Srebrenica and the Dutch Wikipedia

With the case against Ratko Mladic underway, the collective blind spot developed by the Dutch for the massacre of Srebrenica is again taking central stage in news coverage.  The argument for a supposed blind spot may be backed up by analyzing the Dutch Wikipedia entry on the Srebrenica massacre, this post offers two visualization tools that can help doing this. Firstly, controversy (or lack thereof) is measured by way of the Digital Methods Bubble Lines Tool. Secondly, the Dutch entry is studied by using the Wordle word cloud tool.

Former British foreign secretary David Owen, when interviewed for the Yugoslavia Tribunal in The Hague, declared that Slobodan Milošević had warned him early 1993 about the possibility of the massacre of Srebrenica if Bosnian Serbians would invade the Enclave. [1]

Despite the prospect of an invasion, tension was broken by the French UN General Phillipe Morillon who told those seeking refuge in Srebrenica: “Don’t be afraid. You are now under the protection of UN troops, we will not abandon you.” After which the UN flag was raised, and the general received honorary citizenship.  Evacuation began in March 1993 but after approximately 23.000 women, children and elderly had been evacuated Muslim leader Naser Orić decided that the enclave should remain strong and 40.000 Muslims were ordered to stay behind. [2]

According to Dutch journalist Geert Mak, all parties involved had given up on maintaining control over the town. In his book In Europe he writes that the lack of courage of the Americans and Western European nations was illustrated by the scarcity of UN troops on the ground, 7000 for all of Bosnia where 35000 were recommended by military analysts.  The Dutch UN troops took over the responsibility for Srebrenica as of February 1994 with no more than 400 lightly armed soldiers. It was no surprise then that on the 11^th of July Ratko Mladić entered the city without encountering any noteworthy resistance, and could start “transporting” the remaining Muslim refugees. [3]

For many weeks Dutch politicians and media insisted that the transportation of trapped Bosnian Muslims had proceeded in an orderly fashion, and that the Dutchbatters responsible for the changing of guard could consider the transition a success. Only when reports of panic and confusion in the surrounding camps reached the outside world, did the press get wind of the genocide that had taken place[4] and started paying attention to the less than heroic role of the Dutch soldiers.

When studying the role of the Dutch soldiers, the cowardice of the in-command  and the gross underestimation of real-life combat situation by the political authorities are frequently part of the explanation.

Former French president Jacques Chirac suggested that ‘l honneur de la nation’ of the Netherlands was damaged, and  French General Bernard Janvier even ventured that ‘things would have gone differently had the French soldiers been stationed there’[5]. He was heavily criticized for this hypocrisy later; it was Janvier who stated during the turbulent days prior to the massacre that United Nations Protection Force (UNPROFOR) troops were there to keep the peace, not defend the enclave in case of an all out attack. In other words, the Dutch were the only ones willing to take on the impossible task of watching over the town, who was he to pass judgment?

Looking at the discrepancies between Wikipedia entries across languages regarding the Srebrenica massacre, the first remark concerns the page title. In most of the researched languages the events described above are displayed under “the Srebrenica massacre”. The Dutch Wikipedia entry (along with the majority of Dutch media coverage)  however, is titled “the fall of Srebrenica”. The Bosnian Wikipedia entry is the only one within the analyzed set of entries which refers to the events as “The Genocide of Srebrenica”. While analyzing the performative dimensions of this quibbling over semantics might fall well outside the scope of this blogpost it is worth mentioning in the context of investigating the implications of the lexicon of collaborative historical reconstruction.

For the Bubble Lines Tool the English Wikipedia entry is used to serve as a point of reference, it being the biggest and most used.  The Dutch entry is used here to test a previously mentioned hypothesis,  to check for blind spots in their collective encyclopedic awareness of the terrible events. The French have been included in the Bubble Lines Tool comparison on account of the aforementioned remarks and because of the  shared responsibility in the UNPROFOR activities in Srebrenica . The Serbian and Bosnian entries are used for the obvious purpose of analyzing the recordings and interpretations of the parties directly involved in the massacre.  The German entry is included in this comparisons because Germany went through a series of well documented changes in foreign military policy during the early years of the Balkan conflict.[6]

Below the number of words for the Wikipedia article per language on the Srebrenica massacre, collected Dec. 2010. The word count runs from the title page up to the See Also section.

The obvious and preliminary conclusion would be that the Dutch entry in Wikipedia for the Srebrenica massacre indeed revealed  a lack of self-examination and awareness of the role of the Dutch (and perhaps the French) in this tragedy. This hypothesis is not easily tested using the Bubble Lines Tool. The circle isn’t a convenient means  to signify differences due to the inflated perception of inequalities in surface area, which just happens to be part of its geometry. Or as prominent expert in the world of statistical graphics, Edward Tufte, writes in the second chapter (on visual integrity in general) of his book The Visual Display of Quantitative Information:[7]

The representation of numbers as physically measured on the surface of the graphic itself, should be directly proportional to the numerical quantities represented. Clear, detailed and thorough labeling should be used to defeat graphical distortion and ambiguity.

In that respect the tool fails us, however it can be used to jumpstart further investigation into the deviation that is visualized here. To zoom in on the Dutch entry is justified by the seemingly enormous difference in word count.

To confirm a larger than average difference,  which could be reason to further investigate the Dutch entry on the Srebrenica massacre, one would first have to take similar articles and configure a baseline from which further contextualization of the differences can be derived. Doing so entails a collection of similar events, and compare the difference between the English and the Dutch entries in terms of word count to confirm a larger than average difference for the Srebrenica entry.

The website History & Policy is a collaboration between the University of Cambridge, The Institute of Historical Research, and The London School of Hygiene and Tropical Medicine. On it, an article by Senior Lecturer in History Lisa Pine[8] can be found, classifying the Srebrenica Massacre as one of the major genocides of the 20^th century.[9] Which leads to a set of four historical events serving as a reference point for what could be stipulated to be the standard difference between the Dutch and English Wikipedia entries on 20^th century genocide.

As illustrated above, the larger than average difference observation initially observed from the Bubble Lines Tool, holds when ascertaining through comparison with similar articles. But like the Bubble Lines Tool, these irregularities only serve to initiate further investigation.

The Bubble Lines Tool is used to generate indicators of societal concern and controversy, and in the case of the mentioned Wikipedia entry it does just that.  It does not provide for anything beyond an initial signaling of a present silence, and one would be hard pressed to mould this into a workable hypothesis.

Time and space do not permit me to find theoretical foundation for the assumption that word count, spent pixels and other signifiers, is tied up into overall levels of concern. To further investigate the manifest text as well as on the lapses, distortions, silences and absences one needs to move beyond word count alone. Move from how much is being said, to what is being said. The only visual method of doing this is by using an online tool like Wordle. This tool is used to generate “word clouds” from user-provided texts, such as plain text files or, in this case, Wikipedia entries. The clouds below, generated December 2010, give greater prominence to words that appear more frequently in the source text:

Hesitant as I am to ignore professor Tufte’s guidelines,  and because of the possibility that over-reliance on the familiar word cloud tool has turned it into that hammer that desperately needs a nail to pound, there still are noteworthy differences in the visual representation of both entries to point out. In particular the size of the word genocide, in the top right corner of the Dutch word cloud, and the absence of a Dutch translation for the English word massacre. The number of military related terms in the Dutch entry, General/Airsupport/Dutchbat/Soldiers/Armored vehicle, are also relevant when coming to a possible hypothesis based on the word cloud. In the broadest possible strokes that hypothesis could be; the Dutch entry is concerned with a technical recollection of the diplomatic and military proceedings surrounding the fall of defenses around the town while the English entry describes the massacre itself, the plan, the deportation of Bosnians and the mass executions.

Further study into the present text of the Dutch entry would, in light of the proposed hypotheses, focus on the section dealing with claims of responsibility and blame. The inclusion of a section on those responsible (De Verantwoordelijken) in the Dutch entry may be seen as confronting  a possible claim without the risk of giving it dangerous and unnecessary visibility. In  this way, the criticisms are answered without the questions themselves having been formally posed. This is the sort of observation that those familiar with the analysis of visualized information and committed to a specific model for knowledge acquisition, probing for the symptoms of a text , can make and try to contextualize.

Including the paragraph on the  civil law actions, initiated by the Bosnian plaintiffs in order to rule that the UN and the State of the Netherlands breached their obligation to prevent genocide, hints at a similarly motivated transparency. The omissions in the Dutch entry however, show another kind of silence, one that is easily hypothesized to be very personal rather than hegemonic or ideological. At present  this paragraph in the  Dutch entry only deals with a rationale for the absence of air support and not with the more specific and personal accusations by former UN employees of Dutch complicity in the massacre as seen in the English entry.

[2] Mak, Geert. In Europa. Uitgeverij Atlas: Amsterdam, 2004.

[3] Ibid.

[5] As recorded by the German weekly magazine Der Spiegel. Ludwig, Udo. & Mering, Ansgar. ‘Screbrenica Survivors Sue Netherlands, United Nations.’ Online publication, 2007.  Available at: <http://www.spiegel.de/international/world/0,1518,486755,00.html> Accessed December 2010.

[6] In the light of Srebrenica, it was now widely accepted in the German political elite that the legacy of German history should not only be to call for ‘No more wars!’ (‘Nie wieder Krieg!’) but also for ‘No more Auschwitz!’ (‘Nie wieder Auschwitz’!). This move has been argued for and described in; Bauman, Rainer & Hellmann, Gunther. ‘Germany and the Use of Military Force: ‘Total War’, ‘the Culture of Restraint’, and the Quest for Normality.’ German Politics, 10, volume 1, 2001.

[7] Tufte, Edward R., The Visual Display of Quantitative Information. Cheshire: Graphics Press, 1983.

[8] Senior Lecturer in History in the Department of Social and Policy Studies at London South Bank University.

[9] Pine, Lisa. ‘Genocide: Twentieth-Century Warnings for the Twenty-First Century.’ A Policy Paper for History & Policy. 2008. Available at: <http://www.historyandpolicy.org/papers/policy-paper-69.html> Accessed December 2010.