This is an Annotation for Transparent Inquiry (ATI) data project.
The annotated article can be viewed on the publisher's website.
The publication undertakes an analysis of the visual politics of war experience within the wider social (re)production of war. Within this it traces in particular how dominant accounts of war can be complicated and contested. In order to explore these issues the article focuses on the release by WikiLeaks of what became known as the ‘Collateral Murder video’ – gunsight footage from an American Apache helicopter showing an air to ground attack in July 2007 in New Baghdad, Iraq. The analysis for the article centres on the WikiLeaks website devoted to the video, and associated materials - chiefly press coverage from the time of the video’s release. Discourse analysis was used to identify ways of seeing (conceptualised as visual - or scopic - regimes) interacting within the body of texts. The WikiLeaks website itself consists of eight pages, but within these pages are four embedded videos (between 1:30 and 39:14 minutes in length), collections of still images, transcripts, timelines, and links to associated materials, as well as contextualising narrative text. This website, along with the media reports accompanying its release, therefore formed a very rich text, comprised of various visual elements and accompanying written description (the use of the phrase ‘Collateral Murder’ being one over-arching example). These elements, and their contrasting visual perspectives, were juxtaposed across the website and it was the politics of this juxtaposition that formed the focus of the analysis.
Handling and analysing the empirical material
The practical management of this article’s evidentiary material was as follows. The core text (the WikiLeaks ‘Collateral Murder’ website) was augmented with related relevant material, chiefly reporting and commentary in the media at the time of the release of the footage. All of this material was in the public domain and readily accessible. Whilst media reporting of the release was initially considered for its broader context it ultimately evidenced key points made for the article. The collection of media reports were compiled using general search engines and internal searches of news and media organisation websites.
The material (with the obvious exception of the footage) was printed and then qualitatively annotated in hard copy over numerous sessions of analysis. Analysis was grounded in a discourse analytic approach to the identification of particular ways of seeing (conceptualised as scopic or visual regimes) and their juxtaposition and interaction. Having watched the footage repeatedly I was able to identify key visual moments that were particularly relevant to the questions being asked and that could be used to evidence the arguments being made. These key moments were rendered as stills and printed for further annotation in hard copy (two of these stills is reproduced within the article). Through this process of textual immersion and the use of a discourse analytic approach, I was able to identify three interacting modes of seeing within the texts: the view from above, the view from below, and the view from the on-the-ground eyewitness. The analytical objective was to trace how different modes of seeing were juxtaposed to create the footage as ‘Collateral Murder’, an intervention that disrupted the ‘view from above’ through which western audiences typically ‘know’ war. This enabled broader conclusions to be drawn concerning the ways in which war experiences are a material of political currency, invoked, appropriated, and ‘written’ in particular configurations to produce knowledge of, and underpinning, war.
The logic of annotation employed for this publication is as follows. Firstly, the annotations are designed to make the underlying evidentiary material easier to find. More specificity is therefore added to delineate which precise elements within the Wikileaks website (in particular) are drawn upon in specific passages of the article. Secondly, further source excerpts are provided to deepen the evidence base and contextualise key quotes or examples already appearing in the text.
Further considerations
The account above tells a neat story about the publication, the broader question it seeks to answer and how doing so was enabled by an evidentiary base. However, in a manner that will be recognisable to many, these elements of neatness belie a complexity and richness in the process of engagement with the material that is not captured, but is instead smoothed over, by the account above. A neat narrative of how the collection, handling and analysis of data figures within the timeline of a piece of research is often some permutation of the following. The researcher has a research question. They then survey the field of available and relevant data, ‘sample’ in some manner to narrow the field and then proceed to the collection of the data, which is then prepared for analysis. The data is analysed and the results written up in the form of the article.
I first watched the ‘Collateral Murder’ video when it was released in April 2010. I was a person watching over a dozen other people being dismembered and destroyed by a 30mm cannon. It is important not to elide things like this when we talk about ‘data’ and ‘empirics’. ‘Collateral Murder’ seemed very significant, but I was unsure about the form of that significance. I returned to ‘Collateral Murder’ repeatedly. Inevitably I was attempting to make sense of it through questions that frame my academic work, questions about the production of knowledge underpinning war. I found ‘Collateral Murder’ very challenging to theorise. I went back to the material again and again over five years. The phrase I used above, ‘numerous sessions of analysis’, does not quite capture the way in which this material lived with me over those years. My allusion to qualitative annotation in hard copy does little to evoke the iterative, open-ended journeying between questions, ideas, concepts, emotions, printed pages, and dynamic footage that would manifest in those notes and be further played out over the multiple drafts of this article. Whilst I would hope that the neater account of empirical management and analysis that I provide above is clarifying and pragmatically useful, it is only a part of a richer, life of and with the ‘data’ that exceeds clear narrative.