An Exploratory Study on the Impact of Electoral Participation upon a Terrorist Group’s Use of Violence in a Given Year

An Exploratory Study on the Impact of Electoral Participation upon a Terrorist Group’s Use of Violence in a Given Year

by Stephen McGrath and Paul Gill

 

Abstract

Recent studies seeking to understand the determinants of terrorism tend to focus upon situational, rather than structural measures. Typically these studies examine the interaction of terrorist attacks and repressive state actions. However, we know very little about other situational measures that may impact upon a group’s scale of violent activity within a particular year. This preliminary study analyses terrorist attacks committed by both the Provisional IRA (PIRA) and ETA and the electoral performances of the groups’ political wings, Sinn Fein and Batasuna, from 1970 to 1998 and from 1978 to 2005 respectively. More specifically, this paper examines whether the nature and content of terrorist attacks differ in the build-up to that group’s political-wing participating in elections. In other words, this article is a preliminary study of the influence of electoral participation on attack frequency and target selection. Results suggest that PIRA significantly decreased their attacks in an election year and this had a positive impact upon Sinn Fein’s electoral performance. On the other hand, ETA significantly increased its attacks in an election year and this had no significant impact upon Batasuna’s electoral performance.

Keywords: Terrorism, election, violence, target selection

 

Introduction

Historically, the analysis of the quantity and quality of terrorist attacks has typically focused upon correlations with ‘root causes’. These ‘root causes’ include (but are not necessarily limited to) factors like educational attainment, [1] economic performance [2] and regime type. [3] The relatively easy availability of such measures makes them suitable for large-scale, cross-country comparative approaches. The aggregate nature and relative inflexibility of these measures (e.g. propensity to change from one year to another) makes them unsuitable for single-case study approaches where the dependent variable varies substantially from year to year.

More recently, studies seeking to understand the determinants of terrorism tend to focus upon situational measures that are likely to fluctuate across the time period under consideration. Typically these studies examine the interaction of terrorist attacks and repressive state actions in diverse cases such as Northern Ireland, [4] Palestine, [5] Chechnya, [6] Egypt, [7] Pakistan [8] and Iraq [9]. Apart from blue-team/read-team interactions, we know very little about other situational measures that may impact upon a group’s scale of violent activity within a particular year. This lack of focus is worrying considering the fact that such situational measures are far more malleable to counter-terrorism policies than ‘root-cause’ measures that are usually deeply entrenched.

Given the seeming consensus that terrorism is often deployed to impact upon the political process, it is odd that there have been relatively few quantitative examinations (apart from studies of regime type and regime structure) [10] of how the shifting political sphere impacts upon levels terrorist activity. Newman’s cross-national study illustrates that terrorist violence increases closer to an election date. [11] Bali and Park disaggregated terrorist attacks into domestic and transnational types. [12] They found that while transnational attacks decline prior to elections, domestic attacks increase. Both studies make no differentiation between groups who are participating in these elections (through their political wing) and those who are not. De la Calle and Sanchez-Cuenca, on the other hand, take the case of ETA and highlight how its violent activity impacts upon its political support. [13] The evidence suggests that ETA’s political wing, Batasuna, loses support when ETA kills members of the security forces and non-nationalist politics; while support increases when ETA kills informers and drug dealers. The de la Calle and Sanchez-Cuenca article highlights the importance of (a) quantitative and longitudinal studies based on a single case and (b) disaggregating across target type. Finally, Berrebi and Klor examine how the electorate are sensitive to terrorism. Basically, they conclude that victimized groups become ideologically polarized. [14] Admittedly, their focus is upon the electoral behavior within the victimized group, but it is not a big leap to suggest that it could also impact upon the voting behaviour of the constituency that the terrorist group claims to represent.

In this preliminary study, we analyze the cases of two terrorist groups that also had political wings participating in electoral politics. We analyze terrorist attacks committed by both the Provisional IRA (PIRA) and ETA and the electoral performances of the groups’ political wings, Sinn Fein and Batasuna, from 1970 to 1998 and from 1978 to 2005 respectively. More specifically, we are interested in examining whether the nature and content of terrorist attacks differ in the build-up to that group’s political wing participating in elections. In other words, this article  is an exploratory study of the influence of political participation on attack frequency and target selection. We are interested in answering the three questions: Does political participation in a given year impact upon attack frequency? Does political participation in a given year impact upon attack frequency equally across different targets? Does targeting impact upon electoral results?

ETA and PIRA are a natural duo to compare given their similar ethno-nationalist goals, period of operation and types of violence deployed. Indeed, many studies have compared these groups along factors such as: their social background, [15] levels of public support, [16] strategic orientation, [17] target selection, [18] individual motivations for joining, [19] tactical choice, [20] structure, [21] and legitimisation strategies. [22]

 

Data

We measured aggregated terrorist attack frequency year-on-year against the electoral participation of the group’s political wing. In both cases, the analysis totaled eight years of electoral activity and twenty years of non-electoral activity. We stratified terrorist attacks according to the Global Terrorism Database’s classification. We also evaluated associations between political vote gains and target choice by correlating all target selections with Sinn Fein and Batasuna electoral performance.

Voting records (both District and Parliamentary) were obtained from the Northern Irish “ARK” political research database [23] and the Basque government official website.[24] Data regarding terrorist attack frequency and target selection from Northern Ireland (N = 1874) and Spain (N = 1916) were supplied by the Global Terrorism Database (GTD, START), which consequently provided the disaggregation of attacks into their respective target selections. GTD data was aggregated with vote count and percentage share data acquired by the government election records in the same dataset. A control group was formed in order to measure the attack frequency on non-election years: neutral years within the time period of 1970-2005 were selected provided they did not feature political activity.  In the Spanish case study, election years were: 1979, 1980, 1983, 1986, 1987, 1990, 1991, and 1994, crossing Municipal (District) and Assembly cases. The election years selected for the Northern Irish elections were: 1982, 1985, 1989, 1994, 1996, 1997, 1998, 2003, crossing District and Assembly cases, as well as one European case (1994).  We were unable to utilise local 1993 voting data because GTD possesses no data for that year. The decision was made to use the 1994 European election results for Sinn Fein and the corresponding attack frequency instead. 

 

Results

Figure 1 illustrates the frequency of IRA and ETA attacks from 1975 to 1999. It highlights general fluctuations in both the Northern Irish and Spanish case studies, while an overall decline in attacks is observed towards the late 1990s in both cases.

Figure 1 – Attack frequency of the IRA and ETA (1975-1999)

First we test whether the levels of terrorism experienced were affected (either positively or negatively) by electoral participation in a given year. We conducted an independent t-test to determine whether there was a significant difference in the mean number of attacks in election and non-election years for both cases. Both cases reflected drastically different results. PIRA engaged in a (partially) significant lower number of attacks within election years than non-election years. Indeed, Table 1 illustrates that PIRA engaged in almost 50% fewer attacks in election years (45 attacks vs. 84 attacks). On the other hand, ETA engaged in significantly more attacks in electoral years than non-electoral years (94 attacks vs. 45 attacks). Table 2 highlights the fact that ETA engaged in more than double the number of attacks in electoral years. 

Table 1 – Independent t-test of attack frequency against (District and Parliament) electoral participation - Northern Ireland    

 

Table 2 – Independent t-test of attack frequency against (District and Parliament) electoral participation  - Spain

Together the results suggest that the violence conducted by both groups were significantly different in election and non-election years although the effects were entirely different. Next we test whether this difference remained consistent across all target types or whether some targets were attacked significantly more (in ETA’s case) or less (in PIRA’s case). In other words, we tested whether the established increase/decrease in attack methods was disproportionately due to an increase/decrease in attacks on one type of target or whether it was a general trend.

Figure 2 illustrates PIRA’s relative target selection across the time period under consideration. The results reflect the changing picture of target selection with certain strategies fluctuating consistently with overall attacks (Military Attacks), and others taking on more unpredictable patterns (Business, Citizen Attacks).

Figure 2 – IRA target selection over time (1973-2001)

We then conducted a series of Independent t-tests to analyze the frequency of attacking these targets in election and non-election years. The results demonstrate that while all target types are attacked less in election years, this effect is particularly prominent and statistically significant with regards to both attacks on businesses and attacks on the military.

Table 3 – Independent t-test for selected targets against electoral participation

Turning to the case of ETA, we applied the same procedures. Figure 3 illustrates the strong focus on targeting businesses and police throughout the time period analysed. Attacks on citizens and infrastructure are inconsistent chronologically and do not appear to establish any general trends. 

Figure 3 – ETA target selection over time (1970-1999)

The series of Independent t-tests illustrate that ETA’s increased use of terrorist attacks in electoral years was predominantly carried out against business, police and military targets. 

Table 4 – Independent t-test for selected targets against electoral participation

We also tested whether these differences were apparent both a week before and after the election and a month before and after the election. No significant differences were apparent. Evidently, the two groups engaged in significantly different tactical practices in election and non-election years, but not months or weeks. This may reflect long-term strategic planning (in terms of the increase/decrease in attacks occurring much earlier) and/or the level of time it takes for strategic plans to be put into force (in terms of the increase/decrease in attacks occurring much later after the election).

The above results suggest that in Northern Ireland, target selection decreases universally with particular reference to business and military, whilst in Spain, these same targets (in addition to the police) are affected in a significant positive direction. This may be due to a number of reasons including opportunity costs, counter-terrorism practices, and target hardening. It could also be due to the impact of targeting practices on electoral performance, an aspect which we turn to next.

To explore the relationship between voting patterns and attack strategy, we utilised a Pearson’s product-moment Bivariate Correlation, in which the vote counts for Sinn Fein and Batasuna (indicated as “1” in the matrix) were correlated with each target choice in turn. General assumptions were met prior to the undertaking of the test, including the normal distribution of the variables. Table 5 outlines the results for PIRA.

Table 5 – Matrix of target selection and vote quantity (PIRA)

It is important to clarify that all associations are negative; the less a particular target is attacked, the higher vote tally Sinn Fein achieved in that year. In other words, the results suggest that the more discriminate PIRA was in its use of violence, the greater the benefit for Sinn Fein. The fact that the comparison of means test (above) illustrated a negative downturn in attacks (by almost a half) in electoral years and that these downturns corresponded with Sinn Fein gaining electoral points, suggests that more was at play than more sensitive counter-terrorism practices in the immediate build-up to the election itself. The significant downturn in attacks against businesses and the military also reflect PIRA’s strategic orientation of improving their political position impacting upon day-to-day tactical operations. The results also run contrary to the large-N analyses conducted by Newman [11] and Bali and Park [12] who found that terrorist attacks increased when elections are held.

Table 6 outlines the results of the same test applied to the ETA case. Again, the associations are all negative. The more ETA engaged in attacks on all targets (apart from citizens) the less Batasuna achieved in the election although none of these associations displayed significant scores. While ETA engaged in significantly more attacks on particular targets in election years, there is no evidence to suggest that this resulted in greater political support. In other words, PIRA’s strategy paid off while ETA’s made no discernible difference at all.

Table 6 – Matrix of target selection and vote quantity (ETA)

The results differ from the de la Calle and Sanchez-Cuenca [13] findings, outlined above, for a couple of reasons. First, it must be pointed out here that the category of ‘citizen’ utilized by GTD is quite all-encompassing and includes informers and drug dealers – two categories of target that they showed significantly increased Batasuna’s vote. Second, de la Calle and Sanchez Cuenca measured the number of people killed as opposed to our measure of number of attacks committed.

 

Conclusion

The contrasting results highlight the utility of examining both situational drivers of terrorist activity as well as longitudinal quantitative case studies. Despite their structural, ideological and aspirational similarities, PIRA and ETA displayed markedly different behaviours in years that their political wing engaged in elections. PIRA’s violent attacks significantly declined while ETA’s significantly increased. While PIRA’s strategy paid dividends politically, ETA’s did not. PIRA’s propensity to reduce the frequency of their attacks on both businesses and the military may have been driven by a need to avoid negatively impacting their constituency during the build-up to the election, either by disturbing their routine activities or provoking a military backlash. Interviews with former militants and other qualitative research may be helpful for validating this.

It would also be interesting to see how these dynamics play out in a non-European setting. Both ETA and PIRA were relatively discriminate compared to some Middle Eastern groups many of whom possess multiple rivals vying for political and social support. Focusing on Middle Eastern groups would add further layers of complexity to our understanding of the link between political participation and the frequency of terrorist attacks, although relatively few of those groups have put forward a political wing to participate in electoral processes. Additionally, further studies necessitate factoring in the potential impact of a government’s counterterrorism efforts, and also accounting for the geographic distribution of potential targets in relation to both groups’ constituency base. [14]

One thing is for certain however; engaging in electoral politics in a given year corresponded to a change in both ETA and PIRA’s violent output on both an aggregate level (against all targets) and a disaggregated level (against specific targets). Such results reflect organizational behavior and strategy of terrorist groups, and can help feed into security decisions concerned with risk assessment of particular targets in the build-up to an election. The same is true for directing finite resources toward protecting targets that may be particularly valued by a terrorist organization in a given context. The results also illustrate the importance of context when making policy-making decisions. While large-N, comparative and aggregated studies can help elucidate general trends, they may miss some of the subtlety, nuance and case-specific drivers of behaviours.

Our exploratory study is just a minor step in what seems a general trend toward this kind of analysis. [22] Its main finding is that despite the fact that PIRA and ETA are very similar organisations, with both of them having ‘democratic’ political parties, their attack and targeting strategies seem to be related to electoral cycles, but in divergent ways, necessitating very different counter-terrorism strategies and practices.

About the Authors: Stephen McGrath is currently undertaking a PhD in Politics and International Relations at the University of East Anglia. He holds an MSc in Security and Crime Science from University College London, and a BA in Politics from the University of Bath. Paul Gill is a lecturer at University College London's Department of Security and Crime Science. His book on lone-actor terrorism is forthcoming with Routledge - http://www.routledge.com/books/details/9781138787568/


Notes

[1] Krueger, A. B., & Maleckova, J. (2003). Education, poverty and terrorism: Is there a causal connection?. The Journal of Economic Perspectives, 17(4), 119-144. Testas, A. (2004). Determinants of terrorism in the Muslim world: an empirical cross-sectional analysis. Terrorism and Political Violence, 16(2), 253-273. Brockhoff, S., Krieger, T., & Meierrieks, D. (2010). Ties That Do Not Bind (Directly)–The Education-Terrorism Nexus Revisited. Center for International Economics Working Paper, (2010-02).

[2] Abadie, A. (2004). Poverty, political freedom, and the roots of terrorism (No. w10859). National Bureau of Economic Research. Bravo, A. B. S., & Dias, C. M. M. (2006). An empirical analysis of terrorism: Deprivation, Islamism and geopolitical factors. Defence and Peace Economics, 17(4), 329-341. Gries, T., Krieger, T., & Meierrieks, D. (2011). Causal linkages between domestic terrorism and economic growth. Defence and Peace Economics, 22(5), 493-508. Piazza, J. A. (2006). Rooted in Poverty?: Terrorism, Poor Economic Development, and Social Cleavages 1. Terrorism and Political Violence, 18(1), 159-177.

[3] Chenoweth, E. (2013). Terrorism and Democracy. Annual Review of Political Science, 16, 355-378.  Savun, B., & Phillips, B. J. (2009). Democracy, foreign policy, and terrorism. Journal of Conflict Resolution, 53(6), 878-904. Wade, S. J., & Reiter, D. (2007). Does democracy matter? Regime type and suicide terrorism. Journal of Conflict Resolution, 51(2), 329-348.

[4] Lafree, Gary, Laura Dugan and Raven Korte. 2009. “The Impact of British Counterterrorist Strategies on Political Violence in Northern Ireland: Comparing Deterrence and Backlash Models.” Criminology 47(1): 17-45. Asal, V., P. Gill, K. Rethemeyer and J. Horgan. (Forthcoming, 2014) “Killing Range: Explaining Lethality Variance Within a Terrorist Organization”. Journal of Conflict Resolution

[5] Hafez, Mohammed M. and Joseph M. Hatfield.  2006.  “Do Targeted Assassinations Work?: A Multivariate Analysis of Israel’s Controversial Tactic During the Al-Aqsa Uprising.”  Studies in Conflict and Terrorism.  29(4): 359-382. Dugan, L., & Chenoweth, E. (2012). Moving Beyond Deterrence. The Effectiveness of Raising the Expected Utility of Abstaining from Terrorism in Israel. American Sociological Review, 77(4), 597-624. Maoz, Zeev. 2007. “Evaluating Israel’s Strategy of Low-Intensity Warfare, 1949-2006.” Security Studies 16: 319-349. Benmelech, Efraim., Claude Berrebi and Esteban Klor. 2010. “Counter-Suicide-Terrorism: Evidence from House Demolitions”. NBER Working Paper 16493.Kaplan, E.H., A. Mintz, S. Mishal and C. Samban. 2005. “What Happened to Suicide Bombings in Israel? Insights from a Terror Stock Model”. Studies in Conflict and Terrorism 28(3): 225-235.

[6] Lyall, Jason. 2009. “Does Indiscriminate Violence Incite Insurgent Attacks? Evidence from Chechnya”. Journal of Conflict Resolution 53(3): 331-362.

[7] Fielding, D. and Anja Shortland. 2010. “‘An Eye for an Eye, a Tooth for a Tooth’: Political Violence and Counter-Insurgency in Egypt”. Journal of Peace Research 47(4): 433-448.

[8] Johnston, Patrick. 2012. “Does Decapitation Work? Assessing the Effectiveness of Leadership Targeting in Counterinsurgency Campaigns.” International Security 36(4): 47-79. Ismail, A., & Amjad, S. (2014). Determinants of terrorism in Pakistan: An empirical investigation. Economic Modelling, 37, 320-331.

[9] Condra, L. and Jacob Shapiro. 2012. “Who Takes the Blame? The Strategic Effects of Collateral Damage”. American Journal of Political Science 56(1): 167-187. Braithwaite, A., and Johnson, S. D., 2012. “Space-Time Modelling of Insurgency and Counterinsurgency in Iraq”. Journal of Quantitative Criminology, 28(1): 31-48. Linke, Andrew, Frank Witmer and John O’Loughlin. 2012. “Space-Time Granger Analysis of the War in Iraq: A Study in Coalition and Insurgent Action-Reaction”. International Interactions: Empirical and Theoretical Research in International Relations 38: 402-425.

[10] Li, Q. (2005). Does democracy promote or reduce transnational terrorist incidents?  Journal of Conflict Resolution, 49(2), 278-297. Chenoweth, E. (2010). Democratic competition and terrorist activity. The Journal of Politics, 72(01), 16-30. Young, J. K., & Dugan, L. (2011). Veto players and terror. Journal of Peace Research, 48(1), 19-33.

[11] Newman, L. S. (2013). Do Terrorist Attacks Increase Closer to Elections?. Terrorism and Political Violence, 25(1), 8-28.

[12] Bali, V. A., & Park, J. (2014). The Effects of the Electoral Calendar on Terrorist Attacks. Electoral Studies.

[13] De la Calle, L., & Sánchez-Cuenca, I. (2013). Killing and voting in the Basque country: An exploration of the electoral link between ETA and its political branch. Terrorism and Political Violence 25(1):94-112.

[14] Berrebi, C., & Klor, E. F. (2006). On Terrorism and Electoral Outcomes Theory and Evidence from the Israeli-Palestinian Conflict. Journal of Conflict Resolution, 50(6), 899-925. Berrebi, C., & Klor, E. F. (2008). Are voters sensitive to terrorism? Direct evidence from the Israeli electorate. American Political Science Review, 102(03), 279-301.

[15] Waldmann, P. (2005). The radical community: A comparative analysis of the social background of ETA, IRA, and Hezbollah. Sociologus, 55(2), 239-257.

[16] Hewitt, C. (1990). Terrorism and public opinion: A five country comparison. Terrorism and Political Violence, 2(2), 145-170.

[17] Sanchez-Cuenca, I. (2007). The dynamics of nationalist terrorism: ETA and the IRA. Terrorism and Political Violence, 19(3), 289-306.

[18] De la Calle, L., & Sánchez-Cuenca, I. (2006). The production of terrorist violence: analyzing target selection within the IRA and ETA (Vol. 230). Centro de Estudios Avanzados en Ciencias Sociales, Instituto Juan March de Estudios e Investigaciones.

[19] Alonso, R. (2006). Individual motivations for joining terrorist organizations: a comparative qualitative study on members of ETA and IRA. Tangled Roots: Social and Psychological Factors in the Genesis of Terrorism, 11, 187.

[20] Kalyvas, S., & Sánchez-Cuenca, I. (2005). Killing without dying: the absence of suicide missions. Making sense of suicide missions, 209-232. Rooney, N. (2007). Violent Nationalism in Catholic Communities: The Provisional IRA and ETA. Studies in Ethnicity and Nationalism, 7(3), 64-77.

[21] Waldmann, P. (1992). Ethnic and socio-revolutionary terrorism: a comparison of structures. International Social Movement Research, 4, 237-257.

[22] Sullivan, C. M. (2012). Blood in the Village: A Local-Level Investigation of State Massacres. Conflict Management and Peace Science, 29(4), 373-396.

[23] http://www.ark.ac.uk/elections/

[24] http://www.euskadi.net/elecciones/

 

 



Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

ENHANCING SECURITY THROUGH COLLABORATIVE RESEARCH

Perspectives on Terrorism is  a journal of the Terrorism Research Initiative and the Center for Terrorism and Security Studies

ISSN  2334-3745 (Online)

Disclaimer, Terms and Conditions