they will be randomly distributed throughout the year, meaning that a certain [proportion], which will head toward one-twelfth as years go by, will fall in the month before elections. Citing a few attacks that occured around election time is evidence of nothing.
Another difficulty is that the big one, 9/11, occurred nearly one year after a major U.S. election. The 1993 bombing of the World Trade Center, which bore links to al-Qaida, occurred three months after a major U.S. election. Al-Qaida's 1998 embassy bombings in Tanzania and Kenya occurred nearly two years after a presidential election and three months before a midterm congressional election in which the biggest issue was the fallout from President Clinton's affair with a White House intern. Election-cycle theorists finesse most of this by arguing that the danger period lasts through the first year of a new presidency. That's because a chief executive still learning the ropes is likelier to blunder either in defending against an attack or responding to it. Both proved particularly true of President Bush.
Might President Obama be similarly vulnerable? His secretary of state thought so during the primary campaign. "I don't think it was by accident that Al-Qaeda decided to test [Britain's] new prime minister," candidate Hillary Clinton said in January 2008, referring to an al-Qaida-linked car bombing at Glasgow airport mere days after Gordon Brown moved into 10 Downing Street. "They watch our elections as closely as we do." During the general campaign, Vice President Biden made a similar point. "Mark my words," he said. "It will not be six months before the world tests Barack Obama." Biden didn't say the test would come from Osama Bin Laden, but that's certainly possible. I give this theory the penultimate bead because if Biden is right, then we have entered a period of maximum danger.
Next: "The Time-Space Theory," in which we'll examine whether al-Qaida has been biding its time in a manner predicted by rational-choice theory.