Transparency and Casualty figures
Harshula <harshulaj@gmail.com>
Last updated: January 20, 2011
1 Those who seek transparency must also be transparent
Sri Lankans in general are very skeptical about politics and politicians. This skepticism now appears to extend to the United Nations (UN) and international non-governmental organisations (INGOs), for a number of reasons. One of those reasons is the lack of transparency.
The casualty figures for the last few months of the separatist war in Sri Lanka are mired in controversy. The UN has made no attempt to publish their methodology nor address the concerns of those that contradict the UN figures.
John Holmes, Under-Secretary-General for Humanitarian Affairs and Emergency Relief Coordinator of the United Nations, in a CNN interview on December 14, 2009, states:
AMANPOUR: Let me ask Sir John Holmes. There are, obviously, huge allegations about the — the force that the government employed to crush the Tamil Tigers. What are you saying? The ambassador is denying that there were 7,000 or more deaths.
HOLMES: Well, we always said that we didn’t know exactly how many people have been killed. Seven thousand was an internal estimate. The number may be lower; it may be higher. What is clear is the civilian casualties were unacceptably high.
A number of months earlier, an anonymous UN source claimed that 20,000 civilians died:
Confidential United Nations documents acquired by The Times record nearly 7,000 civilian deaths in the no-fire zone up to the end of April. UN sources said that the toll then surged, with an average of 1,000 civilians killed each day until May 19, the day after Velupillai Prabhakaran, the leader of the Tamil Tigers, was killed. That figure concurs with the estimate made to The Times by Father Amalraj, a Roman Catholic priest who fled the no-fire zone on May 16 and is now interned with 200,000 other survivors in Manik Farm refugee camp. It would take the final toll above 20,000. “Higher,” a UN source told The Times. “Keep going.”
The figure and the methodology reported by The Times (London) was criticised by others within the UN according to an article in The Guardian (London):
Citing a UN source, the Times reported that 20,000 people were killed in the final throes of the civil war, mostly by government shelling. However, one UN official described the method used to calculate the toll – which appeared to involve multiplying the figure recorded by doctors by five – as dangerous extrapolation. The Sri Lankan government has rejected the Times report.
...
Officially, the UN spokesman in Colombo, Gordon Weiss, would say only that "we have always said many thousands of people died during the conflict". But privately, UN staff admitted they were puzzled by the methodology used to achieve the new death toll.
"Someone has made an imaginative leap and that is at odds with what we have been saying before," one official said. "It is a very dangerous thing to do to start making extrapolations."
It was also reported in Le Monde that the 20,000 figure was an “unaudited” “extrapolation”:
But Vijay Nambiar has been informed that the final toll "may exceed the 20 000 dead". This figure, unaudited, is an extrapolation of the limited information available.
To add to the confusion, the former UN spokesman in Sri Lanka, Gordan Weiss claimed in an interview with SBS:
GORDON WEISS: I believe that between 10,000 and 40,000 is a reasonable estimate. I think most likely it’s somewhere between 30,000 to 40,000.
The controversial circumstances in which Weiss’ employment with the UN was terminated politicises the issue further.
It was reported in The Guardian (London) that the UN relied on Tamil Government doctors working within the Liberation Tigers of Tamil Eelam (LTTE) controlled areas for accurate casualty figures:
Gordon Weiss, a UN spokesman, said reliable reports from inside the war zone had dried up after the "courageous" doctors who had been working out of the last makeshift hospital at Mullaivaikal East primary school were forced to abandon the building in the face of heavy fighting on Friday.
Later it was reported by the BBC that the doctors claimed the casualty figures they reported were the result of pressure by the LTTE, however, this claim was made whilst they were in Government custody:
"Every day the LTTE (Tamil Tigers) people came to the hospital, they gave the list," he said. "This amount got injured, this amount dead, this area shells fell. We had to tell that list. Read it out. The list was wrong, exaggerated number."
Dr V Shanmugarajah said that on one day, some 60 people were killed but they were instructed by the rebels to say 1,000 were dead.
It is completely plausible that the doctors were threatened by the LTTE to mislead the UN, but it is also plausible that they were under duress by the Government to recant their previous statements. Or the truth could be a combination of both.
Then there is the International Crisis Group’s estimate which claims that the death toll is “certainly high enough to triple the UN’s internal figure of 7,000”. Though they state it is based on “credible evidence” that there were 330,000 internally displaced persons (IDPs), no sources are provided:
While determining a conflict death toll is always a task to be approached with caution, there are multiple reasons to believe that an estimate of tens of thousands of civilians killed is reasonable.
First, a comparison of the number of civilians known to be in the Vanni during the fighting with the number known to have reached government internment camps by the end of the fighting provides a rough estimate of those killed or missing.
Crisis Group has credible evidence that there were 330,000 displaced civilians in the second NFZ and adjacent areas as of mid- to late-February. At that same time, according to the government, there were between 33,000 and 38,000 internally displaced persons (IDPs) from the Vanni already in camps in government-held areas. That brings the total of known civilians to about 365,000.
By the end of the fighting, the highest number of survivors registered in government-run camps was 290,000 as of 26 May 2009. It is not clear to what extent this number includes the more than 10,000 who had surrendered or were detained on suspicion of involvement with the LTTE. There may also have been several thousand people who fled by boat to India or otherwise escaped the war zone but avoided being detained in government camps.
Accepting these uncertainties, there is a plausible case that as many as 75,000 persons remain unaccounted for. Even if the figure of 330,000 is reduced by as many as 30,000, or some adjustment to the difference between 330,000 and 290,000 is made to account for civilians who may have been killed lawfully because they were directly participating in hostilities at the time they were targeted – a number Crisis Group believes is very low – or to account for some number that may have avoided the government camps, it is still difficult to arrive at a figure for the killed or missing that is lower than 30,000.
Moreover, Crisis Group has evidence from various individuals who were in the NFZs until the very end of the fighting to suggest that the scale of death was much higher than reported at the time, and certainly high enough to triple the UN’s internal figure of 7,000. Crisis Group also believes that all but a small portion of these deaths were due to government fire. Eventually, an independent and impartial survey of those still living in the Northern Province will be needed to establish this part of Sri Lanka’s history.
Lastly, the University Teachers for Human Rights (Jaffna) (UTHR(J)) reports the following:
3.5 A Note on Casualty figures
Almost any discussion of figures has been fraught with so much polemics that the truth lay buried. In the first place the Government failed to allow in international observers and severely underestimated population figures to justify harsh military action. The Government claimed the population in the No-Fire-Zone created on 12th February 2009 was 70 000. Most others agreed this was an underestimate, but in their own estimates tended to err on the side of caution. When the International Crisis Group in early March quoted the ICRC figure of 150 000, Prof. Rajiva Wijesinghe of the Government Peace Secretariat launched an acerbic attack on the ICG. When the Army breached the NFZ on 19th April resulting in about 100 000 civilians escaping from the LTTE, the Foreign Minister said that only 15 to 20 thousand people remained in the truncated NFZ. When this too was breached about 80 000 survivors came out. Against the 70 000 originally stated by the Government there must have been nearly 200 000.
Challenged by government obfuscation, anyone interested in what was going on had to take casualty figures given by the LTTE seriously as a first approximation knowing that its intention was to play up the charge of genocide. Apart from this there was a real problem with gauging civilian casualties because the LTTE figures or figures it influenced appeared to conflate civilians, conscripts and combatants. Following the end of the war, the Government claimed that nearly 23 000 “terrorists” were killed. Taking this figure, along with how the LTTE was making up its fighters, it would not be wide off the mark to say that more than 13 000 of this number were made of recent conscripts who did not want to fight and family men forced to dig bunkers. This would account for a huge chunk of the total dead. (This is why we said that a responsible government should have relied primarily on political measures that have not even been tried.)
Thus although figures attributed to the UN have been considered authentic, there are many imponderables, such as who gave the figures from the ground and whether they adequately distinguished between civilians and (enforced) combatants. From the third week of January 2009 until the end of March civilian casualties were relatively high because there was constant fighting as the Army advanced and the LTTE fired from among civilians.
The ICG citing unofficial figures from the UN gave the number of civilian dead from late January to the first week of March as 2300. One did not find it easy to reconcile this with other indicators. During the same period the ICRC transferred 1500 surgical cases from the war zone to Trincomalee Hospital. This would suggest naively about 800 civilian dead during this period. For another indicator, TamilNet reported on 11th April that hospital records showed that 599 civilians died over nearly the past month (from 13th March) and went on to comment that the total may exceed 1500 taking into account deaths not reported to the Hospital. That gives a wide margin. This is why it is viciously stupid for the Government to impugn and detain the doctors who were giving fairly objective reports with conservative casualty figures.
The Timesonline [10] report of 29th May citing UN sources, which gave the total killed for the year as 20 000, with 13 000 of the dead in May, said “the numbers killed by shelling doubled from March to April, with an average of 129 every day.” In contrast our Bulletin No.47 published on 17th April stated, “Persons in regular touch with those who have escaped confirm that an average of 15 to 20 people die each day; either killed by shells or shot by the LTTE attempting to drive fear into would-be escapees.” That huge discrepancy cannot be bridged even if one adds combatant deaths. Also averages are misleading. April was a relatively quiet month until the Army entered the NFZ on 19th April. TamilNet said its correspondent personally witnessed 300 bodies while fleeing the area and an email from one of the doctors said that 700 injured had been taken to the Mullivaykkal temporary hospital. TamilNet reported again on the 29th April and 2nd May that 300 were feared killed in the first instance and 64 in the second.
To get some idea of the casualties, after 19th April, the ICRC removed nearly 1000 patients and carers in two trips until 7th May and TamilNet reported that a further 500 injured were waiting for transportation. The next two major incidents involving civilian casualties from army action were on 9th and 15th May. On the 9th TamilNet estimated 2000 dead with 257 dead bodies and 814 injured brought to hospital. Dr. Shanmugarajah told AP that the dead and dying brought to hospital accounted for 430 deaths but the total was likely closer to 1000. He called it also the bloodiest day he had witnessed. On the 15th, a volunteer doctor reported seeing 800 bodies in the area where he moved. We quoted the TamilNet only as a useful indicator, although figures give by the doctors were, when available, significantly lower.
We may say that there was a fairly regular pattern until the end of March while the Army advanced and the LTTE and the people were pushed east. There was in the first 18 days of April a lower level of deaths due to army shelling, while the LTTE shooting at escapees continued. From 19th April there were as indicated above days on which deaths were high, 300 or more, notably on 9thand 10th May when it touched nearly 1000. On the 14th many escapees died due to LTTE shelling. On the 15th there was heavy fighting until about mid-day, resulting in heavy casualties among combatants, as well as civilians caught up in it. Allowing for some exaggeration by TamilNet, the civilian dead was probably below 1000. As discussed earlier, the heaviest civilian casualties were during the last battle from the 17th evening to the 18th morning. Wile many civilian casualties resulted from LTTE shelling, the Army too by its harsh approach on this occasion caused many deaths totalling 1000 to 4000.
In conclusion, the civilian casualties after 19th April need careful research. Speaking of averages is meaningless and of a hidden massacre is misleading. Such an expression immediately puts the government forces in mind. We know that on the May 14th and May 17th night, the LTTE was to a large extent responsible for civilian deaths. When an organisation fell apart in that manner, one section joined the civilians and escaped, another section feeling abandoned by the leadership and facing certain death from a Government that did not want to accept surrender, lost its balance and was angry with the others seeking to protect their life. We also pointed out that in giving casualty figures, the distinction between civilians, conscripts and cadres has not been clearly made.
We do not wish to underplay numbers, but focusing on numbers may result in unnecessary polemics to the cost of the deeper issues. If we avoided a worse catastrophe, it was due to unplanned factors. For the Government to launch the operation in this way was wrong and utterly irresponsible. No one could think well of a government that bombs and shells its own civilians. That point need not be made by quoting high figures of casualties. The only accurate means of finding out casualties is to count and alternatively to have a clear idea of what was happening on the ground. In their absence, technology and statistical formulae may turn out to be very misleading. Another important indicator is that the people who escaped during the last week of the conflict blame largely the LTTE, towards which their anger is directed.
...
What gives ground for satisfaction is that international pressure had a benign effect in protecting civilians. In the context of present war which took a heavy toll on the lives of soldiers, these ordinary men have shown remarkable restraint towards civilians when they come to contact with them. The civilians are uniformly scathing about the LTTE, and frequently found the Army helpful and considerate. So far we know of no major discrepancy in the civilian population before the fall of Killinochchi and after the war’s end. The population before was estimated at 300 to 350 thousand. Presently nearly 300 000 are accounted for in IDP camps. It would be some time before all are registered and detailed checks could be made. There is no evidence of genocide. It is hard to identify any other Army that would have endured the provocations of the LTTE, which was angling for genocide, and caused proportionately little harm.
Until the UN can provide some clarity about the methodology and sources used to arrive at the estimated death toll, one should not be surprised if Sri Lankans question the credibility of the UN.
2 Version History
-
Version 1 - December 14, 2010
-
Version 2 - January 20, 2011
Footnotes