Social network analysis and counterinsurgency: A counterproductive strategy?

Roger Mac Ginty

Research output: Contribution to journalArticlepeer-review

Abstract

Social network analysis has been adopted by a number of governments in their counterinsurgency campaigns. By using network analysis, security agencies claim they can render militant groups impotent by targeting 'nodal points' or key links in insurgent networks. The article makes three arguments on the potentially counterproductive nature of social network analysis-assisted counterinsurgency campaigns. Firstly, social network analysis may be 'too successful' in stripping militant movements of a cadre who could negotiate a peace accord. Secondly, social network analysis-assisted campaigns fail to address the root causes of violent conflict. Thirdly, by denuding communities of social capital and social entrepreneurs, social network analysis - as a counterinsurgency tool - may condemn communities to underdevelopment and failed post-war reconstruction. In short, the 'magic weapon' of social network analysis might actually prolong the conflict it is supposed to help quell. The article employs the government of Sri Lanka's social network analysis-assisted counterinsurgency campaign as a case study, though it also has application to other cases. It concludes by considering if social network analysis can be put to more constructive uses, specifically in the rebuilding of communities after violent conflict. © 2010 Taylor & Francis.
Original languageEnglish
Pages (from-to)209-226
Number of pages17
JournalCritical Studies on Terrorism
Volume3
Issue number2
DOIs
Publication statusPublished - 2010

Keywords

  • Counterinsurgency
  • Peace-building
  • Post-war reconstruction
  • Social network analysis (SNA)
  • Sri Lanka

Research Beacons, Institutes and Platforms

  • Humanitarian and Conflict Response Institute

Fingerprint

Dive into the research topics of 'Social network analysis and counterinsurgency: A counterproductive strategy?'. Together they form a unique fingerprint.

Cite this