Proofiness - Charles Seife [6]
For example, on October 16, 1995, Louis Farrakhan, leader of the Nation of Islam, held an enormous rally: the “Million Man March.” Of course, the gathering was named long before anyone knew whether a million men would actually attend the rally. When buses and trains and planes filled with men began converging on the National Mall in Washington, D.C., it was a huge event, but did it really live up to its name? Quite naturally, Farrakhan said that it did; his unofficial count of the crowd topped one million people. However, Farrakhan’s count was a Potemkin number. It was a foregone conclusion that the organizers of the Million Man March would declare that a million men were in attendance, regardless of how many actually attended—anything else would be embarrassing. The Park Service, which was in charge of giving an official estimate of the crowd size, was already feeling the pressure to inflate the numbers. “If we say [the crowd] was 250,000, we’ll be told it was a half-million,” a U.S. Park Police officer told the Washington Post shortly before the rally. “If we say it was a half-million, we’ll be told it was a million. Anything short of a million, and you can probably bet we’ll take some heat for it.” Nevertheless, the Park Service dutifully peered at aerial photos and counted heads in an attempt to size up the crowd. As predicted, when the official tally came in—400,000 people, give or take 20 percent or so—a furious Farrakhan threatened to sue.4 As a result, the Park Service stopped estimating crowd sizes for more than a decade, giving rally organizers free rein to make up Potemkin-number crowd estimates without fear of contradiction from a reliable source.
Whenever there’s a big public event, someone has a vested interest in making up a number that makes the crowd look huge. Pretty much every year, the organizers of Pasadena’s Rose Parade announce (presumably for the ears of sponsors) that they estimate a crowd of one million spectators; the real number is probably half that. Estimates of the crowd at Barack Obama’s inauguration topped five million at one point—probably between double and triple the number of people actually in attendance. The right is just as guilty as the left of fudging crowd numbers. In September 2009, a number of right-wing commentators claimed that ABC News had estimated that antiadministration “tea party” protests in D.C. had drawn more than a million people; in fact, ABC had pegged the crowd at between 60,000 and 70,000 attendees. Two months later, Republican representative Michele Bachmann bragged on the conservative Sean Hannity Show that a protest she organized had attracted between 20,000 and 45,000 angry citizens—and a video montage of swirling crowds seemed to support her assertions. However, the estimate was way off—the Washington Post estimated that the crowd was about 10,000. Worse yet, some of The Sean Hannity Show’s footage was recycled from the much larger September protest, making the crowd look much more substantial than it actually was.5
Creators of Potemkin numbers care little about whether their numbers are grounded in any sort of reality. From afar, however, they seem convincing, and a Potemkin number can be a powerful tool to prop up a sagging argument or to bludgeon a stubborn opponent. Even the flimsiest of them can do tremendous damage. Joe McCarthy’s famous claim to know of 205 communists in the State Department, for example, was transparently false, yet it made him a national figure.
Using Potemkin numbers is the most overt form of proofiness. It takes the least skill to pull off. After all, it’s incredibly easy to fabricate a number that suits whatever argument you are trying to make. (This is why Potemkin numbers are so common: 78 percent of all statistics are made up on the spot.)6 However,