Analyzing the Spell of War: A War/Peace Framing Analysis of the 2009 Visual Coverage of the Sri Lankan Civil War in Western Newswires

Abstract

The goal of this study was to analyze the extent to which the visual coverage of the final stages (April/May 2009) of the long-lasting Sri Lankan Civil War relied on war and peace frames. Based on the revolutionary conceptual work of Norwegian scholar Johan Galtung, who viewed war and peace journalism as two competing frames in covering conflicts and wars, we examined news photographs available from Associated Press (AP), Reuters, and Getty/Agence France-Press (AFP). To date, this topic has been discussed from mostly normative viewpoints, and only little research combined the peace journalism concept with visual framing and/or the role of newswires as gatekeeper of information. We tested this concept empirically by using content analysis of editorial news photographs of the conflict in the three leading Western newswires. By and large, our results suggest that overall visual coverage of the conflict waxed and waned over time, but was primarily driven by visuals originated in the Sinhalese-dominated regions of Sri Lanka. Further, newswires are serving very different purposes and news markets: the AP–and to a lesser extent also Getty/AFP–with their focus on external events (therefore qualifying for peace journalism) and Reuters with a stronger focus on the conflict itself (thus qualifying for war journalism). Overall, the stock photo agency Getty/AFP was found to be the newswire that is most likely to provide media outlets with photographs highlighting peace frames, with the most balanced coverage between the two conflict parties and a particular emphasis on peace demonstrations worldwide, negotiations and summit meetings.

Notes

¹This article is based on an earlier version which was presented at the AEJMC annual meeting in Denver, CO, in August 2010, where it received the Robert L. Stevenson Award (Second Place) in the International Communication Division. The authors would like to thank the anonymous reviewers and Editor Dr. Stephen Perry for their valuable feedback and suggestions for improving this work.

²The Sri Lanka Department of Census and Statistics distinguishes between Sri Lankan Tamils (4.3% of the total population or 732,000) and Indian Tamils (5.1%, or 855,000) (available at http://www.statistics.gov.lk/PopHouSat/PDF/Population/p9p8%20Ethnicity.pdf).

³These photos were taken by Reuters photographers (Sharif Karim, Issam Kobeisi, Mohamed Azakir, among others), and one photograph with a Mickey Mouse in Tyre, Lebanon by an AP photographer (Ben Curtis) (see Morris, 2010).

⁴The fairly high number of photos provided by Getty/AFP is due to the fact that the company (unlike Reuters and AP) is an explicit stock photo agency that has acquired other photo agencies (e.g., Tony Stone Images) or entered into partnerships with already established news agencies (Agence France-Presse [AFP] since 2003) to better promote their editorial material.

⁵It should be noted that many of the newswire photographs were repetitive because the same event was shot from different angles, the photographers used different image orientations (landscape/portrait), additional black and white colorations, or varying image resolutions. In particular, this was often the case for images originated in the Tamil regions, i.e., the major zone of conflict.

⁶Photographs that were taken in the following provinces were coded as ‘Tamil territory’: Northern Province (including all five administrative districts of Jaffna, Kilinochchi, Mannar, Mullaitivu, Vavuniya) and Eastern Province (including the two administrative districts of Batticaloa and Trincomalee). According to 2001 Census estimates, Tamils held the majority in these two provinces (particularly in the North). Additionally, the area that was claimed for the state of Tamil Eelam is largely congruent with the ethnic composition of those districts and provinces. All other photographs originated in the remaining provinces of Sri Lanka were coded as ‘Sinhalese territory’ (for further information see http://www.statistics.gov.lk).

⁷Given children's particularly vulnerable and innocent status in a crisis situation like this, we have argued that the presence of minors in visuals draws the viewer's attention. The exceptions to this, however, are photographs that portray crowds of people (such as demonstrators)–often long-shot photographs which might include minors, but without being the focus of those activities. In addition, the angle from which the picture was taken often helped determine the dominant age group.

⁸Reliability checks for all variables yielded acceptable agreement levels according to Neuendorf (2003).

⁹Given the timeframe used for this analysis and the fact that the Sri Lankan Civil War is a war spanning more than 25 years, we were unable to test whether the newswires adopted a reactive (war frame) or proactive (peace frame) approach to the overall conflict.

Note. Differences in depicting physical harm between Tamils and Sinhalese showed statistical significance, χ²(2) = 38.33, p < .001. Differences in depicting emotional suffering of Tamils versus Sinhalese showed statistical significance, χ²(2) = 54.28, p < .001.

¹⁰The focus on ethnicity (instead of region) was considered more meaningful given that during the selected timeframe Sinhalese soldiers were present in the Northern Province (predominately Tamil) and a great number of Tamil civilians escaped from the Northern regions into the Central and Western (mostly Sinhalese) regions of Sri Lanka.

¹¹It should be noted that our continuum does not range from ‘peace promotion’ to ‘war promotion’. Despite the news value conflicts bear, we simply assume that no journalist pursues the goal of promoting war. Instead, forms of peace promotion may be found along a continuum ranging from more direct forms to rather indirect forms. Therefore, depictions of war cruelties including military movements and civilian suffering do not ‘promote’ war, but may contribute to peace on a more indirect route (for example, by publicly evoking a call for action). Thus we labeled these as ‘war frames’. These may emphasize the status quo more than do peace frames. However, this is not to say that these depictions did not lead to peace at the end. In fact, they might do this vicariously via public responses to a conflict and the resulting public pressure on political elites and international actors.

Note. Differences between the four roles showed statistical significance: χ²(6) = 32.21, p < .001. Differences between war and peace frames showed statistical significance, χ²(2) = 30.51, p < .001.

Additional information

Notes on contributors

Rico Neumann

Rico Neumann (M.A., University of Arizona, 2010) is a graduate student at the Institute of Communication and Media Studies at the University of Leipzig (Germany). He currently works as a research collaborator at the United Nations-Mandated University for Peace (Costa Rica). His academic interests include political discourse and behavior, public opinion and media performance during times of conflict.

Shahira Fahmy

Shahira Fahmy (Ph.D., University of Missouri, 2003) is a tenured Associate Professor in the School of Journalism at the University of Arizona. She holds an adjunct/courtesy joint appointment with the Department of Communication and an additional one at the School of Middle Eastern & North African Studies. Her research interests focus on visual communication and conflict reporting with a specific focus on the Middle East and North Africa.