In 2005, observers of the Middle East and North Africa (MENA) described a series of elections around that time as a “desert spring.” Although events such as municipal elections in Saudi Arabia, granting of women’s suffrage in Kuwait, parliamentary elections in Oman, peaceful elections in Yemen, and electoral and constitutional changes in Algeria and Morocco--were largely top-down, analysts celebrated them as democratic stirrings. Subsequent setbacks, such as Syria’s and Egypt’s sham elections in 2007 and 2010, respectively, Iran’s contested 2009 election, and strong performance of Islamist parties such as Hamas brought renewed skepticism about the region’s long-term prospects for democracy. When mass uprisings began to sweep the region in late 2010, another optimistic term, “Arab spring,” emerged. Work seeking to explain it and why social scientists and regional scholars failed to predict it is proliferating. Drawing on literatures on the role of violence in institutional change and “democratization by elections," I advance the hypothesis that the timing and nature of the current revolutions are primarily the outcome of a process that began with earlier cosmetic electoral reforms--dating even to pre-independence colonial elections--and the cycles of participation, expectations, and violence they produce. I argue that they are comparable to democratization processes that occurred in Europe and elsewhere between 1792 and 1920, particularly given the relative insulation the region has enjoyed from broader international norms and trends that some argue make European and modern democratization qualitatively different. Despite leaders’ best efforts to retain power through what Marc Lynch calls “defensive democratization,” subsequent election fraud and manipulation following phony or incomplete openings legitimizes opposition use of violence to actualize the principles that even the most superficial reforms represent. I test the hypothesis by analyzing the timing of election fraud, violence, and revolution using an original global dataset(1800-2012) and qualitative and event data analysis from elections in Algeria (1894-2012) and Egypt (2005-2012).