The consequences of victimization on political identities: Evidence from Spain

Politics and Society

Volumn 40, Issue 3, 2012, Pages 311-347

  Balcells, Laia ^a,b  

^a Barcelona)   (Spain)

^b BARCELONA GRADUATE SCHOOL OF ECONOMICS   (Spain)

Author keywords

Civil wars; Dictatorship; Political identities; Spain; Victimization

Indexed keywords

EID: 84864952281     PISSN: 00323292     EISSN: 15527514     Source Type: Journal    
DOI: 10.1177/0032329211424721     Document Type: Article

References (112)

1. David R. Mayhew, Electoral Realignments: A Critique of an American Genre (New Haven, CT: Yale University Press, 2002), 164.
2. Christopher Blattman and Edward Miguel, "Civil Wars," Journal of Economic Literature 48, no. 1 (2010): 3-57 p. 42.
3. David R. Mayhew, "Wars and American Politics," Perspectives on Politics 3, no. 3 (2005): 473-493.
4. John Bellows and Edward Miguel, "War and Local Collective Action in Sierra Leone" (Manuscript, University of California, Berkeley, 2008).
5. Christopher Blattman, "From Violence to Voting: War and Political Participation in Uganda," American Political Science Review 103, no. 2 (2009): 231-247.
6. Steven Shewfelt, "The Legacy of War: Wartime Trauma and Post-conflict Political Life" (PhD diss., Yale University, 2009).
7. Geoffrey Wallace, "When Do People Back the Perpetrators? Popular Support for the Use of Violence during War" (paper, Midwest Political Science Association annual meeting, Chicago, April 1-4, 2011).
8. Kristin Bakke, Xun Cao, John O'Loughlin, and Michael Ward, "Social Distance in Bosnia-Herzegovina and the North Caucasus Region of Russia: Inter and Intra-ethnic Attitudes and Identities," Nations and Nationalism 15, no. 2 (2009): 227-253.
9. Maarten Voors, Eleonora Nillesen, Phillip Verwimp, Erwin Bulte, Robert Lensink, and Daan van Soest, "Does Conflict Affect Preferences? Results from Field Experiments in Burundi" (Ecore Discussion Paper 17, 2010).
10. Dora L. Costa and Matthew E. Kahn, Heroes and Cowards: The Social Face of War (Princeton, NJ: Princeton University Press, 2008), 21.
11. The civil war in Spain started as the result of a failed putsch organized by General Francisco Franco on July 18, 1936, which led to two different areas of military control: one by the rebels or Nationalists (i.e., the supporters of the coup) and the other by the loyalists or Republicans (i.e., the detractors of the coup). These areas were highly determined by the outcomes of the coup at subnational level (i.e., successful or not), which were in turn determined by a number of factors, a lot of them apparently more contingent than systematic. For example, the success of the coup in an area depended on idiosyncratic features of military leaders in the garrisons (i.e., people who were willing and/or capable of undertaking the specific orders sent by Franco) and on the evolution of the events taking place right after, namely the reaction of the local political elites and the popular masses. Furthermore, "chance" played a crucial role in this. In short, we can argue that the main front lines of the conflict did not follow a pattern that could have possibly been predicted before the coup.
12. Adam Thompson, "The New Powers in College Football Carry Old Baggage: As Showdown Looms, Kansas and Missouri Fans Re-fight the Civil War," Wall Street Journal, November 20, 2007, A1+A16.
13. Peter Hart, The I.R.A. at War 1916-1923 (New York: Oxford University Press, 2003).
14. In fact, in contemporary Ireland, a number of politicians have generational connections with the leaders of the parties that fought in the 1920s (Paul Staniland, pers. comm., May 2007).
15. J. J. Sigal, D. Silver, V. Rakoff, and B. Ellin, "Some Second Generation Effects of Survival of the Nazi Persecution," American Journal of Orthopsychiatry 43, no. 3 (1973): 320-327.
16. Gary D. Allinson, Japan's Postwar History (Ithaca, NY: Cornell University Press, 1997).
17. Shale Horowitz, "War after Communism: Effects on Political and Economic Reform in the Former Soviet Union and Yugoslavia," Journal of Peace Research 40, no. 1 (2003): 25-48.
18. Stathis N. Kalyvas, The Logic of Violence in Civil War (New York: Cambridge University Press, 2006).
19. Stathis N. Kalyvas, The Logic of Violence in Civil War (New York: Cambridge University Press, 2006).
20. Laia Balcells, "Rivalry and Revenge: Violence against Civilians in Conventional Civil Wars," International Studies Quarterly 54, no. 2 (2010): 291-313.
21. I will explore physical violence as well as other sources of victimization (such as displacement, labor repression, imprisonment, and similar acts), which I will group in two categories: moderate and severe victimization (see below). The study of the political effects of displacement is still underdeveloped, with the exception of only a few recent contributions.
22. Abbey Steele, "Seeking Safety: Avoiding Displacement and Choosing Destinations in Civil Wars," Journal of Peace Research 46, no. 3 [2009]: 419-429.
23. The implications of social and economic change taking place during and after war has become a field of study in itself e.g., MICROCON, Households in Conflict Network.
24. Patricia Justino, "Poverty and Violent Conflict: A Micro-Level Perspective on the Causes and Duration of Warfare" [Working Paper 46, Households in Conflict Network, 2008].
25. However, I am unaware of any contributions linking wartime-related economic and social change to political identities.
26. Stevan E. Hobfoll, Daphna Canetti-Nisim, and Robert J. Johnson, "Exposure to Terrorism, Stress-Related Mental Health Symptoms, and Defensive Coping among Jews and Arabs in Israel," Journal of Consulting and Clinical Psychology 74, no. 2 (2006): 207-218.
27. Voors et al., "Does Conflict Affect Preferences?"
28. Herbert C. Kelman, "Conflict Resolution and Reconciliation: A Social-Psychological Perspective on Ending Violent Conflict between Identity Groups," Landscapes of Violence 1, no. 1 (2010): article 5.
29. Daphna Canetti-Nisim, Eran Halperin, Keren Sharvit, and Stevan E. Hobfoll, "A New Stress-Based Model of Political Extremism: Personal Exposure to Terrorism, Psychological Distress, and Exclusionist Political Attitudes," Journal of Conflict Resolution 53 (2009): 363-389.
30. Phuong Pham, Harvey Weinstein, and Timothy Longman, "Trauma and PTSD Symptoms in Rwanda: Implications for Attitudes toward Justice and Reconciliation," Journal of the American Medical Association 292, no. 5 (2004): 602-612.
31. James L. Gibson, Overcoming Apartheid: Can Truth Reconcile a Divided Nation? (New York: Russell Sage, 2004).
32. Monika Nalepa, Skeletons in the Closet: Transitional Justice in Post-communist Europe (New York: Cambridge University Press, 2010).
33. Paloma Aguilar, Laia Balcells, and Héctor Cebolla, "Determinants of Attitudes towards Transitional Justice: An Empirical Analysis of the Spanish Case," Comparative Political Studies 44, no. 10 (forthcoming).
34. Hobfoll, Canetti-Nisim, and Johnson, "Exposure to Terrorism."
35. These authors refer to the impact of PTSD on political beliefs such as authoritarianism or ethnocentrism.
36. Carl Von Clausewitz, On War, ed. Colonel F. N. Maude, trans. Colonel J. J. Graham (1832; repr., London: Routledge and Kegan Paul, 1968).
37. Canetti-Nisim et al., "New Stress-Based Model": 364.
38. Elisabeth J. Wood, Insurgent Collective Action and Civil War in El Salvador (New York: Cambridge University Press, 2003).
39. Wallace, "When Do People Back the Perpetrators?"
40. Canetti-Nisim et al., "New Stress-Based Model."
41. Pham, Weinstein, and Longman, "Trauma and PTSD Symptoms," 611.
42. Jennifer Lerner and Dacher Keltner, "Fear, Anger, and Risk," Journal of Personality and Social Psychology 81 (2007): 146-159.
43. Daniel Bar-Tal and Daniela Labin, "The Effect of a Major Event on Stereotyping: Terrorist Attacks in Israel and Israeli Adolescents' Perceptions of Palestinians, Jordanians, and Arabs," European Journal of Social Psychology 31 (2001): 265-280.
44. Wood, Insurgent Collective Action.
45. Canetti-Nisim et al., "New Stress-Based Model."
46. Demobilization or apathy has been very commonly observed in victims of sexual violence during war.
47. Cheryl Bernard, "Rape as Terror: The Case of Bosnia," Terrorism and Political Violence 6, no. 1 [1994]: 29-43.
48. Elisabeth J. Wood, "Variation in Sexual Violence during War," Politics & Society 34, no. 3 [2006]: 307-342.
49. Kimberly Theidon, "Gender in Transition: Common Sense, Women and War," Journal of Human Rights 6, no. 4 [2007]: 453-478.
50. The term political interest usually means "the degree to which politics arouses a citizen's curiosity".
51. Jan W. Van Deth, "Interest in Politics," in Continuities in Political Action: A Longitudinal Study of Political Orientations in Three Western Democracies, ed. K. Jennings and J. W. Van Deth [Berlin: de Gruyter, 1989], 275-312.
52. It must be noted that radicalization or polarization of political opinions, which has been analyzed in some conflicts.
53. David A. Jaeger, Esteban Kflor, Sami H. Miaari, and M. Daniele Paserman, "The Struggle for Palestinian Hearts and Minds: Violence and Public Opinion in the Second Intifada" [Working Paper 52, Households in Conflict Network, 2008].
54. Indonesia: Shewfelt, "Legacy of War", can be integrated in the first effect (i.e., rejection). In the case of conflicts in which there are very clear boundaries between groups, acceptance will rarely take place, and we can think that attitudes of the victims will swing among "no effect," "demobilization," "mild rejection," and "strong rejection"; the latter would be what has been conceptualized as radicalization or "polarization."
55. Wood, Insurgent Collective Action.
56. Kalyvas, Logic of Violence.
57. Matthew Kocher, Thomas Pepinsky, and Stathis Kalyvas, "Bombing as an Instrument of Counter-insurgency in the Vietnam War," American Journal of Political Science 55, no. 2 (2010): 201-218.
58. Elisabeth Wood, "The Social Processes of Civil War: The Wartime Transformation of Social Networks," Annual Review of Political Science 11 (2008): 539-561.
59. Wallace, "When Do People Back the Perpetrators?"
60. Sigal et al., "Some Second Generation Effects."
61. Kent M. Jennings and Richard G. Niemi, Generations and Politics: A Panel Study of Young Adults and Their Parents (Princeton, NJ: Princeton University Press, 1981).
62. Annick Percheron and Kent M. Jennings, "Political Continuities in French Families: A New Perspective on an Old Controversy," Comparative Politics 13 (1981): 421-436.
63. Aguilar, Balcells, and Cebolla, "Determinants of Attitudes."
64. Antony Beevor, The Spanish Civil War (London: Orbis, 1982).
65. In February 1936, the national elections opposed a left-wing political coalition (the Frente Popular, which grouped all left-wing parties, including the Anarchists) and a right-wing coalition (the Frente Nacional, which grouped all right-wing parties). The left won those elections with 42.9 percent of votes and 60.5 percent of seats.
66. Juan J. Linz and Jesús M. De Miguel, "Hacia un análisis regional de las elecciones de 1936 en España," Revista Española de la Opinión Pública 48 [1977]: 27-68.
67. Issues around land distribution were some of the main catalysts of political violence during the Second Republic and in fact have been considered to be one of the main determinants of the Spanish Civil War.
68. Edward Malefakis, Reforma Agraria y Revolución Campesina en la España del Siglo XX [Barcelona: Ariel, 1976].
69. Josep M. Solé i Sabaté and Joan Villarroya, L'ocupació militar de Catalunya. Març 1938-Febrer 1939 (Barcelona: L'Avenç, 1987).
70. The exceptions were the main right-wing nationalist political parties in Valencia and the Basque Country, which supported the Republican government.
71. Among the total death toll, 122,000 are estimated to have been civilian victims of intentional lethal violence: approximately 81,095 were killed by the right and 37,843 were killed by the left (Santos Juliá, Víctimas de la guerra civil [Madrid: Temas de Hoy, 2004]). These figures have to be read with the caveat that only half of the Spanish provinces have been researched in depth to date (historians are still involved in debates about estimations). Data on refugees are also very fragmentary and should be viewed with caution.
72. People born in 1930 were six years old when the war started, and nine when it finished. I made the decision to make 1930 the cutoff year to avoid interviewing people who were too young to possibly have any memories of the conflict.
73. In the first wave, I contacted people through snowball techniques, and I conducted the interviews mostly in Madrid and Barcelona (capital cities). In the second wave, the selection process was more focused at a community level: I interviewed people whom I managed to contact in the municipalities/counties where I was conducting dissertation fieldwork.
74. The complete interview protocol is available from the Online Appendix at: http://balcells.iae-csic.org.
75. The regions of origin and where people lived during the war were more varied (e.g., Canary Islands, the Basque Country, Castile, and Andalusia).
76. Lee Ann Fujii, "Shades of Truth and Lies: Interpreting Testimonies of War and Violence," Journal of Peace Research 47, no. 2 (2010): 231-241.
77. For example, although some people were reluctant to openly report their political identities, this could be inferred from comments or attitudes displayed throughout a long conversation.
78. Fear was broadly more prevalent among people who experienced the civil war in Nationalist territory, rather than among those who experienced it in the Republican zone. And it could be displayed as more or less generalized: that is, either about a possible return to civil war or about potential reprisals at the individual level. This fear could be perceived in the way subjects talked about the subject (for example, if they were reluctant to talk about it openly or if they explicitly asked me not to record the conversation). Sometimes, the subjects explicitly mentioned their fear.
79. Again, in this civil war, recruitment to one or the other side depended mostly on the location of the person, as conscription was routinized on both sides.
80. Laia Balcells, "Behind the Frontlines: Identity, Competition, and Violence in Civil Wars" [PhD diss., Yale University, 2010].
81. Although this qualitative evidence is not conclusive, it is worth mentioning that it is the opposite of what has been found in other settings, for example, Sierra Leone (Bellows and Miguel, "War and Local Collective Action"), Uganda (Blattman, "From Violence to Voting"), and Indonesia (Shewfelt, "Legacy of War"), and suggests that the results that have been obtained in short-term settings (i.e., immediately after the war ended) might not necessarily apply to long-term contexts.
82. There are several contextual elements of the Spanish case that need to be accounted for: First is the existence of a long-term repressive dictatorship, which not only limited the political options but also promoted a particular view of the civil conflict (in the official Francoist discourse there was only one victimizer: the left). The dictatorship also had a demobilization intent; for example, Edward Hansen (Rural Catalonia under the Franco Regime [Cambridge, UK: Cambridge University Press, 1977]), argues that the growth of public apathy is the most profound consequence of Franco's regime. Second is the generational or so-called period effect: those people I interviewed were elders who may have reduced their political involvement because of their age. Third, because of the "pact of silence" that accompanied the transition to democracy (Paloma Aguilar, Memoria y olvido de la guerra civil española [Madrid: Alianza, 1996]), during the democratic period there may have been an absence of political options satisfying victimized individuals, and they may have lost interest in politics because of this.
83. In fact, there is a greater rate of nonresponses to the question about "family identification" than there is for "individual identification"; this indicates that there may not be as much of a projection of current individual identities on reported family loyalties as we initially suspected.
84. I have widened the frame of years here (as compared to the semistructured interviews) so as not to shrink the sample too much.
85. For those victimized, we asked who was to blame for the events: the Nationalist side, the Republican side, or both. The distribution of the responses, both in absolute numbers and in column percentages, is shown in Table A1 of the appendix. The Nationalist side is reported as the perpetrator of violations to a greater extent than the Republican side-and this is particularly the case for "having to leave Spain" (72.7 percent of cases), "having to hide" (44.8 percent of cases), and "was imprisoned" (45.4 percent). The only respondent who reported having been sacked from work blamed the Nationalist side.
86. In Table A2 of the appendix, I explore the placement of elders on the ideological scale, as reported on the survey (on our scale, 1 is extreme left and 10 is extreme right). I distinguish between victims (of each side, as well as of either side) and nonvictims to check if there are any differences between these subgroups. I find that there are ideological differences between victimized and nonvictimized people, and-more specifically-that differences exist between groups of victimized people depending on the side that perpetrated the violations: those victimized by the Nationalists are on average much more leftist than the remainder of the sample; conversely, those victimized by the Republican are much more conservative (or rightist) than the rest of the sample.
87. I code this from a self-placement question in the survey. In a set of robustness checks, I operationalize the dependent variable (Leftscale) as an ordinal variable, with 10 representing extreme left and 1 representing extreme right.
88. Angus Campbell, Philip Converse, Warren Miller, and Donald Stokes, The American Voter (Chicago: University of Chicago Press, 1960).
89. Russell J. Dalton, Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies (Washington, DC: CQ Press, 2006).
90. In this case, I operationalize nationalism with voting behavior (i.e., voting for a nationalist party) because the scale for nationalist identity has been widely argued to have measurement issues and is not perfectly comparable to the left-right scale.
91. I include individual victimization in this variable so that in the regressions we do not lose cases of people who were directly victimized but who did not have anyone in their family or a close friend victimized.
92. I would have liked to distinguish between victimization of immediate family members and that of friends and extended family members, but there was not a question specifying these differences in the survey.
93. Aguilar, Balcells, and Cebolla, "Determinants of Attitudes."
94. Table A4 of the appendix depicts the descriptive statistics of all the variables included in the forthcoming regressions.
95. This is a contextual variable more than a sociodemographic one. However, I included it here because it operates practically as a sociodemographic control. The categories are the following: 1 =fewer than or equal to 2,000 inhabitants, 2 =between 2,001 and 10,000, 3 =between 10,001 and 50,000, 4 =between 50,001 and 100,000, 5 =between 100,001 and 400,000, 6 =between 400,001 and 1,000,000, 7 =more than 1,000,000.
96. I proxy it with the question: "Could you tell me if you are interested in politics in general?" Possible responses are 4 =very much, 3 =quite a lot, 2 =a little bit, 1 =not at all.
97. For education, 1 =primary education or less, 2 =secondary education, 3 =university degree. Education and interest in politics are two mandatory controls since individuals scoring higher in one of them are much less likely to give a "don't know" or "does not answer" sort of answer (Aguilar, Balcells, and Cebolla, "Determinants of Attitudes").
98. This is a scalar variable that goes from 1 to 6, where 1 is nonreligious (the respondent identifies herself or himself as atheist or nonreligious) and 6 is highly religious (the respondent says that she or he goes to mass several days a week).
99. This is a scalar variable that goes from 1 to 4 and that is operationalized as follows: "When you were a child or adolescent, how much was politics talked about at home?" The response options are 4 =very much, 3 =quite, 2 =a little bit, 1 =not at all.
100. The regressions are run with only those cases for which there are no missing values in any of the variables; thus, all the does not know/does not answer responses in the crucial explanatory variables are not included in the analysis.
101. In M5, the marginal effects (dy/dx) of the socialization and victimization variables are -0.14 for Family Nationalist, 0.27 for Family Republican, 0.06 for Victim Nationalists, -0.07 for Victim Republicans, 0.072 for Victim Francoism. Thus, while the victimization variables are statistically significant, the socialization variables are relatively more relevant, in substantive terms.
102. The only changes are that the variables Talked Politics (in all models) and Francoist Victimization (in model 5) are not significant.
103. 2(3) =10.51.
104. Kalyvas, Logic of Violence, 405-407.
105. Interestingly, if we run the regressions for subsamples of age groups or cohorts (not included here, but available on request), the victimization variables are highly significant for the cohorts of people between eighteen and thirty, and between thirty-one and forty-five, but these are not significant for people between forty-five and fifty-five. For cohorts between fifty-six and seventy and for those older than 70, only the dictatorship victimization is significant. Thus, while victimization during the civil war has an impact on younger generations, older generations are affected by victimization during the dictatorship.
106. Table A6 of the appendix includes the results of regressions with Moderate/Severe Victimization for Nationalist vote: these do not raise major differences as compared to Table 10, except that Nationalist Severe Victimization and Moderate Victimization demonstrate a positive effect on peripheral nationalist vote in the Basque Country.
107. In a set of additional regressions (available on request), I checked for the interactive effect of age with each of these victimization variables: they again display no support for hypothesis 3. The results are in fact not very consistent with the interactive analyses above: the only interactions that are significant are Age and Francoist Moderate Victimization (the effect increases with age) and Age and Severe Republican Victimization (the effect decreases with age).
108. The probability of being a leftist for an average individual in the sample, everything else equal, increases by 0.36 if she or he has a Nationalist family; the exact reverse effect is observed if she or he has a Republican family. Moderate victimization under the Franco regime increases the probability of being a leftist by 0.18, severe victimization by the Nationalists during the civil war increases it by 0.2, and severe victimization by the Republicans during the civil war decreases the likelihood of being a leftist by 0.21. The results of the postestimations are available on request.
109. For example, Aguilar argues that a crucial factor explaining the resurgence in the debate on the Historical Memory in Spain in the 2000s was the entrance into the public sphere of the "grandchildren of the war," who were free from the fears and guilt of their parents.
110. Paloma Aguilar, "Transitional or Post-transitional Justice? Recent Developments in the Spanish Case," South European Society & Politics 13, no. 4 [2008]: 417-433.
111. In fact, moderate violations are those most reported within the sample (i.e., 22 percent, as shown in Table 7), and especially among the oldest generation (almost 30 percent of people over seventy years old report moderate victimization during the dictatorship, the highest percentage among all the age groups).
112. Kalyvas, Logic of Violence, 403.