The UN’s count is based on data from these 8 sources:
‘1. The March 15 Group
2. The Syrian government
3. The Syrian Center for Statistics and Research
4. The Syrian Network for Human Rights
5. The Syrian Observatory for Human Rights
6. The Syrian Revolution General Council
7. The Syria Shuhada Website
8. The Violations Documentation Centre, the documentation arm of the Local Coordination Committees’.
All of them, barring the Syrian government, are opposition sources. If you don’t want to take my word for it, even the most cursory of Google research efforts will reveal that to be true. The report states that ‘each data collection organization determines their own methods for data gathering and verication‘ (p.4, emphasis mine).
It hardly needs pointing out, then, that problems could arise when the vast majority of your raw data is being provided by one side in the conflict.
The report itself does touch on this potential for innaccuracy, when it says that:
‘There are a variety of ways in which a record may be potentially inaccurate. For example, some records may describe people who died of non-conlict-related causes; in the context of a database of killings, these records are potentially inaccurate. For example, victims of accidents, or illness mis-takenly included in lists of conflict-related killings would be one kind of inaccurate record. Another example includes victims who were believed to be dead but are later discovered to be alive. Individuals who were missing following a violent event, or who disappeared for some time may have been mistakenly recorded on a list of conflict-related killings. There is also the possibility that some records are fabricated, that is, records of victims who do not in fact exist at all. Although HRDAG is only aware of a small number of specic examples of inaccurate records (those provided by VDC), it is possible that some additional inaccurate records are included in these data’.
Indeed, the report says that people can contribute to the database of one of the sources used for the count, the Syrian Center for Statistics and Research, by filling ‘out a form on the SCSR website to add victim information’ (p.18) – which doesn’t strike me as the most rigorous of data collection methods.
Might some of these groups have a vested interest in exaggerating the death toll to make the Other Side look bad? To drum up support for greater ‘international’ military intervention? Questions which seem to me – as someone who admittedly doesn’t know a great deal about statistical analysis – obviously worth asking.
And – who knows? – while these figures may well be broadly accurate, the key point should be that the methodology used to come up with them won’t be subjected to anywhere near as much criticism and scrutiny as the Lancet studies on Iraq were by state-corporate media, for reasons which should be obvious.