With the public debate over military action against the Bashar al-Assad regime in Syria following the use of chemical weapons, on August 29 the website usvsth3m posted an interactive online game they called “Where’s Damascus? (Don’t Ask Us).” Using Google Earth, the game calls for players to “click where you think Damascus is” on a map of the whole planet. Players are then told how close they came (in miles), and the percentile in which they fall.
The next day, the site posted a story titled “The US Department of Defense isn’t sure where Damascus is: statistical proof.” They analyzed the first 100,000 plays of the game. There were 139 from within the UK Houses of Parliament, and 59 from the Pentagon. Some 64% of the former and 57% of the latter got it right within 200 miles, compared to a total average of 52.5%. (The figure of within 200 miles was taken as a correct hit because of the scale of the map and the difficulty of inputting a precise spot using a touchscreen on a phone or tablet).
It could be argued that ignorance among the general public about Syria suggests that public disapproval of military strikes is not a compelling reason to forgo them, and that the media is not doing its job reporting on the issues.Jeffrey Mazo
The implication of the article is clear from the title, and from the opening sentence: ‘You’d expect folks in the U.S. Department of Defense to know the location of the place they’re probably about to bomb.” But it’s not just the DoD, or parliamentary staff; the overall figures are pretty discouraging. Although the figures are not broken down by country, the game asks players to click on the map to show where they are playing, the article points out that Texans perform below average (49.3%) and readers of the UK daily newspaper The Independent perform well above average (64.3%). It’s a UK website, and the whole tone of the exercise, and the article, is in keeping with the general “meme” that Americans are dangerously (or hilariously) ignorant. The meme crops up in stories about the confusion of Iraq and Iran among public and politicians, for example.
Of course there is some truth to this, as polls and other evidence shows. For example, when I helped teach a medieval history course at a top American public university back in the 1980s, we gave students a test on the first day of class: they had to mark 20 countries and cities, including Paris and London, on a blank map of Europe. The results were appalling and amusing at the same time. Survival published an article a few years ago on the state of U.S. education and its implications for intelligence analysis. But whether this phenomenon is uniquely American, or particularly strong there, is another question, and one that would require robust analysis rather than anecdotal reportage, however entertaining.
Usvsth3m is a humorous website, dedicated to publishing both original and linked “topical, internet LOLs” (i.e. ‘laugh-out-loud’ items). As I write this, the top story on the site is a collection of video mash-ups of Star Trek characters such as Captain Kirk and Mr. Spock reacting to Miley Cyrus’s notorious performance at the Video Music Awards. The site does not need to be taken seriously. But the “Where’s Damascus?” game and its reported results do reflect some real issues.
The question of possible action against Syria could not be more serious. It could be argued that ignorance among the general public about Syria suggests that public disapproval of military strikes is not a compelling reason to forgo them, and that the media is not doing its job reporting on the issues. On a deeper level, one reason for public disapproval is suspicion of the motives of the policymakers who advocate action, and the idea that those in power are uninformed, only reinforced by the game and the way it is reported, increases that suspicion.
But beyond the immediate question of Syria, the “Where’s Damascus?” game reflects some issues about evidence and methodology that apply to more serious analysis. Despite the website’s headline, the results of the game aren’t “statistical proof” of anything. For example, can we really conclude that the difference between the success rates of the public in general, the Pentagon players and those from the Houses of Parliament are meaningful? The small numbers in the latter groups suggest not, although rigorous statistical testing would need to be done to prove it either way and that is conspicuously lacking here. And the detailed results from the first 100,000 plays, displayed on zoomed-in maps in the online article, do not necessarily support the conclusions that have been drawn.
The overall results show a solid bulls-eye covering Syria, Jordan, and parts of Iraq, Turkey, Saudi Arabia, Lebanon and Israel, surrounded by a penumbra of hits dropping in density as the distance from Damascus increases. The wide area covered by the bulls-eye is of course a feature of the coarseness of the resolution with which a player can choose a location, which is why getting within 200 miles is considered a success. And the penumbra is almost circular, centered on Damascus, and appears to fall off in intensity in a normal distribution. Again, this would require statistical analysis to confirm, but it looks as if the people who located Damascus in Egypt or Greece, for example, were simply a bit sloppy in placing their cursor, or their finger. It’s the same sort of pattern we’d expect if 100,000 people who knew exactly where Damascus was were throwing darts at it on a map.
Other methodological issues include the possibility of deliberate “spoofing” of the game – picking wildly inaccurate places as a joke. A good number of people, for example, picked the towns of Damascus in Oregon and Maryland, and probably went out of their way to find places called Damascus outside of Syria. The number of other, apparently random, places in the United States that were picked also suggest people were simply clicking randomly, or there are bugs in the software. The detailed maps of the choices of DoD and Parliamentary players strengthen this impression: the Pentagon map shows the ‘bad aim’ pattern centered on Damascus, with outliers in South Africa and India that are apparently jokes. The parliamentary map shows the same dartboard pattern, with outliers scattered apparently randomly in clusters in the equatorial Atlantic and Africa, the Southern Ocean and Siberia/Mongolia/China.
The game designers are almost certainly wrong, then, when they speculate that a relatively high density of hits in Greece reflects “ignorance of European geography” or that in Libya and Egypt it means that “people seem aware there’s trouble in that part of the world and just plopped a marker down in some sort of vague trouble spot.” In both cases, the errors are probably just down to a sloppy design – or at least design that would be considered sloppy if this game was intended as a real data-gathering exercise rather than a silly game.
The fact that the bulls-eye around Damascus is not perfectly circular, and the penumbra extends slightly west of circular into western Libya and Tunisia and east into Afghanistan, does suggest some genuine, if minor, geographic confusion. But to my eye, the only real anomaly is a cluster of hits east of the Caspian Sea, in Kazakhstan, Turkmenistan and Uzbekistan, showing up on both the general and Pentagon-specific maps of choices. This cluster does seem to reflect a real misunderstanding of Damascus’ location. Even here, I’m not convinced by the explanation offered by the folks at usvsth3m: “We reckon that people vaguely remembered that Damascus is on a western shore, but ‘flubbed’ which particular sea it is. After all, it couldn’t be the friendly old Mediterranean, where we all go on our holidays.”
For the record, when I played the game, I used the zoom feature on Google Earth to get in as close as possible to the area I knew it lay in (the game does not allow zooming in as closely as Google Earth). I got within 20 miles, but this error was down to weak overhead imagery analysis skills rather than ignorance: I mistook what’s probably a dry salt lake to the southwest of the city for an urban area (both are grey blotches at the scale I was looking at). Good thing I’m not in charge of targeting decisions.
This article was first published in the IISS blog on Sept. 11, 2013.
Jeffrey Mazo is IISS Consulting Senior Fellow for Environmental Security and Science Policy, and Consulting Editor of Survival. He has worked in scholarly journals publishing since 1987, most recently before joining the IISS at Frank Cass Publishers in London. Until May 2013 worked full-time as the IISS Research Fellow for Environmental Security and Science Policy and Managing Editor of Survival. Interests include the effects of environment and climate on long-term social, cultural and political development, ethnic and identity politics, and science and technology policy.