Content uploaded by Michael C. Zeller
Author content
All content in this area was uploaded by Michael C. Zeller on Sep 23, 2021
Content may be subject to copyright.
PATTERNS OF DEMOBILIZATION: A QUALITATIVE COMPARATIVE
ANALYSIS (QCA) OF FAR-RIGHT DEMONSTRATION CAMPAIGNS
*
Michael C. Zeller
†
Scholarship on social movement lifecycles has focused on mobilization processes, with
relatively less attention on the ends, demobilization. The intuitive connection between origins
and ends has sometimes led to a conceptualization of demobilization as simply the failure to
continue mobilizing, obscuring the distinct causal processes underlying demobilization. This
article adds to recent studies foregrounding demobilization by studying the negative
demobilization of large, far-right, demonstration campaigns. Using a subset from this popu-
lation of cases—campaigns in Germany, England, and Austria between 1990 and 2015—the
article applies qualitative comparative analysis (QCA) to this causally complex phenomenon. I
find that demobilizing is conjunctural, with evidence of four patterns: closing opportunity,
coercive state repression, civil countermobilization, and militant anti-far-right action. This
article addresses an important—and conspicuously ubiquitous—population of cases, far-right
demonstration campaigns and presents findings that reflect on critical issues in the study of far-
right sociopolitics.
Social movement mobilization processes have received much scholarly attention, demobilization
processes comparatively little. The obvious connection between these concepts—even the
parallelism of the terms, “mobilization” and “demobilization”—can encourage conceptualizations
of demobilization as a failure to mobilize. Yet it is a truth borne out repeatedly in empirical
observation: the endings of movement campaigns, organizations, fields do not simply mirror their
beginnings. Demobilization occurs through distinct causal processes—not merely a failure to
mobilize.
Notwithstanding the relative paucity of research, demobilization is covered in several recent
studies. At the microlevel of individual participation in movement activity, research from Fillieule
(2009, 2015) and from Gorski and Chen (2015) delve into individual disengagement and activist
burnout. To date, Davenport (2015) offers the most thoroughgoing theorization of organizational
demobilization—albeit one marked in some parts by induction from a single case study. Others
(e.g., Heaney and Rojas 2011; Lasnier 2017) have examined the demobilization of whole
movements or movement fields. In the areas of terrorism research (e.g., Cronin 2009) and studies
on repression, too, demobilization is covered, but often in an inescapably particularistic manner:
terrorism is skewed toward the circumstances of armed contention against the state, excluding
more common forms of movement activity; and repression, too often treated solely as the province
of the state (see Earl 2006), encompasses only one grouping of external demobilizing pressures.
Yet one form of demobilization is especially underexamined, that of campaigns. The scarcity
of research on campaign demobilization is surprising; part of the mesolevel of analysis, campaigns
are the crux of movement activity (Staggenborg and Lecomte 2009; Tilly 2004: 4), but little
research has focused on the ending of this activity. This article adds to research filling this lacuna
*
I thank Mobilization’s editor and anonymous reviewers, and especially Carsten Schneider for his invaluable com-
ments on the preparation of this article, particularly on the application of QCA. Additional comments from Fabian
Virchow and Andrea Krizsan were helpful in making QCA terminology more readily comprehensible. An early version
of this study was presented to the research seminar of the Hochschule Düsseldorf’s Forschungsschwerpunkt Rechts-
extremismus/Neonazismus (FORENA) and to the Central European University’s Center for Policy Studies.
†
Michael Zeller is a doctoral candidate in the Department of Political Science at Central European University and head
of the Organization Research Unit at the Centre for the Analysis of the Radical Right (CARR). Please direct corres-
pondence to zeller_michael@phd.ceu.edu.
© 2021 Mobilization: An International Quarterly 26(3): 267-284
DOI 10.17813/1086-671X-26-3-267
Mobilization
268
(Demirel-Pegg 2015, 2017; Zeller 2021) by examining the negative demobilization of a subset of
demonstration campaigns. The article applies qualitative comparative analysis (QCA) to this
causally complex phenomenon. It provides a model for further cross-case study sorely needed in
social movement research, which is saturated with case studies that provide rich empirical data
but scant capacity for generalization. Moreover, by examining large, far-right, demonstration
campaigns, the article addresses an important—and conspicuously ubiquitous—population of
cases, and presents findings that reflect on key issues in the study of far-right sociopolitics.
The article proceeds as follows: the next section puts forward a theorization of social move-
ment campaign demobilization and applies that theory to the population of large, far-right,
demonstration campaigns. A subset of this population, from Germany, England, and Austria
between 1990 and 2015, provides the empirics for the study. The article reviews the data and
methods employed, giving particular attention to the collection of protest event data, how this is
used to form a dataset of campaigns (consisting of series of protest events), and how QCA is an
especially suitable analytical tool. The article proceeds to the QCA analysis and parses the
meaning of the results to reveal four patterns of demobilization that can be termed (1) closing
opportunity, (2) coercive state repression, (3) civil countermobilization, and (4) militant anti-far-
right action. It concludes by discussing several tentative conclusions and lines for further research.
THE POPULATION OF LARGE, FAR-RIGHT DEMONSTRATION CAMPAIGNS
Broadly, demobilization is the process whereby social movement activity—of whole fields, of
organizations and activists, or of particular campaigns—decreases and ultimately ends. Note that
this overarching conception encompasses decline (a processual phenomenon) and cessation (a
discrete outcome). Davenport (2015: 22) provides a crucial distinction in his theorization of
organizational demobilization, between positive and negative demobilization. The former denotes
demobilization resulting from “winning,” a perceived success of activity; the latter, “collapse,
implosion, hindrance, or explosion.” In other words, there are three discrete sets that cover the
universe of potential developments of movement campaigns: (1) non-demobilization (i.e.,
continued mobilization), (2) positive demobilization, and (3) negative demobilization. This last
form, unintended, undesired demobilization is the focus of the present article. In set-theoretic
terms, therefore, the outcome of interest is negative demobilization whereas the non-outcome (i.e.,
not negative demobilization) encompasses both non-demobilization and positive demobilization.
Scholars have conceptualized demobilization at macro- (e.g., Tarrow 2011: 190) and micro-
levels (e.g., Fillieule 2015: 278). The intermediate, mesolevel of analysis has received less
attention. Davenport (2015) offers a valuable study and theorization of organizational de-
mobilization. But the demobilization of campaigns remains an understudied part of the mesolevel
of analysis. Social movement campaigns consist of four essential elements: (1) a constant
organizing actor (typically a movement organization, either solely or in coordination with others),
(2) temporal boundedness, (3) strategically linked actions (i.e., a series of events), and (4) the
intention to advance movement goals (Staggenborg and Lecomte 2009).
Campaigns are how movements move, but how that motion stops or changes direction
remains unclear. Existing scholarship addresses campaign demobilization somewhat obliquely.
Theorization on protest policing is especially noteworthy: Della Porta and Reiter (1998: 4) list
some of the central variables within protest policing phenomena. Yet study in this area seldom
focuses on demobilization per se—often it is the underexamined outcome of what research
sometimes casts as a monocausal process. Koopmans’s (1997) insightful study teases out aspects
of protest policing, drawing a crucial distinction between situational and institutional repression—
but it is situated at the macrolevel wherein policing and repression have an effect on society-wide
levels of mobilization. For all their merits, such studies obscure demobilization processes and tend
to fall into the trap of framing demobilization as the inverse of mobilization, rather than a separate
set of phenomena characterized by different causal processes.
Patterns of Demobilization
269
Earl’s (2006) conceptualization of social control is more suitable to study of demobilization.
Earl (2004, 2006) differentiates social control along three dimensions
1. Identity of repressive agent: (a) state agents closely connected to national political elites
(i.e., national state agents); (b) state agents distantly connected to national political elites
(i.e., local state agents); (c) private agents.
2. Character of repressive action: (a) coercion—direct repression; the threat or use of force;
(b) channelling—indirect repression (e.g., resource deprivation, problem depletion).
3. Visibility: (a) overt/manifest—observed, explicit, obvious repressive action; (b) covert/
latent—unobserved, concealed, veiled repressive action.
This conception provides two significant advantages. First, it moves away from the exclusive
focus on repressive action by the state to include manifestations of social control from private
(i.e., nonstate) agents. These too can exert demobilizing pressure. Second, Earl’s conceptual-
ization provides for the conjunctural nature of demobilization: whether one form of social
control—say, overt coercion from local state agents—materialises does not exclude any other—
such as latent channelling from private agents. Indeed, inspecting cases of campaign de-
mobilization reveals that, rather than monocausal and linear patterns, demobilizing factors
typically manifest in combinations. Conjunctions of causal factors produce the outcome of
demobilization.
1
Extant research focused on campaign demobilization bears out this causal nature. Demirel-
Pegg (2015) provides case studies of protest campaign demobilization, both developing causal
mechanisms of demobilization processes: by “brutal and indiscriminate repression” and by
“critical events” that alter the strategic opportunities available to campaign organizers (2017). In
both cases, the main causal factor represents only the most crucial in a conjunction of conditions.
Zeller (2021) identifies a causal mechanism of coercive countermobilization triggering state social
control and producing demobilization. These studies provide depth and rich analysis—but, like
all single case studies, have limited potential for generalizable inference.
Other research is less useful, if only for its glancing attention to demobilization. For instance,
in an otherwise illuminating article, which recognizes the conjunctural quality of campaign
outcomes, Huang and Sun (2019: 419) give very short shrift to protest campaign failure. In their
analysis of contentious campaigns, Chenoweth and Stephan (2011) only inspect failed cases as a
way of enhancing their analysis of successful ones. This is typical of much research, eschewing
analysis of failure and demobilization in favor of mobilization and success. Thus, existing research
reveals little consideration of campaign demobilization and (the perennial problem in social
movement studies) a general dearth of cross-case study.
Large, Far-Right, Demonstration Campaigns
Attempting to derive a general theory of campaign demobilization would be misguided;
campaigns are too varied in their form and surrounding context, and the process by which
campaigns demobilize are too multitudinous.
2
Given the impossibility of a general model of
campaign demobilization, the goal of scholarship in this area should be to examine subsets from
that vast universe of cases. These subsets should, first, explicitly represent specific populations in
themselves so that findings about demobilization processes (from within-case studies) and
patterns (from cross-case studies) are set in a meaningful empirical constellation. Second, subsets
ideally would offer connections to other, similar populations; rather than being particularistic,
findings should relate to other cases of demobilization processes.
This study examines a subset that satisfies these criteria: large, far-right, demonstration
campaigns. Demonstrations
3
are, as King and Soule (2007: 415) assert, the “quintessential tactic”
of social movements. That whole campaigns consist of series of demonstrations is not surprising.
They serve several important purposes for movement organizations: raising awareness, attracting
new members, keeping existing members connected, promoting group solidarity, and facilitating
networking. Beyond such instrumental ends, demonstrations provide meaningful references to
Mobilization
270
past events, preceding mobilizations within social movements, and can fulfill other symbolic
functions. They are exhibitions of movement strength and the centerpiece of the contentious
repertoire in modern society (Tarrow 2011; Tilly 2008). As such, demonstration campaigns are a
meaningful—though still overlarge—subset of campaigns to study demobilization processes.
Demonstration campaigns are particularly common and particularly important within the far-
right movement field. Almost certainly, demonstrations serve all or most of the instrumental
purposes listed above, but for the far right, symbolic motivations are perhaps even more sig-
nificant. Marching down the main thoroughfare of a city or town, or convening a large rally harks
back to the far right’s “glorious past”: the fascist regimes of the interwar years and their mass
displays of martial pomp. By demonstrating far-right movements boldly claim their space in the
public sphere, unwilling to accept—as most far-right movements had to after the Second World
War—a more furtive existence. As a result, there is typically not much internal pressure to
abandon demonstration campaigns. Far-right demonstration campaigns have a common property
of inertia: not tending to stop or change unless acted on by an outside force. For these reasons,
far-right demonstration campaigns are particularly suitable to examine causal processes related to
external demobilizing factors, both affording opportunities of generalizing to other populations
and representing in itself an important population of cases.
“Large” is an ancillary scope condition that reinforces the focus on external demobilizing
factors. Larger campaigns attract more attention from other actors, including from the state (Biggs
2018; McCarthy, Mcphail, and Smith 1996). The threshold of what constitutes a large demon-
stration is vague. This study refers to a mobilization of at least 1000 participants at the zenith of a
campaign, which is a threshold that is often recognized (e.g., Huang and Sun 2019; McCarthy,
McPhail, and Smith 1996), but the qualitative distinction is certainly not that clear. In any case,
because of the relationship between demonstration campaign size and the attention garnered, it is
likely that there is a qualitative difference between the demobilization of large and smaller
demonstration campaigns.
In this subset of cases, negative demobilization takes on one of two clear manifestations.
First, a demonstration campaign ceases its demonstrations. If the previously regular demon-
strations no longer occur, the campaign has demobilized. Second, demonstration campaigns live
and die by the participants mobilized for the event; in some respects, it does just come down to
numbers (Biggs 2018; Denardo 1985). So when a demonstration campaign experiences decreased
participation, it undergoes a degree of negative demobilization—all the more so when the decrease
is large and sustained over successive demonstrations.
DATA AND METHODS
Cases of large, far-right, demonstration campaigns were drawn from three countries, Germany,
England,
4
and Austria, between 1990 and 2015. Far-right demonstration campaigns are strikingly
ubiquitous; they mobilize and manoeuvre in most societies that are sufficiently liberal. Indeed,
general openness for mobilization, a minimally permissive opportunity structure, is necessary for
far-right movement organizations to emerge. Liberal societies’ tolerations and rights succour
demonstrations of many doctrinal stripes. Unsurprisingly, then, the liberal societies of Europe are
commonly distinguished by the presence of boisterous and sometimes potent far-right movements
and their demonstration campaigns. Together, within the broader grouping of European liberal
democracies, the geographic bounds of Germany, England, and Austria represent an illuminating
cross-section of contexts: on the bases of far-right party and movement sector strength, Michael
Minkenberg (2013a) distinguishes these countries by strong parties and weak movements
(Austria), weak parties and moderately strong movements (western Germany), and weak parties
and strong movements (eastern Germany
5
and the United Kingdom, see table 1). Furthermore, the
development of specific instruments to address far-right activism and demonstrations (Germany
and Austria), and the absence thereof (England) varies the presence of forms of state social
control. This reduces methodological problems that stem from limited diversity.
Patterns of Demobilization
271
Table 1. Country Contexts of Demonstration Campaigns.
Far-Right Movement Strength
Low
Medium
High
Far-Right
Party
Strength
Strong
Austria a
Weak
Germany (west) a
Germany (east) a
England
a The country contexts in bold typeface have specific legal instruments to address far-right activism and demonstrations.
Adapted from Minkenberg (2013a: 12).
The temporal bounds of 1990 and 2015 delimit an important and particularly energetic period
of far-right activity. A rising tide of far-right mobilization in the public sphere followed the
disintegration of the Soviet Union. As Minkenberg (2013b: 12) writes, “the notion that the
mobilization of the radical right or xenophobic movements often occurs in times of accelerated
social and cultural change provides a fruitful starting point for explaining relevant trends both in
Eastern Europe after 1989 and in Western Europe before and after that momentous year.” The
spur of changes in the late 1980s and early 1990s coincided with renewed far-right mobilization
in many European countries—not least in Germany, after the reunification of East and West.
Similarly, the end of 2015 coincided with a shift of far-right activism in Europe. The
following year, 2016, saw the far right celebrate the results of the United Kingdom’s vote to leave
the European Union and, in the United States, the election of Donald Trump.
6
More specifically,
2016 marked a clear shift in England, where the government used powers of the Terrorism Act of
2000 to ban a far-right organization (the so-called “National Action” group) for the first time since
the Second World War. In Germany, large inward migration peaked in the refugee crisis of 2015;
politicized to a phenomenal degree, it prompted a wave of far-right demonstrations and other
forms of mobilization. This wave largely subsided—apart from the persistent campaign of
PEGIDA (Patriotische Europäer gegen die Islamisierung des Abendlandes, or Patriotic Euro-
peans Against the Islamisation of the Occident) in Dresden—as the Alternative für Deutschland
(AfD) party radicalized and provided an electoral voice for far-right grievances. Austria, too, was
gripped by politicized immigration issues in 2015; the governing coalition of centrist parties was
troubled by internal strife and decreasing popularity, the start of a development that yielded the
2017 government of a more conservative center-right party in coalition with the Freiheitliche
Partei Österreichs (FPÖ).
Drawing from these three countries yields a set of thirty-two cases of large, far-right,
demonstration campaigns (see figure 1 and appendix 1). The method of fuzzy-set QCA (see
below) enables a meaningful analysis of this midrange number of cases. But it also demands
detailed empirical knowledge of each case—no mean feat—in order to perform the iterative
process, the “back and forth between ideas and evidence” that is essential to qualitative research
(Ragin 2008: 78). The campaigns vary considerably in length and number of participants (above
the threshold of mobilizing more than 1000 people at least once), but they all represent far-right
movement campaigns consisting of series of demonstrations.
Protest Event Data to Campaign Dataset
Demonstration campaigns are a unit of analysis composed of several protest events, so
assembling these data began with the creation of a protest event dataset (PED).
7
A search for large,
far-right demonstrations in the specified countries and period was undertaken. This search
provided the basis for targeted data collection on associated events: demonstrations by the same
far-right organizer(s). The project adopted a “blanketing strategy” (Hutter 2014) for this data
collection: information on events was drawn from multiple sources, chiefly police and state
security agency records, as well as local and national newspaper coverage, and information from
involved social movement actors. In all, this yielded a PED of nearly 500 manually coded demon-
stration events.
Mobilization
272
Figure 1. Timeline of Demonstration Campaign Cases in Descending Order of Size
Note: * denotes campaign initiation before the period of interest, ** denotes campaign continuation beyond period of
study (i.e., past 2015), *** denotes campaign initiation before and continuation after the period of interest.
PED was grouped into campaign data based on continuity of organizing actor and topical
focus.
8
In some instances, the resultant campaigns manifest in annual demonstrations in the same
(or nearby) locations; in others, more temporally condensed campaigns focusing on some pressing
topic or development. Most campaigns (twenty-five of thirty-two) occurred in Germany, where
far-right organization and mobilization has long been exceptionally robust; three of the four cases
from Austria
9
are initiatives of far-right fraternities (Burschenschaften), while the other is a
commemoration of World War II soldiers (a form mirrored in several campaigns in Germany);
and in England, where the far-right scene is typically smaller and more fractious, campaigns were
organized by the British National Party (BNP) and the English Defence League (EDL).
Campaign Data in Qualitative Comparative Analysis (QCA)
Research in various areas of social movement studies indicate several distinct causal
characteristics of demobilization. First, demobilization results from conjunctural causation
(Schneider and Wagemann 2012); it is the product of a causal process of combined conditions.
Second, demobilization is equifinal (ibid.); that is, there are multiple pathways to demobilization,
multiple demobilizing processes. Third, demobilization processes are asymmetrical (ibid.),
meaning that the absence of conditions leading to the outcome may not lead to the absence of the
outcome. Furthermore, this implies multifinality (ibid.): causal conditions of demobilization, such
as forms of social control, may be causally relevant for demobilization and non-demobilization.
(The inconsistent effects of repression—by turns, deterring and inciting movement activism—
epitomises multifinality.) In other words, demobilization is characterized by causal complexity.
Set-theoretic methods are well suited to these ontological notions of causality; and QCA
provides a means of cross-case study of causally complex phenomena. This study employs fuzzy-
Patterns of Demobilization
273
set QCA (fsQCA),
10
which allow the cases’ membership in sets to be partial (i.e., representing
degrees of the concept of interest), to incorporate dimensions of time into the study—avoiding
temporally flattening out campaign data. As explained below, the rationale for this is that causal
conditions will only exert causal force when they are proximate to the case outcome and (to a
lesser extent) when they manifest repeatedly.
In this study, the outcome of interest is negative demobilization (ND). Calibrating this set, in
line with the aforementioned definition of negative demobilization, consists of a composite of two
sets: (1) observation or not (i.e., crisp set) of ceased demonstration events where the organizing
actor has not had some demand or raison(s) d’être satisfied (which would signify positive
demobilization); (2) observation of decreased demonstration size in a campaign (i.e., fuzzy set)
over repeated events. Given the importance of sheer participation numbers to demonstration
campaigns, large and sustained decreases in demonstrator numbers represents a degree of negative
demobilization. Like asserting 1000 participants constitutes “large” for campaigns, the threshold
for negative demobilization by participatory decline is somewhat arbitrary. Yet if a campaign that
once mobilized more than 1000 participants now mobilizes fewer than 100 participants in
successive events, it is fair to say this fully represents the concept of negative demobilization (i.e.,
full set membership, 1). If fewer than 200 participants (but more than 100), then this partially
represents negative demobilization (i.e., 0.67 set membership). If fewer than 500 participants (but
more than 200), then this somewhat represents negative demobilization (i.e., 0.33 set mem-
bership). And if participation does not drop below 500, then it is not justified to speak of
participation-based negative demobilization. As ever with fuzzy sets, the qualitative anchor of 0.5
divides between cases that represent the concept, even if imperfectly (> 0.5), and cases that do not
represent the concept, even if traces of it manifest (< 0.5). Within the composite outcome set,
cessation of demonstration events takes precedence; if demonstrations have ceased, whether
participation has decreased is irrelevant. In absence of that distinct negative demobilization,
decreased participation may qualify. Demonstration campaigns in which events continued past
2015 were coded as non-negative demobilization.
Four conditions (summarized in table 2, next page) are derived from Earl’s conceptualization
of social control with the intention of isolating different forms of counteraction from different
agents. Each of these conditions exact a causal force most when they are, chiefly, proximate to
the outcome and, secondarily, repeated. These characteristics allow for the fuzzification of the
four social control conditions with the direct assignment method of calibration: proximate and
repeated observable manifestations represent full set membership (1); proximate but not repeated
observable manifestations, more in than out (0.67); repeated but not proximate observable
manifestations, more out than in (0.33); and no or only isolated observable manifestations that are
not proximate to the outcome (0). The four-value set is “especially useful in situations where
researchers have a substantial amount of information about cases, but the nature of the evidence
is not identical across cases” (Rihoux and Ragin 2009: 90).
State channelling social control (SCH) refers to measures by state actors that restrict the
operation of the demonstration campaign but stop short of invoking the state’s coercive apparatus
to prohibit demonstrations.
11
SCH may include banning certain locations or speakers for the
campaign, bans on organizations involved in the demonstration campaign, and new laws (national,
regional, or local) that diminish the organizer’s opportunity for action—state actions that do not
preclude demonstration events per se but do encumber them.
State coercion social control (SCO) denotes use of the state’s coercive powers or threat of
violence: banning demonstration events, arrest and detention of numerous demonstrators.
12
SCO
is typically legally ordered or permitted police enforcement or prevention measures.
Private channelling social control (PCH) refers to nonstate
13
(i.e., the initiative of private
actors) actions that restrict the operation of the demonstration campaign, but without resorting to
coercive measures. PCH often takes the form of simultaneous counterdemonstrations that seek to
meet the demonstrative power of the far-right campaign with a countervailing public presence, or
to undermine the attention-seeking far-right demonstration with a larger, more appealing event,
as well as disruptive tactics, such as when counterdemonstrators play loud, boisterous, discordant
Mobilization
274
music during a far-right group’s solemn memorial march. PCH also includes private agents
making legal challenges or lobbying for legal changes; these efforts likewise attempt to deprive
far-right campaigns of resources or opportunities essential to their demonstrations.
Private coercion social control (PCO) denotes the presence of nonstate coercive action:
private violence or threat of violence. PCO of course includes instances where private agents
violently engage far-right demonstrators; but it also covers instances where they blockade far-
right march routes or demonstration venues: this too constitutes a physical confrontation intended
to prevent a far-right demonstration.
In addition to those four conditions, the QCA model includes a temporal condition. Long
duration (LD) refers to demonstration campaigns lasting for ten or more events (i.e., a crisp set).
(Three events is considered minimum to constitute a demonstration campaign; up to nine events
are arguably midrange campaigns. This threshold was checked with robustness tests, which
revealed that minor alterations to this threshold did not significantly affect the QCA results.)
Unlike the other causal conditions, which existing theory suggests should contribute to negative
demobilization, long duration is more ambiguous. Far-right campaigns may grow and thrive as
long continuation begets senses of tradition and ritual. Annual gatherings and commemorations
are very much in this mold. Yet long duration may also constitute a demobilizing pressure,
especially when campaign events are rapidly recurrent, demanding significant personal invest-
ment from activists. Thus, whereas the four preceding conditions have positive directional ex-
pectations, long duration cannot be expected to contribute to negative demobilization or non-
negative demobilization.
Table 2. Calibration Summary of Conditions and Outcome.
Set Label
Abbr.
Set Scores and Empirical Manifestation
State Channelling
SCH
0 - zero or only one observation of SCH that is not proximate to outcome
0.33 - repeated observations of SCH but not proximate to outcome
0.67 - one observation of SCH that is proximate to outcome
1 - repeated and proximate to outcome observations of SCH
State Coercion
SCO
0 - zero or only one observation of SCO that is not proximate to outcome
0.33 - repeated observations of SCO but not proximate to outcome
0.67 - one observation of SCO that is proximate to outcome
1 - repeated and proximate to outcome observations of SCO
Private Channelling
PCH
0 - zero or only one observation of PCH that is not proximate to outcome
0.33 - repeated observations of PCH but not proximate to outcome
0.67 - one observation of PCH that is proximate to outcome
1 - repeated and proximate to outcome observations of PCH
Private Coercion
PCO
0 - zero or only one observation of PCO that is not proximate to outcome
0.33 - repeated observations of PCO but not proximate to outcome
0.67 - one observation of PCO that is proximate to outcome
1 - repeated and proximate to outcome observations of PCO
Long Duration
LD
0 - campaign lasting fewer than 10 events
1 - campaign lasting 10 or more events
Negative
Demobilization
ND
0 - no observations of negative demobilization (continued mobilization or
positive demobilization)
(Outcome)
0.33 - somewhat diminished participation in demonstration campaign,
fewer than 500 participants (but more than 200) at events
0.67 - significantly diminished participation in demonstration campaign,
fewer than 200 participants (but more than 100) at events
1 - observation of ceased demonstration events (without observations of
positive demobilization) or nearly total lost participation (i.e., fewer than
1000 participants at demonstration events)
Patterns of Demobilization
275
QCA RESULTS
Analysis of Necessity
Following recommended QCA best practice, this study conducted a test of necessity prior to
analysis of sufficiency (Schneider and Wagemann 2012: 278). First, each condition was tested.
No single condition proved necessary for the outcome of negative demobilization, which is
consistent with most theorization on demobilization. Then, a test for necessity of disjunctions (i.e.,
of combinations of conditions) was conducted. One should also note that disjunctions “tend to be
big sets in which most (and sometimes all) cases are members” (Schneider 2018: 248). In other
words, disjunctions may not actually be necessary—just exceedingly common. Researchers
should therefore attend to the Relevance of Necessity (RoN) measure.
The disjunction of state coercion or private channelling (denoted as: SCO + PCH) passes the
common consistency threshold of 0.9 with a score of 0.927; the RoN is 0.594, which is passable.
The necessity of this disjunction is plausible: private channelling against a far-right campaign or
state coercion could be needed to bring about negative demobilization. However, plotting this
disjunction (figure 2) facilitates a closer inspection of the supposed necessity relation. The plot
gives lie to the necessity of state coercion or private channelling: more than half of all inconsistent
cases (i.e., the second Wehrmachtsausstellung and the Ulrichsberg-Gedenkfeier campaigns) are
deviant cases for consistency in kind, which directly contradict the statement of necessity
(Schneider and Rohlfing 2013; Schneider 2018: 247). Thus, no atomic conditions, nor any dis-
junctions of conditions are necessary for the outcome of negative demobilization.
Figure 2. Necessity Plot of SCO + PCH.
Analysis of Sufficiency
In QCA truth tables are the central analytic tool. They allow “researchers to visualise and
analyse central features of causal complexity, such as equifinality or conjunctural causation and
the presence of INUS or SUIN conditions” (Schneider and Wagemann 2012: 9). In a truth table
(table 3 on the next page), each column denotes a different set (either causal condition or out-
come); “each row denotes a qualitatively different combination of conditions, [that is], the dif-
ference between cases in different rows is a difference in kind rather than a difference in degree”
(Schneider and Wagemann 2012: 92, emphasis in original). The truth table is sorted by outcome
Mobilization
276
Table 3. QCA Truth Table.
SCH
SCO
PCH
PCO
LD
OUT
n
Incl
PRI
Cases
8
0
0
1
1
1
1
3
1.000
1.000
BNP Red, White and Blue festival;
Trauermarsch Dresden,Wiener
Korporations Ball
31
1
1
1
1
0
1
2
1.000
1.000
2. Waffen-SS commemoration;
Heidenau hoert zu
4
0
0
0
1
1
1
1
1.000
1.000
2. Wehrmachtsausstellung
5
0
0
1
0
0
1
1
1.000
1.000
Lichtellaeufen Schneeberg
9
0
1
0
0
0
1
1
1.000
1.000
Sachsentag (Sommerfest)
11
0
1
0
1
0
1
1
1.000
1.000
1. Waffen-SS commemoration
18
1
0
0
0
1
1
1
1.000
1.000
Ulrichsberg-Gedenkfeier
21
1
0
1
0
0
1
1
1.000
1.000
HoGeSa
22
1
0
1
0
1
1
1
1.000
1.000
Pressefest der Deutsche Stimme
23
1
0
1
1
0
1
1
1.000
1.000
AN Antikriegstag
24
1
0
1
1
1
1
1
1.000
1.000
Bad Nenndorf Trauermarschen
27
1
1
0
1
0
1
1
1.000
1.000
1. Hess Gedenksmarsch
29
1
1
1
0
0
1
1
1.000
1.000
2. Hess Gedenksmarsch
7
0
0
1
1
0
1
3
0.876
0.859
Fest der Voelker; Magdeburg Bombing
Commemoration; Mourn Lee Rigby
6
0
0
1
0
1
0
7
0.556
0.530
Deutsche Volksunion Congress;
PEGIDA Dresden; Legida; Linzer
Burschenbundball; Mittenwald
Gebirgsjaeger Pentecost; EDL rally;
Freital steht auf
12
0
1
0
1
1
0
1
0.248
0.248
1. Wehrmachtsausstellung
1
0
0
0
0
0
0
2
0.198
0.000
Freigeist; REP
Aschermittwochsveranstaltung
3
0
0
0
1
0
0
2
0.000
0.000
Tag der deutschen Zukunft; Wien
Akademiker Ball
2
0
0
0
0
1
0
1
0.000
0.000
Einsiedel sagt Nein
10
0
1
0
0
1
?
0
-
-
13
0
1
1
0
0
?
0
-
-
14
0
1
1
0
1
?
0
-
-
15
0
1
1
1
0
?
0
-
-
16
0
1
1
1
1
?
0
-
-
17
1
0
0
0
0
?
0
-
-
19
1
0
0
1
0
?
0
-
-
20
1
0
0
1
1
?
0
-
-
25
1
1
0
0
0
?
0
-
-
26
1
1
0
0
1
?
0
-
-
28
1
1
0
1
1
?
0
-
-
30
1
1
1
0
1
?
0
-
-
32
1
1
1
1
1
?
0
-
-
(OUT) and inclusion (Incl.). The nonsequential row numbers in the first column reflect the basic
ordering of the configurations, from no present conditions (i.e., row 1 consists only of zeroes) to
all conditions present (i.e., row 32 consists of all ones. See Dusa 2019). Furthermore, PRI means
“proportional reduction in inconsistency” and indicates relevance, that is, “how much it
[analytically] helps to know that a given X is specifically a subset of Y and not a subset of ~Y”
(Schneider and Wagemann 2012: 242).
This sorting shows which rows (or combinations of conditions) co-occur with the outcome
and how consistently; inclusion, in other words, shows how some rows, like row six, contains
cases that have and do not have the outcome. Empirical cases can be represented by one—and
Patterns of Demobilization
277
only one—row, or combination of conditions. This is true for cases with fuzzy set membership,
too, because the qualitative anchor of 0.5 separates set members and nonmembers: a case with
0.67 membership in a set falls into a truth table row where that set is present (1); a case with 0.33
membership, into a row where the set is absent (0). Thus, the campaign cases are sorted into the
rows that reflect conditions in their case: the n column records the number of cases in each row;
the cases column lists them. Rows that represent an unobserved combination of conditions have
no cases in them (i.e., the bottom thirteen rows); the outcome is uncertain (thus, the OUT column
records a “?”) in such instances because there are no empirical observations.
14
Take note of row six, which is clearly below the raw consistency threshold: that is, within
this row there are cases that are part of the outcome set (i.e., Deutsche Volksunion Congress,
Legida, EDL rally, and Freital steht auf) and cases that are nonmembers of the outcome set (i.e.,
PEGIDA Dresden, the Linzer Burschenbundball, and the Mittenwald Gebirgsjäger Pentecost).
Given the high degree of inconsistency, this row is excluded from the logical minimisation process
(i.e., the derivation of a sufficiency solution) (Schneider and Wagemann 2012: 121–22).
To produce an intermediate sufficiency solution
15
from the truth table, the study applies the
“directional expectations” (see Schneider and Wagemann 2012: 168–74) about the causal
influence of conditions in the model. (Recall from above, each form of social control is expected
to contribute to demobilization processes; “long duration” is ambiguous, suggesting no directional
expectation.) Boolean minimization of the truth table yields four configurations of conditions that
generate negative demobilization:
𝑆𝐶𝐻 ∗𝐿𝐷 +𝑆𝐶𝑂 ∗∼ 𝐿𝐷 +𝑃𝐶𝐻 ∗∼ 𝐿𝐷+∼ 𝑆𝐶𝑂 ∗𝑃𝐶𝑂 ∗𝐿𝐷 => 𝑁𝐷
This mathematical representation reads as follows: state channelling (SCH) and long duration
(LD) or state coercion (SCO) and not long duration (~LD) or private channelling (PCH) and not
long duration (~LD) or not state coercion (~SCO) and private coercion (PCO) and long duration
(LD) are sufficient to produce negative demobilization. Each of the four solution terms, separated
by the logical OR (denoted by a “+”), represents a sufficient configuration of conditions. Table 4
displays the coverage
16
and consistency
17
of the terms within the solution, as well as the cases
covered by each term. Figure 3 provides a graphical illustration of the solution for negative
demobilization. Figure 4. presents a plot of this solution.
Table 4. Intermediate Sufficiency Solution to Negative Demobilization (Unique cases bolded)
Solution
term
Higher-Order
Concept
Coverage
Consistency
Typical cases
SCH*LD
closing
opportunity
0.145
0.910
Pressefest der Deutsche Stimme; Bad Nenndorf
Trauermarschen
SCO*~L
D
coercive
state
repression
0.247
1.000
1. Hess Gedenksmarsch; Sachsentag
(Sommerfest); 1. Waffen-SS commemoration;
Heidenau hört zu; 2. Hess Gedenksmarsch; 2.
Waffen-SS commemoration
PCH*~L
D
civil
counter-
mobilization
0.363
0.962
AN Antikriegstag; Fest der Voelker; HoGeSa;
Lichtellaeufen Schneeberg; Mourn Lee Rigby; 2.
Hess Gedenksmarsch; 2. Waffen-SS
commemoration; Heidenau hört zu
~SCO*P
CO*LD
militant
anti-far-right
action
0.232
1.000
BNP Red, White and Blue festival;
Trauermarsch Dresden; Wiener Korporations
Ball; 2. Wehrmachsausstellung; Bad Nenndorf
Trauermarschen
[Solution]
0.840
0.967
Mobilization
278
Figure 3. Graphic Illustration of the Solution for Negative Demobilization.
The first configuration consists of state channeling and long enduring campaigns (SCH*LD).
It is suggestive of a well-known phenomenon in social movement studies: closing opportunity.
Basically, state authorities alter the legal strictures on demonstrations or affect the availability of
resources for performing demonstrations. The cases fitting within this configuration seem to bear
out this interpretation. In the Ulrichsberg-Gedenkfeier
18
case in Austria, which honored soldiers
killed in the Second World War (most prominently and controversially including Wehrmacht and
SS soldiers), the state withdrew the army’s participation in the event along with the support of
some other resources; participation numbers for the event shrank to no more than a couple hundred
by 2015. The Pressefest der Deutsche Stimme, a press junket for the newspaper of the far-right
National Democratic Party of Germany (Nationaldemokratische Partei Deutschlands), negatively
demobilized in the aftermath of revelations about terrorist acts by the so-called National Socialist
Underground, whereupon several German states enacted measures (including several
organizational bans [Zeller 2020]) that limited the mobilizing capacity of far-right groups and
activists. (Admittedly, though, the Pressefest demanded a lot of resources and mobilizing energy,
so internal movement pressures may be more important than external pressures.) The Bad
Nenndorf Trauermarsch campaign similarly evinces this causal configuration, though there is an
issue of overdetermination since it is also covered by another solution term (~SCO*PCO*LD).
Within-case study could disentangle these conditions and assess what truly had a causal effect.
State coercion and a not-long enduring campaign (SCO*~LD) comprise the second sufficient
term in the solution. This configuration seemingly accounts for coercive state repression: bans
and prohibitions, arrests, and prosecutions. For instance, in the case of the first Waffen-SS
commemoration campaign (in Halbe, Germany in the early 1990s), negative demobilization
occurred when local state authorities imposed a ban on the events and there was mass police
deployment to prevent any attempt by the far right to demonstrate. The INUS
19
condition “not of
long duration” (~LD) in this term is intriguing. Partially, it is an artefact of limited diversity: only
one truth table row representing both state coercion social control and long duration (i.e.,
SCO*LD) fits any case (row twelve and the first anti-Wehrmachtsausstellung campaign),
meaning there is no empirical evidence for the seven other rows representing this conjunction.
This could suggest that state authorities do not commonly apply coercion to large, far-right,
demonstration campaigns that are longstanding, and/or that coercion against such longstanding
campaigns is not effective in causing negative demobilization.
The third solution term denotes private channelling and not-long enduring campaigns
(PCH*~LD). This essentially amounts to civil countermobilization. Inspecting the cases covered
by this term, several are marked by the presence of other causal conditions that seem relevant to
Patterns of Demobilization
279
Figure 4. Sufficiency Solution a
Note: a Cases at point 1 are fully covered and consistent with the solution formula. The cases at point 2 are individually
irrelevant cases (IIR). Importantly, the deviant cases for coverage (Legida, Deutsche Volksunion Congress, EDL rally,
and Freital steht auf) are intriguing cases for further study because they are not covered by the solution formula.
Within-case study could uncover the causal conditions that explain negative demobilization in those campaigns.
the negative demobilization outcome. For instance, the second Hess Gedenkmarsch campaign,
which occurred in the first half of the 2000s and honored the memory of Rudolf Hess, a prominent
Nazi leader, negatively demobilized only after a new law criminalized glorification of the Nazi
regime (Virchow 2013). The law was certainly spurred on by private channelling efforts, wherein
residents from the location of the far-right campaign (Wunsiedel, Germany) lobbied national
politicians to adopt a prohibition against “Nazi glorification” (the “NS Verherrlichung stoppen”
campaign)—but other causal conditions (i.e., state channelling social control) were significant.
The combination of private channelling and not-long enduring campaigns may, therefore, signify
a causal trigger, which can spark several different negative demobilization processes.
The fourth solution term is composed of no state coercion, private coercion, and long duration
(~SCO*PCO*LD). Here, we see the application of physical confrontation, threatened or realized,
by nonstate agents against the far-right campaign. It is the militant anti-far-right action pattern of
negative demobilization. The cases covered all display the violence or “violence-ready”
(gewaltbereit) tactics of groups like Antifa and Autonomists, typified by blockades of far-right
demonstration sites and disruption of events by attacking police and/or far-right activists.
DISCUSSION
Taken together, the QCA results suggest several conclusions—but the provisionality of these
conclusions should be underscored at the outset. In all its myriad forms, demobilization is a
process. Longstanding assertions (e.g., Tarrow 2011) and recent studies (e.g., Demirel-Pegg 2015,
Mobilization
280
2017) in social movement scholarship hold that causal mechanisms underlie demobilization
processes. Inevitably, even mechanisms of the same type take on unique forms in each case, but
this fact should not lead us into notions of inescapable idiosyncrasy nor, by the same token, deter
searches for cross-case findings. With QCA, we can suss out meaningful causal patterns, but we
cannot thence confirm causal mechanisms. Follow-on process tracing—cases selected in the
manner prescribed by Schneider and Rohlfing (2013, 2016)—is essential to valid causal inference.
With that caveat in mind, the QCA results and re-examination of the analysed demonstration
campaigns point to some important findings. Each form of social control can affect negative
demobilization. That no causal condition is necessary nor solely sufficient reinforces the view of
demobilization processes as multifarious. It also supports presuppositions about demobilization’s
causal nature: it is conjunctural, occurring through the coming together of conditions
In several cases covered by different solution terms, it appears that private agents acted first—
even if subsequent state action was causally decisive. For example, the first Hess Gedenkmarsch
campaign (in Germany) is covered by the coercive state repression term (SCO*~LD), but within
this case coercive countermobilization by antifascist and autonomist activists was crucial in
triggering state action (Zeller 2021). In the Ulrichsberg-Gedenkfeier campaign (in Austria), the
QCA results suggest state channelling, closing opportunity, was decisive. Yet a clutch of counter-
demonstrators (organized in the Arbeitskreis gegen den kärntner Konsens, or “Working Group
against the Carinthian Consensus”) had begun to protest against this event years before the
national government forbade the Austrian army’s participation and before the government
withdrew subsidies for transporting older would-be demonstrators up a mountain to the event
location. Whether due to inattention, indifference, or (particularly in cases where local state
authorities oppose the far-right campaign) insufficient capacity, it appears that sometimes private
forms of social control must first materialize to jolt the requisite state actor(s) out of lethargy or
spur on the requisite state action.
The results also confirm the effectiveness of coercive repression. Such a mild claim is
scarcely contestable. The QCA results, nevertheless, bear out this assertion and, going further,
suggest that both coercive state repression and militant anti-far-right action can bring about
negative demobilization. But inspecting the cases covered by the latter pathway reveals that
brawling, bashing, and “punching a fascist” is perhaps not what is needed. In the private actions
against the British National Party’s Red, White and Blue festival, the Dresden Memorial March
(Trauermarsch), the Bad Nenndorf Memorial March, violent clashes with the far-right
demonstrators or police were marginal, while blockades of march routes and event venues were
tactically central. To be sure, the far right (to whatever extent the possibility of violence
represented a concern and not a welcome invitation to fight) and police in these cases were well
aware of militant anti-far-right activists’ potential for violence; perhaps that, being menacing to a
certain degree, was the crucial characteristic, if not to prompt state action, at least to deter faint-
hearted far-right would-be demonstrators. The cases covered by the fourth solution term suggest
this might be so.
Lastly, the distribution of cases from different countries across the four solution terms is
revealing. German cases are covered in all the terms, but English cases are covered only in the
civil countermobilization (PCH*~LD) and militant anti-far-right action (~SCO*PCO*LD) terms;
the sole Austrian case of negative demobilization is also covered by the militant anti-far-right
action (~SCO*PCO*LD) term. In part, this clustering is a consequence of limited diversity: there
are only two English cases (the Mourn Lee Rigby and BNP Red, White and Blue festival
campaigns) and one Austrian case (the Wiener Korporations Ball campaign) with the full out-
come. Nevertheless, this clustering meets expectations about the demobilization processes in these
contexts. Neither in Austria nor in England was any state social control present. In Austria, this is
attributable to far-right party strength (see table 1): the FPÖ was a driving force behind the Wiener
Korporations Ball campaign, indeed sponsoring a successor campaign (the Wien Akademiker Ball
campaign). In England, the absence of state social control may be due to the lack of specific legal
instruments to deal with far-right activism or, relatedly, the state’s relatively noninterventionist
Patterns of Demobilization
281
posture toward far-right demonstrations. Within-case studies of the English and Austrian cases
could shed more light on the nature of state (in)activity in these instances.
Each of these provisional conclusions demands greater scrutiny, to verify and further specify
pathways and to identify causal mechanisms. More cross-case study could assess whether these
findings hold in other populations. This should form part of the continued development of
demobilization research. In particular, coincidence analysis (CNA, another set-theoretic method)
could be used to reveal cross-case sequencing patterns; further ethnographic study (e.g.,
Pilkington 2016) is sorely needed to assess micro-effects of some causal conditions, such as how
individual activists experience and react to forms of social control. This article’s QCA contributes
to the important research agenda on demobilization.
APPENDIX: OVERVIEW OF CASES
Campaign name
Location(s)
Far-right SMO(s)
Years
AN Antikriegstag
Dortmund
Autonome Nationalisten
2005-2013
Bad Nenndorf Trauermarschen
Bad Nenndorf
NPD and several smaller groups
2006-2015
Deutsche Volksunion Congress
Passau
DVU
1987-2001
PEGIDA Dresden
Dresden
Pegida (Dresden)
2014-
Fest der Völker
Jena, Altenburg
NPD
2005-2010
Freigeist
Sachsen
NPD
2015-2015
1. Wehrmachtsausstellung
Several German cities
NPD and several smaller groups
1997-1999
2. Wehrmachtsausstellung
Several German cities
NPD and several smaller groups
2001-2004
1. Hess Gedenksmarsch
Wunsiedel, nearby
towns
Coalition of far-right groups, led by far-right activists
including Michael Köhnen, Christian Worch, and
Jürgen Rieger
1988-1994
2. Hess Gedenksmarsch
Wunsiedel
Jürgen Rieger’s Aktionsbüro Norddeutschland and
other far-right groups
2001-2004
HoGeSa
Köln, Hannover,
Ludwigshafen
Hooligans gegen Salafisten (HoGeSa),
Gemeinschafts-Stark Deutschland eV
2014-2015
Legida
Leipzig
Legida
2015-2016
Lichtelläufen Schneeberg
Schneeberg
NPD (Sachsen)
2013-2014
Linzer Burschenbundball
Linz
Burschenschaft Arminia Czernowitz zu Linz
1948-
Magdeburg Bombing
Commemoration
Magdeburg
NPD and associated groups
2005-2015
Mittenwald Gebirgsjäger
Pentecost
Mittenwald
Kameradenkreis der Gebirgstruppe
1957-
EDL rally
Several English cities
English Defence League
2009-2014
Pressefest der Deutsche Stimme
Several towns in
eastern Germany
NPD
2001-2012
Red, White and Blue festival
Denby and elsewhere
in central England
British National Party
1999-2009
REP
Aschermittwochsveranstaltung
Geisenhausen
Die Republikaner
1998-2005
Mourn Lee Rigby
Various locations in
England
English Defence League
2013-2013
Sachsentag (Sommerfest)
Locations in Sachsen
Junge Nationaldemokraten
2007-2013
Tag der deutschen Zukunft
Several German cities
NPD, Die Rechte, III Weg
2009-2019
Trauermarsch Dresden
Dresden
Junge Landsmannschaft Ostdeutschland, NPD;
Aktionsbündnis gegen das Vergessen (AgdV)
2000-2013
Ulrichsberg-Gedenkfeier
Klagenfurt
Ulrichsberggemeinschaft
1958-2017
1. Waffen-SS commemoration
Halbe
coalition of far-right groups, led by far-right activist
Christian Worch
1990-1992
2. Waffen-SS commemoration
Halbe
Christian Worch’s Die Rechte, NPD
2002-2007
Wien Akademiker Ball
Wien
Wien FPÖ
2013-
Wiener Korporations Ball
Wien
Wiener Korporations-Ring
1952-2012
Freital steht auf
Freital
Bürgerinitiative Freital wehrt sich - Nein zum Hotelheim
2015-2015
Heidenau hört zu
Heidenau
Bürgerinitiative Heidenau hört zu, NPD
2015-2015
Einsiedel sagt Nein zur EAE
Chemnitz
Bürgerinitiative Einsiedel
2015-2016
Mobilization
282
NOTES
1
Davenport (2015: 39) approaches this in his explanation of “intersection of internal and external sources of
demobilization,” in which pairs of causal conditions, one internal and one external, combine to produce demobilization.
However, demobilizing pressures can and often do occur in more complex configurations, multiple internal and external
factors figuring in the process. It is the work of within-case study to disentangle these conditions and identify causal
mechanisms.
2
Davenport’s (2015) theorization of social movement organization demobilization is more successful—though still
problematic in some parts—because a focus on organizations establishes a less variable unit of analysis. Campaigns
are not like this; they are protean, adopting any number of forms.
3
Casquete (2006: 47) provides a good definition: “a collective gathering in a public space whose aim is to exert political,
social, and/or cultural influence on authorities, public opinion and participants through the [...] expression of an opinion
or demand.” I exclude “disciplined and peaceful” from Casquete’s definition because these are not essential to the
definition of demonstrations—there are plenty of undisciplined and/or nonpeaceful demonstrations.
4
The study specifies England rather than the United Kingdom or Britain so as not to include cases from Northern
Ireland, a unique context with singular dynamics, or contexts in which there are no cases of large far-right demonstration
campaigns (i.e., Scotland and Wales).
5
As noted below, the classification of eastern Germany as representing weak parties and strong movements is
appropriate before 2015. In the years since, the rise of the Alternative für Deutschland Party has transformed the socio-
political landscape; far-right movements remain strong in eastern Germany—but there is now a strong far-right party,
with significant connections to this movement scene.
6
Some research (e.g., Koopmans and Rucht 1995) suggests that far-right mobilization would subside after such
encouraging electoral results. Indeed, such a development may have begun with the 2019 election of Boris Johnson in
the U.K. (Parveen 2020).
7
Protest-event datasets have been criticized on several grounds. See Demarest and Langer (2019) for potential pitfalls—
not least of which is selection bias. This problem typically stems from sole reliance on one or a few national newspapers
for data. The present study avoids this issue by applying the “blanketing strategy” described below.
8
Attending to the consistency of topical focus ensures that demonstration events represent a series of strategically
linked actions (and could thus be grouped as part of one campaign), rather than merely being the coincident initiative
of the same far-right organizer; for social movement actors can operate multiple campaigns simultaneously.
9
The study excludes the case of the so-called Bleiburg Commemorations that take place annually in southern Austria.
While it is a large, far-right (at least partially) demonstration campaign, with the recurrent sponsorship of the Croatian
state and the Catholic Church the Bleiburg Commemorations are an archetypal unique case, characterized by singular
dynamics and processes.
10
Analyses were conducted with the QCA (Dusa 2019) and SetMethods (Oană and Schneider 2018) packages for R.
11
See Koopmans’s (1997) conception of “institutional repression.”
12
See Koopmans’s (1997) conception of “situational repression.”
13
In some cases, private channeling is linked to states—sometimes covertly sponsored by them (e.g., astroturfing).
14
These are called logical remainders. See Schneider and Wagemann (2012) on logical remainders and limited diversity.
15
An intermediate sufficiency solution—unlike parsimonious and conservative solutions—is based on simplifying
assumptions that are easy counterfactuals, that is, they accord with specified directional expectations. For further
explanation, see Schneider and Wagemann (2012).
16
Here, coverage refers to the relation in size of the solution term and the outcome set. In other words, coverage
“expresses how much of the outcome is covered” by the solution term (Schneider and Wagemann 2012: 325).
17
Here, consistency measures the degree to which one solution term is a subset of the outcome.
18
This case is not listed in table 4 because it is only a partial member of the outcome set. Nevertheless, it still represents
the outcome of negative demobilization (due to lost participation). See figure 4.
19
That is, “Insufficient but Necessary part of a condition which is itself Unnecessary but Sufficient for the result”
(Schneider and Wagemann 2012). In other words, INUS conditions are individual representations of conditions within
the terms that comprise a solution. For example, the solution derived in this study is
𝑆𝐶𝐻 ∗𝐿𝐷 +𝑆𝐶𝑂 ∗
)
𝐿𝐷 +𝑃𝐶𝐻 ∗
)
𝐿𝐷 +
)
𝑆𝐶𝑂 ∗𝑃𝐶𝑂 ∗𝐿𝐷 => 𝑁𝐷
. In that solution SCH is an INUS condition of the first term; LD is the other INUS
condition comprising that term.
REFERENCES
Biggs, Michael. 2018. “Size Matters: Quantifying Protest by Counting Participants.” Sociological
Methods and Research 47(3): 351–83.
Casquete, Jesus. 2006. “The Power of Demonstrations.” Social Movement Studies 5(1): 45-60.
Chenoweth, Erica, and Maria J. Stephan. 2011. Why Civil Resistance Works: The Strategic Logic of
Nonviolent Conflict. New York: Columbia University Press.
Cronin, Audrey Kurth. 2009. How Terrorism Ends: Understanding the Decline and Demise of Terrorist
Campaigns. Princeton, NJ: Princeton University Press.
Patterns of Demobilization
283
Davenport, Christian. 2015. How Social Movements Die. New York: Cambridge University Press.
Della Porta, Donatella, and Herbert Reiter. 1998. Policing Protest: The Control of Mass Demonstrations
in Western Democracies. Minneapolis: University of Minnesota Press.
Demarest, Leila, and Arnim Langer. 2019. “How Events Enter (or Not) Data Sets: The Pitfalls and
Guidelines of Using Newspapers in the Study of Conflict.” Sociological Methods and Research
47(1): 1–35.
Demirel-Pegg, Tijen. 2017. “The Dynamics of the Demobilization of the Protest Campaign in Assam.”
International Interactions 43(2): 175–216.
Demirel-Pegg, Tijen, and Scott Pegg. 2015. “Razed, Repressed and Bought Off: The Demobilization of
the Ogoni Protest Campaign in the Niger Delta.” Extractive Industries and Society 2(4): 654–63.
Denardo, James. 1985. Power in Numbers: The Political Strategy of Protest and Rebellion. Princeton:
Princeton University Press.
Dusa, Adrian. 2019. “QCA with R. A Comprehensive Resource.” Basel: Springer International.
Earl, Jennifer. 2004. “Controlling Protest: New Directions for Research on the Social Control of Protest.”
Research in Social Movements, Conflicts, and Change, Special Issue on Authority in Contention 25:
55–83. Bingley, UK: Emerald Group Publishing.
———. 2006. “Repression and the Social Control of Protest.” Mobilization: An International Journal
11(2): 129–43.
Fillieule, Olivier. 2009. “Disengagement from Radical Organizations. A Process and Multi-Level Model
of Analysis.” Pp 1-29 in Movements in Times of Transition, Bert Klandermans and Cornelius van
Stralen, eds. Minneapolis: University of Minnesota Press.
———. 2015. “Demobilization and Disengagement in a Life Course Perspective.” Pp. 277-88 in The
Oxford Handbook of Social Movements, Donatella della Porta and Mario Diani, eds. Oxford: Oxford
University Press.
Gorski, Paul C., and Cher Chen. 2015. “‘Frayed All Over:’ The Causes and Consequences of Activist
Burnout Among Social Justice Education Activists.” Educational Studies 51(5): 385–405.
Heaney, Michael T., and Fabio Rojas. 2011. “The Partisan Dynamics of Contention: Demobilization of
the Antiwar Movement in the United States, 2007-2009.” Mobilization: An International Journal
16(1): 45–64.
Huang, Ronggui, and Xiaoyi Sun. 2019. “Dual Mediation and Success of Environmental Protests in China:
A Qualitative Comparative Analysis of Ten Cases.” Social Movement Studies 19(4): 408–425.
Hutter, Swen. 2014. “Protest Event Analysis and its Offspring.” Pp 335-67 in Methodological Practices
in Social Movement Research, Donatella della Porta, ed.. Oxford: Oxford University Press.
King, Brayden G., and Sarah A. Soule. 2007. “Social Movements as Extra-Institutional Entrepreneurs:
The Effect of Protests on Stock Price Returns.” Administrative Science Quarterly 52(3): 413–42.
Koopmans, Ruud. 1997. “Dynamics of Repression and Mobilization: The German Extreme Right in the
1990s.” Mobilization: An International Quarterly 2(2): 149–64.
Koopmans, Ruud, and Rucht Dieter. 1995. “Social Movement Mobilization Under Right and Left Govern-
ments: A Look at Four West European Countries.” Berlin: Wissenschaftszentrum Berlin für Sozial-
forschung.
Lasnier, Virginie. 2017. “Demobilization and its Consequences: After the Russian Movement Za chestnye
vybory.” Europe - Asia Studies 69(5): 771–93.
McCarthy, John D, Clark Mcphail, and Jackie Smith. 1996. “Images of Protest: Dimensions of Selection
Bias in Media Coverage of Washington Demonstrations, 1982 and 1991.” American Sociological
Review 61(3): 478–99.
Minkenberg, Michael. 2013a. “From Pariah to Policy-Maker? The Radical Right in Europe, West and
East: Between Margin and Mainstream.” Journal of Contemporary European Studies 21(1): 5-24.
———. 2013b. “The European Radical Right and Xenophobia in West and East: Trends, Patterns and
Challenges.” Pp. 9–34 in Right-Wing Extremism in Europe. Country Analyses, Counter-Strategues
and Laboer-Marker Oriented Exit Strategies, Ralf Melzer and Sebastian Serafin, eds. Berlin:
Friedrich Ebert Stiftung Forum Berlin.
Oană, Ioana-Elena, and Carsten Q. Schneider. 2018. “SetMethods: An Add-on Package for Advanced
QCA.” The R Journal 10(1): 507-533.
Parveen, Nazia. 2020. “Threat from Far Right May be Receding since Tory Election Victory.” The
Guardian. March 29, 2020. Accessed October 13, 2020 https://www.theguardian.com/politics/2020/
mar/29/threat-from-far-right-may-be-receding-since-tories-election-victory
Pilkington, Hilary. 2016. Loud and Proud: Passion and Politics in the English Defence League.
Manchester, UK: Manchester University Press.
Mobilization
284
Ragin, Charles C. 2008. Redesigning Social Inquiry: Fuzzy Sets and Beyond. Chicago: University of
Chicago Press.
Rihoux, Benoît, and Charles C. Ragin. 2009. Configurational Comparative Methods. Qualitative Com-
parative Analysis (QCA) and Related Techniques. Thousand Oaks, CA: Sage.
Rohlfing, Ingo, and Carsten Q. Schneider. 2018. “A Unifying Framework for Causal Analysis in Set-
Theoretic Multimethod Research.” Sociological Methods and Research 47 (1): 37–63.
Schneider, Carsten Q. 2018. “Realists and Idealists in QCA.” Political Analysis 26(2): 246–54.
Schneider, Carsten Q., and Claudius Wagemann. 2012. Set-Theoretic Methods for the Social Sciences
(Strategies for Social Inquiry). Cambridge: Cambridge University Press.
Schneider, Carsten Q., and Ingo Rohlfing. 2013. “Combining QCA and Process Tracing in Set-Theoretic
Multi-Method Research.” Sociological Methods and Research 42(4): 559–97.
———. 2016. “Case Studies Nested in Fuzzy-set QCA on Sufficiency: Formalizing Case Selection and
Causal Inference.” Sociological Methods and Research 45(3): 526–68.
Staggenborg, Suzanne, and Josée Lecomte. 2009. “Social Movement Campaigns: Mobilization and
Outcomes in the Montreal Women’s Movement Community.” Mobilization 14(2): 163–80.
Tarrow, Sidney G. 2011. Power in Movement: Social movements and contentious politics. 3rd ed.
Cambridge: Cambridge University Press.
Tilly, Charles. 2004. Social Movements, 1768-2004. Boulder, CO: Paradigm Publishers.
———. 2008. Contentious Performances. New York: Cambridge University Press.
Virchow, Fabian. 2013. “Creating a European (Neo-Nazi) Movement by Joint Political Action?” Pp 197-
214 in Varieties of Right-Wing Extremism in Europe, Andrea Mammone, Emmanuel Godin, and
Brian Jenkins, eds. London: Routledge.
Zeller, Michael C. 2020. “Germany: Is Banning Far-right Groups Enough?” Open Democracy. Acessed
August 10, 2021 at https://www.opendemocracy.net/en/global-extremes/germany-banning-far-right-
groups-enough/.
———. 2021 (in press). “Demobilising Far-Right Demonstration Campaigns: Coercive Counter-
Mobilization, State Social Control, and the Emobilization of the Hess Gedenkmarsch Campaign.”
Social Movement Studies, https://doi.org/10.1080/14742837.2021.1889493
