From Rallies to Riots: Why Some Protests Become Violent

• Date: 01 May 2020
• DOI: 10.1177/0022002719887491
• Author: Brandon Ives,Jacob S. Lewis
• Published date: 01 May 2020
• Subject Matter: Articles

Article

From Rallies to Riots:

Why Some Protests

Become Violent

Brandon Ives

1

, and Jacob S. Lewis

2

Abstract

When do nonviolent protests escalate into violence? Existing literature has focused

primarily on campaign-level escalations and only recently has work begun to examine

protest event-level escalations. We build on this emerging literature and develop an

argument for why some protests escalate to violence. We use statistical analysis and

find that violent escalations are more likely to occur following recent repression and

when protests are unorganized. Our results offer insight into the conditions in which

protests remain peaceful and offer citizens a channel to pursue their goals as well as

the conditions in which protests become violent and destabilizing.

Keywords

conflict, escalation, repression, protest, subnational politics

When do nonviolent protests escalate into violence? Recent scholarship has empha-

sized the benefits of nonviolent protest in effectively challenging nondemocratic

governments. Remaining nonviolent can increase popular support and lead to a

higher likelihood of generating a stable democratic government (Pinckney 2018;

Stephan and Chenoweth 2008). The use of violence, on the other hand, can harden

government resolve, result in arrests and repression, raise the costs for future pro-

testors, and limit mobilizing potential.

1

Yet, despite the upsides of remaining

1

Department of Government and Politics, University of Maryland at College Park, MD, USA

2

Department of Political Science, Pennsylvania State University, the Behrend College, Erie, PA, USA

Corresponding Author:

Brandon Ives, Department of Government and Politics, University of Maryland at College Park, 3140

Tydings Hall, College Park, MD 20742, USA.

Email: bjives@umd.edu

Journal of Conflict Resolution

2020, Vol. 64(5) 958-986

ªThe Author(s) 2019

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/0022002719887491

journals.sagepub.com/home/jcr

peaceful and the downsides of turning violent, protests do sometimes escalate into

violence. Given the detrimental effects of violence and its high costs, what explains

why some individual protest events escalate to violence?

While substantial work in the contentious politics literature examines the strate-

gic escalation of tactics over the course of prolonged campaigns and social move-

ments, less attention has been paid to the mechanisms that lead individual nonviolent

protests to escalate into violence. Yet, better understanding of protest-level

dynamics is important for several reasons. Individual protest events can be consid-

ered the “building block” of larger movements. Many campaigns and movements

emerge organically from a small set of initial protests and can be radically shaped by

the character of their initial engagements. Violent escalation within a peaceful

campaign can potentially derail a movement. Moreover, not all protests occur within

the framework of larger strategic movements. Impromptu and disconnected protests

have not received as much attention within the broad “dynamics of contention”

research program that dominates much of the literature. Finally, understanding how

nonviolent protests escalate into violence allows us to expand the analysis of the

local and microlevel mechanisms of violence beyond the extensive work on strategic

and intentional riots.

In this article, we develop a general theory of violent escalation of a nonviolent

protest. We argue that two important factors—relative costs of violence and levels of

organization—shape why some protests escalate to violence. By relative costs of

violence, we refer to the costs (arrest, torture, death) of engaging in violence versus

the costs of remaining nonviolent. When the relative costs are low, protests attract

violence-oriented participants. By levels of prot est organization, we refer to the

presence of a command hierarchy and the overall density of relationships of the

participants. When protesters are only loosely connected and/or the protest lacks an

identifiable hierarchy, protest leaders struggle to constrain and guide participants.

As a result, the protest is more likely to break out into violence. Together, these

factors create a “gatekeeping effect,” in which the relative costs of violence shape

whether violence-oriented participants are drawn to protests, and the level of orga-

nization of the protest determines whether violence-oriented participants can be

constrained.

We test our theoretical expectations using statistical methods. Drawing from over

6,000 incidents of nonviolent protest across Af rica from 2000 to 2015, we find

strong support for our theory of escalation to violence. When the relative costs of

violence to nonviolence are low, as measured by recent repressive efforts by the

government, protests are significantly more likely to escalate into violence. We also

find that when protests are unorganized, escalation to violence is significantly more

likely to occur.

This article makes several contributi ons to the contentious politics lite rature.

First, we extend the literature by focusing on protest event-level dynamics rather

than campaigns. That is, some protest events are likely to have occurred in the

context of a broader campaign, while other protests occur outside of a more

Ives and Lewis 959