The Weak Points of Statistical and Demographic Analyses in Estimations of War Victims in Bosnia and Herzegovina in the Period 1992-1995

Abstract

In the political and war crisis which embraced Bosnia and Herzegovina in the spring of 1992 with an end of war hostilities in the autumn of 1995 when the "Dayton Peace Agreement" emerged (November 1995), a media war occurred. From the very beginning, this war had an international character. The question on the number of war victims (killed and missing) "exploded" in June of 1993, when Haris Silajdžić stated that there had been 200000 dead among the Muslims. This figure uncritically became the basis for all later media and local "empirical truths" on the number of victims. All statistical and demographic disciplines were exploited to support, if not prove, the propaganda standpoints. Objectivity was oppressed by an ugly "face of the war". Having in mind the experience of the Second World War in Yugoslavia, the question on the number of victims does not cease to be topical for decades after the end of the war. Bosnia and Herzegovina is more than a confirmation. This question seems to intervene (and in a way "feed of") with the most difficult political and international questions and court trials. ("International Court of Justice", indictment of Bosnia and Herzegovina against The Federal Republic of Yugoslavia, namely Serbia).

The methodological analysis of the most important works which deal with the question of the number of victims in the Bosnian war (above all, those done by Bosnian institutes and authors) indicate the "mistakes" made by the character of these works (propaganda). The manipulation with statistical methods and numbers is not new. Methodological and numerical traps can slip even to the most informed.

The use of statistics and social science in court trials seems to show Janus's face of science: on one side the authentic "moral passion" of researchers finds great sense, and on the other side special interests strive to impose themselves through the (most refined) instrumentation of science and knowledge. (The example of Mr. Patrick Ball's testification in the trials in the Hague Tribunal is edifying as regards the question of the reasons for the Albanian exodus in the war crisis on Kosovo and Metohia in 1999).

This analysis points out to the crucial defects of every statistical (and demographic) procedure of deriving the number of war victims in the absence of a comparable population census after the war (which did not take place in Bosnia and Herzegovina). The qualification of the quality of the 1991 Census in Bosnia and Herzegovina is briefly given (the author was an expert and organizational leader of all operations of last censuses in former Yugoslavia, 1991).

Probably the most distinctive point, in the continuous course of deriving numbers and analysis on the number of victims in the Bosnian war so far, is the text of George Kenney published in the NY Times Magazine, April 23^rd  1995.