The inexact science of exact numbers: does anyone know how many archaeological sites have been looted in Syria?

I’m still working on some of the posts that I’d planned to publish before this one, but I’m not going to hold it back any longer.

750 or 1,000 or more or fewer?

More than 750 archaeological sites have been attacked’ in Syria, according to the director of museums in regime territory, Ahmad Deeb. Presumably, those attacks are supposed to be attributed to rebels and jihadists. Curiously, last year (paraphrased by the Times), a former deputy director of Europol, the current president of the Belgian federal police council, Willy Bruggeman, relayed unspecified “international police” forces’ unpublished evidence that ‘1,000 historical sites’ had ‘been raided by Isis and other rebel [sic] factions during the war in Syria’. They are both (fairly) exact numbers. Are they both accurate ones?


Do “international police” – who are not Interpol (but who may be Europol, insofar as they have never replied to information requests that I submitted as an academic and/or a journalist in December 2014 and January and February 2015) – have better local sources than the Directorate-General of Antiquities and Museums and its associated police and security forces? Presumably, in its public relations with the outside world, the embattled regime has not underestimated the destructive and mercenary nature of its enemies.

Do “international police” have more comprehensive satellite data than the Smithsonian Institution and Penn Cultural Heritage Center-supported American Association for the Advancement of Science (AAAS), the U.S. Department of State-backed American Schools of Oriental Research (ASOR) and the United Nations Operational Satellite Applications Programme (UNOSAT)? None of those has made such a claim.


As I say, they are not expressive complaints that “hundreds” of sites have been attacked, they are (fairly) exact numbers. Logically, if these data exist, there must be a list (or, rather, two lists) of looted sites; there must be a count of how many sites have been looted by rebels and how many sites have been looted by jihadists; there must be a record of which rebels and which jihadists have looted which sites; and, indeed, there must be a list of sites that have been looted by desperate civilians.

There must, too, be a list of sites that have been looted by regime forces. It is understandable that cultural heritage workers who labour under the Assad regime could not publish such a list. It is less understandable that international law enforcement would not do so. It is important to heed the cautionary note on remote survey from the Geospatial Technologies and Human Rights Project (GTHRP) at the American Association for the Advancement of Science (AAAS), which warns of both the technological limits on analysis and the ethical limits on reporting. But, if such a list exists, it is inexplicable that no international law enforcement has published it, even a limited list of the sites that have been looted by the Islamic State.

After all, the existence of satellites and the use of satellite monitoring are hardly secrets; satellite evidence of looting in regime-held, rebel-held and jihadist-held territory has been published. The publication of a redacted list of sites that have been looted – without any evidence indicating by whom – could hardly jeopardise police/security operations, otherwise the U.S. and the U.N. would not be funding and publishing the material evidence that they have already.

Are even the contradictory claims true?

Likewise, there has still been no resolution to the problem of claimed-but-unpublished Iraqi intelligence data, which suggested that the Islamic State had made (up to or more than) thirty-six million dollars ($36m) from the trafficking of antiquities from Syria, but which was reported in two contradictory forms by the same reporter.

Police and intelligence/security may – rightly – be withholding evidence in order to protect investigations into organised criminals and paramilitaries. But these contradictory claims cannot both be true. This is why it is not very helpful for such structures to say, ‘[a]ll the data antiquities research has been questioning is real, but confidential due to ongoing operations/investigations.’


2 Responses to “The inexact science of exact numbers: does anyone know how many archaeological sites have been looted in Syria?”

  1. I’m a bit late to this, but your question is an interesting one, which prompted me to develop a crowdsourcing satellite image analysis project, at, which is explicitly looking at Google Maps provided satellite imagery to try to answer this question, or at least approach answering it. I wrote the application to be generic enough that other missions can be created (and missions in Libya and Yemen are in the works). But the initial mission involves looking at more than 2,500 sites in Lebanon, Syria, and Iraq for evidence of looting and other military-related damage like impact craters, construction of air defense sites, etc. The mission has been running for a bit over a month, with voluneers from the GISCorps providing a lot of the initial effort. Here are the preliminary results as of just now: Results Report as of June 19, 2015, 2:15 pm (Arizona) currently has 137 registered users.

    Currently, 57 users have contributed to “The Impact of Military Activity and Looting on Archaeological Sites in Lebanon, Syria, and Iraq”

    92.711% of the seedpoints in the mission have been visited at least once.

    There have been 5,960 visits to 2,404 of 2,593 seedpoints, an average of 2.479 visits per seedpoint.

    Users have made 6,255 observations on 836 seedpoints (34.775% of visited seedpoints), an average of 7.482 observations per seedpoint.


    39 users observed 1,477 instances of “Looting” on/near 437 seedpoints (18.178%).
    19 users observed 232 instances of “Air Defense” on/near 71 seedpoints (2.953%).
    34 users observed 1,750 instances of “Revetment/Berm” on/near 313 seedpoints (13.020%).
    21 users observed 363 instances of “Military Hardware” on/near 149 seedpoints (6.198%).
    24 users observed 486 instances of “Trench” on/near 177 seedpoints (7.363%).
    20 users observed 300 instances of “Bunker/Shelter” on/near 116 seedpoints (4.825%).
    19 users observed 750 instances of “Other Structure” on/near 113 seedpoints (4.700%).
    29 users observed 806 instances of “Impact Crater” on/near 286 seedpoints (11.897%).
    14 users observed 91 instances of “Other” on/near 66 seedpoints (2.745%).

    So more then 18% of the sites we’ve looked at have evidence of looting. Considering that the sample of 2,500 sites is far less than 10% of the number of known sites, you can extrapolate the numbers, and they’re not good. Also, see my article in Slate:


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: