Dynamics of internal resettlement during civil war
| Author | Laia Balcells |
| Published date | 01 March 2018 |
| Date | 01 March 2018 |
| DOI | http://doi.org/10.1177/0022343317750464 |
| Subject Matter | Regular Articles |
Dynamics of internal resettlement during
civil war: Evidence from Catalonia (1936–39)
Laia Balcells
Department of Government, Georgetown University
Abstract
This article explores the dynamics of internal resettlement in times of civil war by using a novel dataset of all
municipalities of Catalonia during the Spanish Civil War (1936–39), which includes information on the total
number of internally displaced from other parts of Spain that sought refuge in Catalan localities during the civil
war. The dataset, which also incorporates information on relevant covariates, is built with a combination of primary
(i.e. archival) and secondary sources. The results of the multivariate analyses indicate that political identities have a
significant impact on resettlement: people tend to relocate to places where they can find others who share their
political and/or ethnic identity; we see this reflected in political and ethnic alignments at the municipal level. In
addition, the article uncovers a relevant dynamic in the diffusion of violence at the local level: the arrival of internal
refugees in a new locality may have the unintended effect of increasing levels of direct violence due to its role in
disseminating credible news of atrocities committed by the other side. The implications of this study go beyond the
Spanish case and make a contribution to unpacking dynamics of violence and internal displacement in civil wars. The
article also sheds light on some of the mechanisms by which refugee flows can play a role in the diffusion of violence
throughout a given country.
Keywords
civil wars, displacement, refugees, Spain, violence
Introduction: Violence and displacement
during civil war
Displacement constitutes one of the most devastating
effects of civil wars. In the current civil war in Syria
(2011–present), for example, over 50% of the country’s
population has been displaced (Amnesty International,
2016). As of September 2017, there are 6.5 million
people who are displaced within Syria, constituting the
biggest internally displaced population in the world
(UNHCR, 2017). Refugees (including the internally
displaced) often live under horrific conditions, and their
return at the conclusion of the conflict is anything but
certain (Serrano, 2010; Pearlman, 2017).
1
Despite the
existence of reports by journalists linking displacement
to civil war violence, we still do not know much about
how violence relates to displacement – whether forced or
voluntary – at the local level.
Displacement can be defined as civilian migration
provoked by one or more armed groups during conflict
(Steele, forthcoming). Displacement is often a deliberate
consequence of violent attacks (or threats of attacks), and
armed groups frequently use it as a substitute for lethal
violence, or sometimes as a complement to the latter
(Azam & Hoeffler, 2002; Kalyvas, 2006; Steele, 2017);
armed groups often, for strate gic reasons, seek to use
collective targeting that leads to displacement in order
to gain control of contested territory (Steele, 2017;
Balcells & Steele, 2016). This can be seen in all types
of wars, including those leading to ethnic cleansing and
Corresponding author:
laia.balcells@georgetown.edu
1
For example, as of September of 2017, hundreds of thousands of
Rohingya refugees, a mostly Muslim minority ethnic group, remain
in limbo, having fled Myanmar in order to escape state repression and
now trying to enter Bangladesh (Safi, 2017).
Journal of Peace Research
2018, Vol. 55(2) 236–251
ªThe Author(s) 2018
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0022343317750464
journals.sagepub.com/home/jpr
genocide (Mylonas, 2012; Bulutgil, 2016; Straus, 2015).
At the same time, displacement and violence are highly
intertwined: violence engenders displacement, and vari-
ous forms of violence can result from displacement
(Hovil, 2008). Displacement often has violent conse-
quences not only in neighboring countries (Salehyan
& Gleditsch, 2006; Salehyan, 2009; Stearns, 2011), but
also within the country where displacement is occurring
(Duncan, 2005; Azam, 2009). Cycles of violence and
displacement do not come to an end easily, and they are
behind the humanitarian catastrophes in some of the
most severe civil wars in recent times (e.g. in Great
Lakes, former Yugoslavia, Colombia, and Syria).
Although cycles of violence and displacement are
hardly new phenomena, to date they have not been
extensively researched at the micro level. There is still
much to learn about the precise re lationship between
displacement and violence dynamics at the local level.
Indeed, while there has been abundant research on the
economic and political impact of immigrants and refu-
gees on the places of origin as well as on the hosting
locations (see, for example, Cebolla & Jime
´nez, 2010;
Braun & Mahmoud, 2011), there is not that much sys-
tematic micro-level research of the impact of resettle-
ment patterns on wartime dynamics (Serrano, 2010).
2
This article is an attempt to contribute to the vast
body of civil war literature by exploring the causes and
consequences of resettlement patterns in civil wars
through the case of Catalonia during the Spanish Civil
War (1936–39). Following Petersen (2001), Kalyvas
(2006), Autesserre (2008), and Christia (2012) –
among others – I consider the local community as the
most appropriate level of analysis to understand war-
time dynamics. During civil war, interactions occurring
at the local level become extremely important, to such
an extent that systems of local governance often arise,
superseding political dynamics at the national level
(Arjona, 2016).
A micro-level approach is thus favored here, that
exploits a novel fine-grained dataset with local-level
demographic, political, and geographic variables. The
dataset, built using a combination of primary (i.e. archi-
val) and secondary sources, includes data on all 1,062
municipalities of Catalonia in 1936. Precise figures on
refugee flows throughout Spain during the civil war are
still non-existent; however, to the best of my knowledge,
this is the first empirical research that uses data for all
municipalities in a single region of Spain.
3
Thus, the
article offers an empirical contribution to the study of
displacement flows during an iconic civil war that took
place in Europe during the interwar period. This manu-
script focuses on refugee influxes in Catalonia, and on
the consequences the arrival of refugees had for dynamics
of violence at the local level.
4
This article is structured as follows: the next section
introduces a set of hypotheses on the relationship
between displacement and violence during civil war. The
section that follows offers a succinct introduction to the
Spanish Civil War and to the refugee flows in/from Cat-
alonia. Afterwards, I present the data and the empirical
analyses of the determinants behind refugee settlement
in Catalonia and their consequences for the dynamics of
violence at the local level. The final section discusses the
results of this study and draws implications for other
civil wars.
Displacement dynamics during civil war
The study of displacement during civil wars is often
intertwined with the study of violence against civilians.
In fact, the predominant hypothesis in the macro-level
literature is that higher levels of violence yield more dis-
placement (Schmeidl, 1997; Davenport, Moore & Poe,
2003; Melander, O
¨berg & Hall, 2009); however, the
mechanism by which these two are interrelated is not
very clear (Steele, 2009, 2017). When it comes to the
micro level, it is now widely accepted that we need to
understand the precise dynamics behind the targeting of
civilians if we aim to understand wartime displacement
(Kalyvas, 2006).
While the first wave of micro-level approaches to
civilian victimization during civil war focused largely
on military factors (Valentino, Huth & Balch-Lindsay,
2004; Kalyvas, 2006), more recent studies have brought
political variables to the fore in order to explain variation
in levels and forms of violence and resettlement. For
example, inspired by Goodwin’s (2006) concept of ‘cate-
gorical terrorism’, Straus (2015) and Steele (2009, 2017)
have used the terms ‘categorical mass violence’ and
2
Some notable exceptions are Stee le (2009, 2017, forthcoming),
Engel & Iba
´n
˜ez (2007), and Iba
´n
˜ez (2008).
3
‘Refugees’ commonly refers to displaced individuals crossing
international borders. In addition to using the term ‘internally
displaced’ and the acronym IDP, in this article, I also refer to
internally displaced people as refugees because they were called
‘refugiados’ (in Spanish) and ‘refugiats’ (in Catalan).
4
Balcells & Steele (2016) have analyzed the various determinants of
refugee flows out of Catalonia during the civil war period, comparing
them to refugee dynamics in the Colombian civil war.
Balcells 237
To continue reading
Request your trial