Josef Stalin is reputed to have said that it is not the people who vote that count, but the people who count the votes. Whether or not he actually made the statement, the statement points out the fundamental truth that real democracy is not possible without confidence in rudimentary rules of election administration: the ability of candidates to campaign and people to organize for political action, the ability to cast a vote freely, and the counting of votes in a fair and impartial manner. Though Americans throw around the rhetoric of “stolen elections” (in connection with Florida 2000 or elsewhere), we in the U.S. generally have confidence that the votes that are cast will be counted in a generally accurate way. (Only in very close elections, when recounts lead to examining the underbelly of American election administration, are the problems with election administration at the margins exposed.) It is always a small miracle to me that after such hard fought presidential elections, no one questions the peaceful transitions of power.
Of course, it is not so everywhere, and in places without longstanding democratic traditions it often takes outside election observers to confirm the fairness of an electoral process. When such outside verification is not available, there usually is very little smoking gun evidence showing lack of fairness. One way to try to ferret out problems with the fairness of election totals, as Juan Cole tries to do with yesterday’s Iranian election, is to look at patterns of voting that don’t fit the conventional thinking about candidate or party support. Walter Mebane has been looking at formal statistical methods to try to ferret out elections stolen by election officials. Ordeshook and Shakin are also making strides in this area. We’ll have to see what these political scientists make of the evidence from the Iranian election.
In the meantime, issues like this may be more likely to be settled in the streets, or through government repression, than through advancements in social science.