Political Bias Lurking in Fatality Statistics Published by Syrian Observatory for Human Rights

CMEPS-J Report No. 1 (Updated on July 24, 2015)

CMEPS-J Report No. 1 (Updated on July 24, 2015)

AOYAMA Hiroyuki and HAMANAKA Shingo

Introduction

In Syria, where the “Arab Spring” flared in turmoil in March, 2011, the situation deteriorated as violence broke out from within and without the country and sanctions and interventions from Europe and the US added to straining local people’s lives further. In the prolonged conflict described as the “century’s worst human rights crisis”, over 230,000 people perished while some 4 million abroad and 6 and a half million in the country have been forced to refugee status, with 10 million people are said to have been suffering. These estimates, especially the fatality figures, are based on the statistical data published by the Syrian Observatory for Human Rights.
  The Syrian Observatory for Human Rights (SOHR, al-Marṣad al-Sūrī li-Ḥuqūq al-Insān in Arabic, http://www.syriahr.com/), is a human rights organization established in 2006 by Rami Abdelrahman, a Syrian resident in England, that has kept transmitting information on the human and property damage done and human rights infringed since the outbreak of the conflict. While almost all similar Syrian human rights networks like the Syrian Network for Human Rights (SNHR, al-Shabaka al-Sūrīya li-Ḥuqūq al-Insān in Arabic, http://sn4nr.org/arabic/), cover only limitedly fatalities suffered by the Syrian regime side and violent incidents caused by opposition forces, the Syrian Observatory for Human Rights appears to collect and report damage done in the conflict exhaustively. Therefore, the publication of the Syrian Observatory for Human Rights has been referenced by various UN organizations as a main source of critical data of conflict damage, and influential media companies like AFP are relying on the publication for their reportage of conflict situations.
  However, certain suspicion has been raised about the neutrality of this organization since the spread of “misinformation” related to the so-called “Houla Massacre” in Hims reported in May, 2012. In this report, we try to focus our attention on the fatality statistics as provided by the Syrian Observatory for Human Rights that has drawn ambivalent evaluations, review how critical data are to be collected and determine whether or not a political bias has worked its way into the published statistics.

What Variances in Fatality Breakdowns Suggest:

First, let’s take a look at the manner of collecting data for fatality statistics employed by the Syrian Observatory for Human Rights. In Table 1 and Figure 1, a summary is given of the major fatality figures published by the Syrian Observatory for Human Rights since the outbreak of the Syrian conflict in 2011.

Figure 1. Changes in Syrian Conflict Fatality Figures

画像の説明
(Source) Home page of the Syrian Observatory for Human Rights, AFP, al-Ḥayāt, etc., as summarized by the authors.

Table 1. Changes in Syrian Conflict Fatality Figures (unit: person)

画像の説明
(Source) Home page of the Syrian Observatory for Human Rights, AFP, al-Ḥayāt, etc., as summarized by the authors.


  Figure 1 indicates clearly that the fatality jumped sharply since the latter half of 2011 as the military strife went into full swing and also that the numbers of civilian victims accounted for over half of the fatalities. Table 1 signifies, especially since 2013 when the fatality breakdowns were made public officially, that confrontations in the conflict gained in complexity as the multi-layer strife became characteristically obvious.
  The above can be confirmed by following the changes in breakdown appellations used in the published data. Table 1 did not make specific entries, but “pro-regime militia”, “other pro-regime alien belligerents” and “alien combatants” (Al Nusra Front, Daish, etc.) in 2014 underwent changes from 2013 in appellations to reflect the added diversity in parties involved.

Table 2. Changes in Breakdown Appellations

画像の説明
(Source) Home page of the Syrian Observatory for Human Rights, AFP, al-Ḥayāt, etc., as summarized by the authors.


  On the other hand, looking at “civilians” in the fatality breakdown, we note that certain special consideration has been made, which could be criticized as politically biased. The Syrian Observatory for Human Rights, since mid-2012, has categorized “civilians” into “civilians and dissident belligerents” (madanī wa muqātil muʻāriḍ”, but in 2013 they created sub-divisions of “minors (younger than 18)” and “women of age 18 and over”, along with “combatants of combat troops” (muqātil al-katāʼib al-muqātila), under the “civilian” category, which at 2014-end was expanded to “combatants of combat troops and Islamist troops” (muqātil al-katāʼib al-muqātila wa al-islāmīya).
  If dissident militia groups and jihadists (Islamist militants) are deemed as “militant civilians”, then pro-regime militias that do not belong to Syrian army or security forces must be included in “civilians”. But, it is suspected that the Syrian Observatory for Human Rights make an arbitrary differentiation between civilians and combatants, depending on the political and ideological propensity of the civilian.

What Quantitative Analyses Suggest:

Next, we will review to determine statistically whether the Syrian Observatory for Human Rights has employed an arbitrary classification of fatalities or not. To begin with, we have calculated the differences in fatalities for one year from the previous fiscal to the current between fiscal 2014 and fiscal 2015, since the changes in fatalities listed in Table 1 represent the cumulative frequency. Figures 2 and 3 list the fatality differences by groups as shown in two groups: one group for minors (younger than 18), women of age 18 and over, Syrian army and pro-regime militia and defectors, and the other for dissident militia group combatants (inclusive of jihadists), Hizbullah combatants, other pro-regime alien combatants, alien combatants (Al Nusra Front, Daish, etc.), and the unidentified (inclusive of combatants). Special care was taken in classifying the groups so that the pro-regime and dissident sides were comingled in each group to avoid biasing total figures by groups.

Figure 2. Changes in Syrian Conflict Fatality Figures (Group 1)

画像の説明
(Source) Created by the authors, based on Table 1.

Figure 3. Changes in Syrian Conflict Fatality figures (Group 2)

画像の説明
(Source) Created by the authors, based on Table 1.


  Looking at these Figures, we note that the publication periods were not constant and the fatalities varied as the fighting intensified. In particular, Figure 3 indicated a reduction in the fatality of dissident militia groups for February 2, 2015, as compared to that for December 2, 2014, which might suggest an arbitrary classification revision by the Syrian Observatory for Human Rights.
  Now, let’s address how we may find systematic evidence to determine if the Syrian Observatory for Human Rights classifies “civilians” by their political and ideological propensity for inclusion in the fatality of combatants. Here, we take any 2 adjoining periods at will and review whether the increase in fatalities during these periods is constant or not for the above two groups, regardless of their breakdowns. The two periods chosen at random were in 4 pairs; (1) April 1 and May 19, 2014, (2) August 21 and December 2, 2014, (3) February 7 and March 15, 2015, and (4) April 16 and June 9, 2015. We conducted the chi-square tests of independence test on two groups for two randomly chosen periods. If the classification is free of arbitrariness, the cumulative fatality should increase monotonously, regardless of the breakdown classifications within the group. Considering the possibility of fatality concentration in a specific group depending on the combat intensity, we conducted the tests on two groups by 4 pairs of period, with the 5% significance level.
  Table 3 shows a cross tabulation and the results of chi-square tests on two groups by 4 pairs of period. Group 1 indicates increases monotonously in cumulative fatalities for all two periods regardless of the breakdown classifications. But, Group 2 shows differences in changes in cumulative fatalities by breakdown classifications for all two periods of the 4 pairs. For all 4 pairs, the null hypothesis, the groups and the periods are statistically independent, was rejected in chi-square tests at the significance level of 5%. These results usually do not happen even if the fatality may concentrate in specific groups.

Table 3. Cross Tabulation of Syrian Conflict Fatalities (unit: person)

画像の説明
(Source) Created by the authors, based on Table 1.


  Based on the results of the quantitative analyses, we may safely conclude it is highly possible that the Syrian Observatory for Human Rights has arbitrarily differentiated civilians from combatants by their political and ideological propensity.

Conclusion

We have confirmed the signs of arbitrary operation by the Syrian Observatory for Human Rights in their fatality statistics in both the variation of data collection mannerism and the changes in statistical figures per se. The Syrian conflict, as has been shown by the “Arab Spring” in Arabic countries that originated it, is generally captured as a confrontation of “the long-term autocratic regime” by “democratization” and a systemic transformation has been considered a matter of course. However, it has become acknowledged at long last in recent years that such a concept of pre-established harmony as based on the rewarding-virtue-and-punishing-vice morality has no place in reality as Arabic nations including Syria deepen their woes in turmoil. That said, however, the political bias that lurks in the data supplied by the Syrian Observatory for Human Rights revitalizes an extremely simplified, stereo-typed “Arab Spring” and is hampering correct understanding of the Syrian situations.
  Data that are politically tendentious should not essentially be referenced in any effort for grasping the true picture of the conflict reality. The Syrian conflict, in particular, has been tainted as an “information war” that is waged through “agitative broadcasting” by satellite TV stations, the Internet and other social networking systems, propagating intelligence and influencing the course of events. As such, neutral information and recognition of current state of affairs based on correct data are crucial. As one of “the most reliable information” sources relied upon by various UN organizations and media companies in Europe and the US, the Syrian Observatory for Human Rights needs to be aware of the political value attached to any and all pieces of information on the Syrian conflict that is known as the “information war”.

CMEPS-J_report1