				(0.00906)
Teachers	No	0.868***	Nonsignificant	0.0812***
		(0.00764)		(0.00530)
	Yes	0.930***		0.0531***
		(0.0179)		(0.0111)
Labor	No	Nonsignificant	Nonsignificant	0.0847***
				(0.00603)
	Yes			0.0649***
				(0.00831)
Peasants	No	0.879***	Nonsignificant	0.0811***
		(0.00728)		(0.00538)
	Yes	0.803***		0.0502***
		(0.0432)		(0.0130)
Indigenous and Afro-descendant people	No	0.872***	Nonsignificant	Nonsignificant
		(0.00721)
	Yes	0.928***
		(0.0251)
Observations		4.717	4.717	4.717

Source: Author’s elaboration.Obs: Standard errors in parentheses.
• ^*** p < 0.01,
• ^** p < 0.05,
• ^* p < 0.1.

IV. Tactics and Human Resources during the Demonstrations Against the Ley Pulpín

This section analyzes the way resources shaped the tactics used in the mobilization against the so-called Ley Pulpín in Peru. This law, passed on December 11, 2014 reduced the mandatory job benefits of 18 to 24-year-old Peruvians. According to its proponents, the goal of the law was reducing that segment’s unemployment and informal employment rates (Ford Cole, 2018, p. 50) but in reality the law followed a trend of reforms in Peru making labor more flexible and unregulated (Fernández-Maldonado, 2015). The movement included several different actors but most of the participants were college students (Fowks, 2015). After five demonstrations, the government repealed the law on January 25, 2015. Variation in the explanatory and dependent variables in a short time span within the same movement provides an excellent opportunity to assess the causal impact (Gerring & McDermott, 2007) of resources on tactics.

The evidence discussed below is based on eighteen in-depth interviews carried out by the author between February and May of 2015 in Lima. Interviewees included current and former student leaders and activists, faculty, and university and government officials. Their insights are used to analyze the movements’ dynamics. This primary evidence is complemented with newspaper articles and other secondary data, which registered some basic facts about the Pulpín rebellion.

IV.1 The Evolution of the tactics used during the youth movement against the Ley Pulpín

The movement consisted of five main episodes of protest (Fowks, 2015; La República, 2015). The first march, which was organized organically through social media, took place on December 19, eight days after the bill was passed. About four thousand people gathered in downtown Lima and marched towards the upscale district of Miraflores. The incidents that took place during the protest demonstrate its organizers’ lack of experience:

The first march against the [Ley Pulpín]was a total chaos. On the one hand, the mass of young people moved about through different streets of Lima with very little organization which, combined with police repression, resulted in more than twenty young people being arrested, and in a large number of injured persons, including a person ran over by a police patrol (Díaz Zanelli, 2015).

The second march on December 23 gathered more than 10,000 people, and included demonstrations outside the headquarters of the National Confederation of Private Business Institutions (CONFIEP), an interest group which had lobbied in favor of the law.¹⁸ The third march was on December 29, and had a similar number of participants and actions. The fourth march occurred on January 15 and headed towards the National Congress. Demonstrators presented congressional officials with a list of demands. During this event “violence marked the day, during which young people denounced the presence of hooded figures” (La República, 2015). The Minister of Interior attributed the violence in an otherwise peaceful demonstration to the presence of certain labor organizations and activists from the Movement for Amnesty and Fundamental Rights (Movimiento por la Amnistía y los Derechos Fundamentales, MOVADEF) (La Prensa, 2015; Mendoza, 2015).¹⁹ When Congress decided to debate the law, President Ollanta Humala used a national television address to defend the bill. However, public opinion had already turned against the Ley Pulpín and had a very positive opinion of the movement (GFK, 2015), so the President’s own party announced it would vote to repeal the law on January 25 (Fernández-Maldonado, 2015). Finally, the law was repealed on January 27, the same day as last of demonstration, which gathered two thousand young protesters.

These protests were relatively unusual in the Peruvian context (Ford Cole, 2018, p. 49). This was the first time a social movement was able to thwart a labor reform since the implementation of a market economy in the country in the 1990s (Fernández-Maldonado, 2015, pp.166-167). Unlike other demonstrations by students and young people since the return to democracy in the 2000s, the protests were massive and largely peaceful, and had an important amount of first-time and low-income participants (Cavero, 2015). The relatively contained character of the movement can be found in the resources activists were able to amass. Initially, opposition parties were important contributors, but most of their resources came from the way young activists adapted their organization, which resulted in larger and more peaceful demonstrations.

IV.2 The minor role of parties, and the advantages of autonomy

Political parties from the opposition, which had a strong presence and leadership in the early stages of the mobilization through the “18D” coordinating group (named after the day before the first march), attempted to “appropriate” (Edwards & McCarthy, 2004, p. 134) the movement. Participants in the movement accused political parties like PCP-PR²⁰ and PAP²¹ of attempting to aparatear (coopt) the movement for their own purposes, and in particular to take advantage of the movement’s popularity in preparation for the presidential and congressional elections of 2016 (Chávez, 2015).²² Other leaders, however, were able to organize the mobilization in such a way that party interests could not capture the movement.

Members of PAP attempted to increase their influence - and prevent other parties controlling the mobilizations - after the first march by proposing to assign the organization of the movement in Lima and Callao to fourteen different zonas (zones).²³ Zonas were geographically disaggregated, and organized movement participants into groups of residents from contiguous municipalities (Paucar Albino, 2017). The initiative backfired for the PAP: the rest of the movement, including several university organizations, which had a strong anti-party stance,²⁴ decided to organize the movement so that that no single participating organization could appropriate the zonas (García & Vela, 2015). Another objective was to weed out infiltrators, who were accused of having connections to the police and causing public disorder during the demonstrations.²⁵

Political party activists, therefore, were forced to participate in their roles as zona residents, and the connection between the movement and parties was weakened. Several sources attribute the movement’s legitimacy and acceptance, its ability to summon large crowds, and its ultimate success in repealing the Ley Pulpín to the distance the organization took from formal politics and parties.²⁶ Indeed, movement members adopted an increasingly anti-party stance at meetings and during the demonstrations.²⁷ Parties and party members, therefore, participated in the mobilizations against the Ley Pulpín but had to accept a secondary role to avoid being left out.²⁸

In sum, the main resource the movement was its organizational strength, which was “unprecedented” (Díaz Zanelli, 2015) by Peruvian standards. Zonas were devised to work in a horizontal, nonhierarchical way, with all decisions made in open assemblies. All members were encouraged (and sometimes compelled if they had been quiet for too long) to voice their opinions. The leaders acted only as spokespersons and had to rotate periodically. Its horizontal nature also promoted the incorporation of many members of college student and workers’ unions, in addition to young people (the targets of the Ley Pulpín). This organizational structure increased the number of participants, and reduced clashes with the police after the first march.²⁹ As mentioned above, the only exception occurred in the fourth march, when some alleged members of MOVADEF and other political groups attacked security forces. They threw paint, rocks and other objects when the police attempted to prevent demonstrators from reaching Congress (La Prensa, 2015), which resulted in dozens of arrests and several injured police officers.³⁰ However, most of the tactics used by the movement included peaceful and sometimes disruptive acts (resulting in some arrests) like demonstrations at Lima’s Plaza Mayor (Main Square) with a Peruvian flag, blockading urban highways, staging protests in the city’s financial district, and displaying banners in shopping malls (Díaz Zanelli, 2015).

The qualitative evidence is aligned with the quantitative results. The case study shows that the positive effect of protest size is due to the larger organizational capacity required to gather more participants. At the same time, the violent actions during the fourth march suggest that “[t]he cost to an individual of joining a riot declines as riot size increases, since the probability of being apprehended is smaller the larger the number involved” (Granovetter, 1978, p. 1422). The case study also suggests that, in order for the effect of additional actors to be positive, their inclusion must be coordinated - violent actions were allegedly carried out by “infiltrators” (La Prensa, 2014). As Piven & Cloward (1995, p. 158) observe, “[i]t is not that disruption and violence are never employed by formally organized groups; it is that, in general, organization constrains against such tactics.”

V. Conclusion

This article shows how the presence of allies and the number of demonstrators, as major resources available to social movements, shape the choice of tactics used in protest events. Unlike other works, which analyze the types of protest tactics as discrete choices (Medel & Somma, 2016), this work recognizes the fact the different types of tactics may occur in the same demonstration. In short, it argues that protests with both more participants and allies should follow the logic of numbers, while reduced participants and the absence of allies make mobilizations more likely to follow the logic of damages.

The quantitative analyses demonstrate that size is an important determinant of the repertoire of contention. Larger numbers of participants are associated with higher likelihoods of peaceful and disruptive tactics, while it decreases the odds of more violent methods. At the same time, however, very large protests also tend to experience more violent actions than medium-sized events (but fewer than the smallest demonstrations). Thus, results show that the factors associated with using, say, peaceful strategies may sometimes overlap with those associated with violent tactics.

The effect of additional actors is less clear-cut. The absence of any actor has a negative effect on the use of peaceful strategies, but its effect on disruptive and violent tactics is not statistically significant. Meanwhile, the presence of some specific actors had the expected effect on all types of protests; participation by other groups had the expected effect but only on one or two tactic variables; finally, a few actors did not conform to expectations. The effect of these actors (high school students and peasants) is in line, however, with the argument the resource-poor groups are less likely to use peaceful actions, and more prone to resort to violent tactics (Medel & Somma, 2016; Piven & Cloward, 1979; Van Dyke et al., 2004). The evidence, therefore, supports the argument that decreased political cohesiveness increases the use of violent tactics (Pearlman, 2012) - both very large numbers of participants and the addition of disparate actors into a mobilization may undermine its cohesiveness.

The case study of the movement against the Ley Pulpín in Peru in 2014-2015 illustrates how participants, allies, and organizational resources in general caused these mobilizations to be relatively contained. After an unsuccessful, disorganized, and violent first march, the movement was organized geographically and horizontally. While opposition parties were the ones initially organizing the protests, young organizers feared that the parties would appropriate the movement’s resources for electoral purposes, and therefore began to organize themselves autonomously. This horizontal organization promoted the recruitment of both more participants and more organizations. The only exception was the participation of certain political groups like MOVADEF, whose involvement in the fourth march a led small number of participants to attack security forces outside the Peruvian Congress. Thus, in general, this movement’s tactics tended to become more contained as it began to follow the logic of numbers, acting independently from particular interests, and amassing more human and organizational resources from additional individual participants and social organizations.

• 1
  I thank Raul Madrid, Belén Cumsille R. and the anonymous reviewers of Revista de Sociologia e Política for their insightful comments on this article. I acknowledge support from the Centre for Social Conflict and Cohesion Studies- COES (ANID/FONDAP/15130009).
• 2
  “Pulpín” comes from the Pulp brand of fruit juice, which is aimed at children and comes in a colorful, odd-shaped container. The word “chibolo” (boy, youngster) was added to create the expression “chibolo pulpín,” which refers to inexperienced, wholesome, naïve, and exploitable young people (Collyns, 2015; Ford Cole, 2018, p. 49).
• 3
  Coined by Tilly (1986) it “comprises what people know they can do when they want to oppose a public decision they consider unjust or threatening” (Della Porta et al., 2017, p. 21), In its most recent iteration, Tilly (2007) incorporated broadened to the concept, after being “criticized for focusing only on public displays of disruptive action” (Della Porta et al., 2017, p. 22).
• 4
  Other exceptions include works emphasizing the importance for tactical choices of movement organizations’ appeals to the general public (Barkan, 1979), their probability of success (Oberschall, 1994), organizational identities (Meyer, 2004), position in inter-organizational networks, (Hadden, 2015), and their targets and claims (Gonzalez Vaillant & Schwartz, 2019; Wang & Piazza, 2016).
• 5
  This combination of large- and small-n approaches in a sequential research design “can help explain the mechanisms - how quantified variables interact - through the use of qualitative data” (Ayoub et al., 2014, p. 69).
• 6
  This is particularly true in the case of allies with high levels of political capital. These groups are very unlikely to adopt more transgressive protest strategies, which may hurt their reputations or turn public opinion against them (Bernstein, 1997; Medel and Somma, 2016; Walker et al., 2008).
• 7
  For a discussion and description of the LASPD, see Disi Pavlic (2020).
• 8
  The LASPD distinguishes between high school and college students, based on the words used to describe participants and the organizations involved in the events.
• 9
  Colleges student are not a unitary actor, as student movements sometimes have to unite members from various types of institutions: public and private; universities and non-university vocational institutions; and schools with and without student unions (Montero Barriga et al., 2017, pp.145-146). Unfortunately, news events in the LASPD rarely made a distinction between different types of college students.
• 10
  Unfortunately, 36.7% of the protest events in the LASPD did not have enough information to assess their size. The “not enough information” category is, therefore, included to the model to keep these observations in the analysis.
• 11
  Coders were in charge of specific countries, so this variable is highly correlated with the country variable. Coders also varied significantly in terms of assigning missing values to the protest size variable.
• 12
  Data and replications codes are available on the author’s website.
• 13
  The analyses were carried out using STATA 15 (StataCorp, 2017).
• 14
  The significance level of the effects of high school students and labor are not consistent across the logistic (I) and the multilevel logit (II) models.
• 15
  This result is not statistically significant in the multilevel logistic model, however.
• 16
  The rest of the variables are used at their means. All the results are statistically significant.
• 17
  The predicted probabilities when there is not enough information about the number of protesters suggests that these events are mid-sized.
• 18
  This is the first time the CONFIEP has been directly targeted by protests in Peru (Fernández-Maldonado, 2015, p. 142, fn. 2; Fowks, 2015).
• 19
  MOVADEF is a political movement that is, according to several observers, a continuation of the Shining Path terrorist organization (Pighi Bel, 2017; Salazar & Tamara, 2011; Sandoval, 2012, p. 1).
• 20
  Communist Party of Peru - Red Fatherland. Not to be confused with the Communist Party of Peru - Shining Path.
• 21
  Peruvian Aprista Party. In turn, APRA is the acronym for American Popular Revolutionary Alliance, founded in 1924.
• 22
  Interview with Álvaro Vidal. Student leader, Pontifical Catholic University of Peru (PUCP) and UNMSM. May 13, 2015.
• 23
  Interview with César Ames. President, Universidad Nacional Mayor de San Marcos (UNMSM) Social Sciences Student Union. The idea also arose out of the unsuccessful experience in the first demonstration (Díaz Zanelli, 2015).
• 24
  Interview with Lilia Ramírez. Lawyer, Legal Defense Institute (IDL). March 3, 2015.
• 25
  Interview with César Ames.
• 26
  Others, however, believed that the movement would not survive if it rejected traditional organization, hierarchy, and cooperation with parties (Cavero, 2015).
• 27
  Interview with Lilia Ramírez.
• 28
  The movement also incorporated labor unions, but was not subjected to them either (Fernández-Maldonado, 2015, p. 142).
• 29
  This undermines an alternative explanation, which argues that activists prefer peaceful tactics because violent ones preclude higher participation - in other words, that tactics affects numbers (Chenoweth & Stephan, 2013; Schock, 2005). The Pulpín rebellion shows that tactics within the same movement became more peaceful after becoming better organized, more numerous and with the presence of more organizations.
• 30
  Most participants did not engage with the police violently, however, and about twenty of them convinced security forces to let them deliver a letter to representatives in Congress (La Prensa, 2015). The activists using violence also attacked other demonstrators (Mendoza, 2015).
• A produção desse manuscrito foi viabilizada através do patrocínio fornecido pelo Centro Universitário Internacional Uninter à Revista de Sociologia e Política.

References

• Almeida, P. (2010) Social Movement Partyism: Collective Action and Political Parties. In: Van Dyke, N. & McCammon, H.J. (Eds.) Strategic Alliances: Coalition Building and Social Movements. Minneapolis: University of Minnesota Press, pp.170-196.
• Ayoub, P.M., Wallace, S.J. & Zepeda-Millán, C. (2014) Triangulation in social movement research. In: Della Porta, D. (Ed.) Methodological practices in social movement research. Oxford University Press, pp.67-96.
• Barkan, S.E. (1979) Strategic, tactical and organizational dilemmas of the protest movement against nuclear power. Social Problems, 27(1), pp.19-37. DOI: 10.2307/800014
  » https://doi.org/10.2307/800014
• Benford, R.D. & Snow, D.A. (2000) Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26(1), pp.611-639. DOI: 10.1146/annurev.soc.26.1.611
  » https://doi.org/10.1146/annurev.soc.26.1.611
• Bernstein, M. (1997) Celebration and Supression: The Strategic Uses of Identity by the Lesbian and Gay Movement. American Journal of Sociology, 103(3), pp.531-565. DOI: 10.1086/231250
  » https://doi.org/10.1086/231250
• Cavero, O. (2015) Reflexiones sobre las movilizaciones en contra del régimen laboral juvenil. In: Movilizaciones en contra del régimen laboral juvenil: ¿Nueva Cultura política juvenil? La Colmena, Lima, Peru.
• Chávez, N. (2015) Movilizaciones contra la Ley Pulpín. In: Movilizaciones en contra del régimen laboral juvenil: ¿Nueva Cultura política juvenil? La Colmena, Lima, Peru.
• Chenoweth, E., Stephan, M.J. (2013) Why civil resistance works: the strategic logic of nonviolent conflict New York: Columbia University Press.
• Coleman, J. S. (2000) Social capital in the creation of human capital. In: Eric L. Lesser (Ed.) Knowledge and Social Capital. Elsevier, pp.17-41. DOI: 10.1016/B978-0-7506-7222-1.50005-2
  » https://doi.org/10.1016/B978-0-7506-7222-1.50005-2
• Cunningham, K.G. (2013) Understanding strategic choice: The determinants of civil war and nonviolent campaign in self-determination disputes. Journal of Peace Research, 50(3), pp.291-304. DOI: 10.1177/0022343313475467
  » https://doi.org/10.1177/0022343313475467
• Della Porta, D. (2013) Repertoires of Contention. In: Snow, D.A., Della Porta, D., Klandermans, B. & McAdam, D. (Eds.) The Wiley-Blackwell Encyclopedia of Social and Political Movements Oxford, UK: Blackwell Publishing Ltd. DOI: 10.1002/9780470674871.wbespm178
  » https://doi.org/10.1002/9780470674871.wbespm178
• Della Porta, D. (2006) Social movements, political violence, and the state: a comparative analysis of Italy and Germany Cambridge: Cambridge University Press.
• Della Porta, D., Diani, M. (2006) Social movements: an introduction Second edition Malden, MA: Ed. Blackwell Publishing.
• Della Porta, D., O’Connor, F.P., Portos García, M., Subirats Ribas, A. (2017) Social movements and referendums from below: direct democracy in the neoliberal crisis. Chicago, IL: Policy Press.
• DeNardo, J. (1985) Power in numbers: The political strategy of protest and rebellion Princeton University Press.
• Disi Pavlic, R. (2020) Policies, parties, and protests: explaining student protest events in Latin America. Social Movement Studies, 19(2), pp.183-200. DOI: 10.1080/14742837.2019.1629281
  » https://doi.org/10.1080/14742837.2019.1629281
• Disi Pavlic, R. (2016) Still Massive? Explaining the Size of Tertiary Student Protests in Latin America. In: XI Congreso Chileno de Ciencia Política Pucón, Chile.
• Edwards, B. & McCarthy, J.D. (2004) Resources and Social Movement Mobilization. In: Snow, D.A., Soule, S.A. & Kriesi, H. (Eds.) The Blackwell Companion to Social Movements Malden, MA: Blackwell Publishing Ltd, pp.116-152.
• Fernández-Maldonado, E. (2015) Perú: de la Ley General del Trabajo al Régimen Laboral «Pulpín». Cuadernos del CENDES, 32(89), pp.141-172.
• Ford Cole, F. (2018) Neoliberalism’s law in Peru: A Model. In: MacNaughton, G. & Frey, D.F. (Eds.) Economic and Social Rights in a Neoliberal World Cambridge: Cambridge University Press, pp.41-58.
• Freyre, M.L. (2014) El problema del desempleo en Argentina y el surgimiento de los Planes de Empleo y sostenimiento de ingresos en la agenda pública. Revista de Sociologia e Política, 22(51) pp.35-54. DOI: 10.1590/1678-987314225103
  » https://doi.org/10.1590/1678-987314225103
• Gamson, W.A. (1990) The strategy of social protest 2º ed. California: Wadsworth Pub.
• García, L. & Vela, J. (2015) Las “zonas” o la inesperada virtud de la anarquía. Argumentos, 9(1), pp.43-49.
• Gerhards, J. & Rucht, D. (1992) Mesomobilization: Organizing and framing in two protest campaigns in West Germany. American journal of sociology, 98(3), pp.555-596. DOI: 10.1086/230049
  » https://doi.org/10.1086/230049
• Gerring, J. & McDermott, R. (2007) An Experimental Template for Case Study Research. American Journal of Political Science, 51(3) pp.688-701. DOI: 10.1111/j.1540-5907.2007.00275.x
  » https://doi.org/10.1111/j.1540-5907.2007.00275.x
• Gonzalez Vaillant, G. & Schwartz, M. (2019) Student movements and the power of disruption. Partecipazione e conflitto, 12(1), pp.112-141.
• Granovetter, M. (1978) Threshold models of collective behavior. American Journal of Sociology, 83(6), pp.1420-1443. DOI: 10.1086/226707
  » https://doi.org/10.1086/226707
• Hadden, J. (2015) Networks in contention: The Divisive Politics of Climate Change Cambridge: Cambridge University Press. DOI: 10.1017/CBO9781316105542
  » https://doi.org/10.1017/CBO9781316105542
• Koopmans, R. (1997) Dynamics of Repression and Mobilization: The German Extreme Right in The 1990s. Mobilization: An International Quarterly, 2(2), pp.149-164. DOI: 10.17813/maiq.2.2.e6g82877674x6048
  » https://doi.org/10.17813/maiq.2.2.e6g82877674x6048
• Koopmans, R. (1993) The Dynamics of protest waves: West Germany, 1965 to 1989. American Sociological Review, 58(5) pp.637-658. DOI: 10.2307/2096279
  » https://doi.org/10.2307/2096279
• Maney, G.M., Kutz-Flamenbaum, RV, Rohlinger, D.A. & Goodwin, J. (Eds) (2012) Strategies for social change Minneapolis: University of Minnesota Press.
• McAdam, D. (1986) Recruitment to high-risk activism: The case of freedom summer. American Journal of Sociology, 92(1) pp.64-90.
• McAdam, D. (1983) Tactical Innovation and the pace of insurgency. American Sociological Review 48(6) pp.735-754. DOI: 10.2307/2095322
  » https://doi.org/10.2307/2095322
• McAdam, D. (1982) The Political Process Model. In: McAdam, D (1982) Political Process and the Development of Black Insurgency, 1930-1970 Chicago: University of Chicago Press, pp.36-59.
• McAdam, D., Tarrow, S.G., Tilly, C. (2001) Dynamics of contention New York: Cambridge University Press.
• McCarthy, J.D. & Zald, M.N. (1977) Resource Mobilization and Social Movements: A Partial Theory. American Journal Sociology, 82(6), pp.1212-1241.
• Medel, R. & Somma, N. (2016) ¿Marchas, ocupaciones o barricadas? Explorando los determinantes de las tácticas de la protesta en Chile. Política y Gobierno, 23(1), pp.163-199.
• Meyer, M. (2004) Organizational identity, political contexts, and SMO action: Explaining the tactical choices made by peace organizations in Israel, Northern Ireland, and South Africa. Social Movement Studies, 3(2), pp.167-197. DOI: 10.1080/1474283042000266119
  » https://doi.org/10.1080/1474283042000266119
• Montero Barriga, V., Muñoz, C. & Picazo, M.I. (2017) Estrategias y Recursos Empleados por el Movimiento Estudiantil en el 2011. Universum (Talca) 32(1), pp.137-157. DOI: 10.4067/S0718-23762017000100137
  » https://doi.org/10.4067/S0718-23762017000100137
• Morgenstern, S., Negri, J.J. & Pérez-Liñán, A. (2008) Parliamentary Opposition in non-parliamentary regimes: Latin America. The Journal of Legislative Studies, 14(1-2), pp.160-189. DOI: 10.1080/13572330801921166
  » https://doi.org/10.1080/13572330801921166
• Oberschall, A. (2017) Social Movements: Ideologies, Interest, and Identities New York: Routledge.
• Oberschall, A.R. (1994) Rational choice in collective protests. Rationality and Society, 6(1), pp.79-100.
• Pearlman, W. (2012) Precluding Nonviolence, propelling violence: The effect of internal fragmentation on movement protest. Studies in Comparative International Development, 47(1), pp.23-46. DOI: 10.1007/s12116-012-9099-2
  » https://doi.org/10.1007/s12116-012-9099-2
• Piven, F.F. & Cloward, R.A. (1995) Collective protest: A critique of resource-mobilization theory. In: Lyman, S.M. (Ed.), Social Movements: Critiques, Concepts, Case-Studies London UK: Palgrave Macmillan, pp.137-167. DOI: 10.1007/978-1-349-23747-0_8
  » https://doi.org/10.1007/978-1-349-23747-0_8
• Piven, F.F. & Cloward, R.A. (1979) Poor people’s movements: Why they succeed, how they fail New York: Vintage Books.
• Popovic, S. & Miller, M. (2015) Blueprint for Revolution: How to use Rice pudding, Lego Men, and other Nonviolent techniques to galvanize communities, overthrow dictators, or simply change the world New York: Spiegel & Grau.
• Salazar, D.,Tamara, L. (2011) MOVADEF: rezagos de un fenómeno violento. Revista Andina Estududios Políticos, 1(4), pp.1-20.
• Sandoval, P. (2012) El genio y la botella: sobre Movadef y Sendero Luminoso en San Marcos. Revista Argumentos, 6(5), pp.1-7.
• Schock, K. (2005) Unarmed insurrections: people power movements in nondemocracies Social movements, protest, and contention. Minneapolis, London: University of Minnesota Press.
• Schwartz, M. (1988) Radical protest and social structure: The Southern Farmers’ Alliance and cotton tenancy, 1880-1890. Chicago: The University of Chicago Press.
• Somma, N. & Medel, R. (2019) What makes a big demonstration? Exploring the impact of mobilization strategies on the size of demonstrations. Social Movement Studies, 18(2), pp.233-251. DOI: 10.1080/14742837.2018.1532285
  » https://doi.org/10.1080/14742837.2018.1532285
• Stearns, L.B. & Almeida, P.D. (2004) The Formation of state actor-social movement coalitions and favorable policy outcomes. Social Problems, 51(4), pp.478-504. DOI: 10.1525/sp.2004.51.4.478
  » https://doi.org/10.1525/sp.2004.51.4.478
• Steedly, H.R. & Foley, J.W. (1979) The success of protest groups: Multivariate analyses. Social Science Research. 8(1), pp.1-15. DOI: 10.1016/0049-089X(79)90011-5
  » https://doi.org/10.1016/0049-089X(79)90011-5
• Su, Y.P. (2015) Anti-Government protests in democracies: A test of institutional explanations. Comparative Politics, 47(2), pp.149-167. DOI: 10.5129/001041515814224444
  » https://doi.org/10.5129/001041515814224444
• Taylor, V. & Van Dyke, N. (2007) ‘Get up, Stand up’: Tactical repertoires of social movements. In: Snow, D.A., Soule, S.A. & Kriesi, H. (Eds.) The Blackwell Companion to Social Movements Oxford, UK: Blackwell Publishing Ltd, pp.262-293. DOI: 10.1002/9780470999103.ch12
  » https://doi.org/10.1002/9780470999103.ch12
• Tilly, C. (2015) Popular Contention in Great Britain, 1758-1834 New York: Routledge.
• Tilly, C. (2007) Democracy Cambridge: Cambridge University Press. DOI: 10.1017/CBO9780511804922
  » https://doi.org/10.1017/CBO9780511804922
• Tilly, C. (1986) The contentious French: four centuries of popular struggle. Harvard University Press.
• Tilly, C., Tarrow, S.G. (2015) Contentious politics (Second revised edition. ed). New York, NY: Oxford University Press.
• Van Dyke, N. (2003) Crossing movement boundaries: Factors that facilitate coalition protest by American college students, 1930-1990. Social Problems, 50(2), pp.226-250. DOI: 10.1525/sp.2003.50.2.226
  » https://doi.org/10.1525/sp.2003.50.2.226
• Van Dyke, N., Soule, S.A. & Taylor, V.A. (2004) The targets of social movements: Beyond a focus on the state. In: Research in Social Movements, Conflicts and Change (ed) Authority in contention United Kingdom: Emerald Group Publishing Limited, Vol. 25, pp.27-51. DOI: 10.1016/S0163-786X(04)25002-9
  » https://doi.org/10.1016/S0163-786X(04)25002-9
• Walker, E.T., Martin, A.W. & McCarthy, J.D. (2008) Confronting the State, the Corporation, and the Academy: The influence of institutional targets on social movement repertoires. American Journal of Sociology, 114(1), pp.35-76. DOI: 10.1086/588737
  » https://doi.org/10.1086/588737
• Wang, D.J. & Piazza, A. (2016) The use of disruptive tactics in protest as a trade-off: the role of social movement claims. Social Forces, 94(4), pp.1675-1710. DOI: 10.1093/sf/sov124
  » https://doi.org/10.1093/sf/sov124

Newspaper articles

• Collyns, D. (2015) Peru labour law sparks backlash from enraged youth amid calls for repeal. The Guardian London, 18.Jan. Available through: <https://www.theguardian.com/global-development/2015/jan/16/peru-labour-law-pulpin-youth-employment-protest>. Accessed at 08 feb. 2021.
  » https://www.theguardian.com/global-development/2015/jan/16/peru-labour-law-pulpin-youth-employment-protest
• Díaz Zanelli, J.C. (2015) Cómo se organizaron los movimientos juveniles que acabaron con la Ley Pulpín. La República. Buenos Aires, 01.Feb. Available through: <https://larepublica.pe/01-02-2015/como-se-organizaron-los-movimientos-juveniles-que-acabaron-con-la-ley-pulpin/>. Accessed at 08 Aug. 2020.
  » https://larepublica.pe/01-02-2015/como-se-organizaron-los-movimientos-juveniles-que-acabaron-con-la-ley-pulpin
• Fowks, J. (2015) Crecen las protestas contra la reforma laboral en Perú. El País Lima, 16.Jan. Available through: <https://elpais.com/internacional/2015/01/16/actualidad/1421384190_409077.html>. Accessed at 08 feb.2021.
  » https://elpais.com/internacional/2015/01/16/actualidad/1421384190_409077.html
• GFK (2015) GFK: el 54% cree que el Congreso derogará la “Ley Pulpín,” América TV Lima, 25.Jan. Available through: <https://www.americatv.com.pe/noticias/actualidad/aumenta-rechazo-ley-pulpin-76-encuestados-esta-contra-n167898>. Accessed at 08 feb.2021.
  » https://www.americatv.com.pe/noticias/actualidad/aumenta-rechazo-ley-pulpin-76-encuestados-esta-contra-n167898
• La Prensa (2015) Ley Pulpín: Jóvenes fueron liberados tras 20 horas de permanecer detenidos. La Prensa Lima, 16.Jan. Available through: <https://laprensa.peru.com/actualidad/noticia-ley-pulpin-marcha-regimen-laboral-juvenil-protesta-37754>. Accessed at 10 Feb. 2021.
  » https://laprensa.peru.com/actualidad/noticia-ley-pulpin-marcha-regimen-laboral-juvenil-protesta-37754
• La Prensa (2014) Policía cree que infiltrados provocaron violencia al final de protesta contra régimen laboral. La Prensa Lima, 23.Dec. Available through: <https://laprensa.peru.com/actualidad/noticia-protesta-regimen-laboral-juvenil-ley-pulpin-daniel-urresti-36576>. Accessed at 10 feb. 2021.
  » https://laprensa.peru.com/actualidad/noticia-protesta-regimen-laboral-juvenil-ley-pulpin-daniel-urresti-36576
• La República (2015) Ley laboral juvenil: así se gestó su derogatoria. La República Buenos Aires, 26.Jan. Available through: <https://web.archive.org/web/20150127152226/http://www.larepublica.pe/26-01-2015/ley-laboral-juvenil-asi-se-gesto-su-derogatoria>. Accessed at 10 Feb. 2021.
  » https://web.archive.org/web/20150127152226/http://www.larepublica.pe/26-01-2015/ley-laboral-juvenil-asi-se-gesto-su-derogatoria
• Mendoza, R. (2015) “Ley Pulpín”: En medio de incidentes jóvenes marchan hacia el Congreso contra Ley Laboral Juvenil D. Correo.
• Paucar Albino, J. (2017) Se cumplen dos años de la derogatoria de la “Ley Pulpín.” La Mula 26. Jan. Available through: <https://redaccion.lamula.pe/2017/01/26/ley-pulpin-peru-marcha-significado-jovenes-derechos-laborales/jorgepaucar/>. Accessed at 10 feb. 2021.
  » https://redaccion.lamula.pe/2017/01/26/ley-pulpin-peru-marcha-significado-jovenes-derechos-laborales/jorgepaucar/
• Pighi Bel, P. (2017) Qué es el Movadef, la polémica organización que vinculan con Sendero Luminoso en Perú y a la que acusan de apología al terrorismo. BBC News Mundo 9. Aug. Available through: <https://www.bbc.com/mundo/noticias-america-latina-40856626>. Accessed at 10 feb.2021.
  » https://www.bbc.com/mundo/noticias-america-latina-40856626

Others sources

• OSAL, (2012) Cronologías del Conflicto Social. Observatorio Social de Américal Latina, Consejo Latinoamericano de Ciencias Sociales
• StataCorp (2017) Stata Statistical Software: Release 15. StataCorp, LP, College Station, TX.

Publication Dates

• Publication in this collection
  09 June 2021
• Date of issue
  2020

History

• Received
  08 Oct 2019
• Accepted
  23 June 2020
• Received
  22 Aug 2020