1. Background Elections are always a high stakes game, and, in many countries, particularly as it seems in many African countries, the effects of losing have uncomfortable consequences for many political parties. These effects increase for the many autocrats that have governed African countries for lengthy periods. As Posner and Young have pointed out, the shift away from undemocratic methods of gaining and maintaining political power has led many African autocrats to seek electoral strategies for keeping power, bolstered in many cases by attempts to manipulate the constitution in order to increase the number of terms that they might serve as presidents2. However, even if ruling presidents can manipulate the constitution in order to keep having the possibility of being re-elected, they still have to win elections, and, where longstanding rulers are unpopular (as is frequently the case), they may have to resort to undesirable methods in order to do so. As a recent analysis of violence in African elections points out: To summarise, the main finding is that significant electoral violence is not as frequent as critics suggest, but frequent enough to cause serious concern for advocates. That significant violence occurs in nearly one fifth of all African electoral contests means that advocates should pay attention to the risk that democratic elections can trigger. A secondary finding, which is perhaps unsurprising but still quite important, is that incumbent forces overall commit more violence than challengers do. However, conforming to expectation, where challengers engage in violence, especially significant violence, they are more likely to do so after an election occurs and, furthermore, if electoral violence occurs after the election it is likely to be at a higher level3. Of the 41 cases of violent elections examined by Straus and Taylor, a high degree of electoral violence took place before polling in 31 cases. Zimbabwe, for example, has shown this for the 2000 Parliamentary elections, the 2002 Presidential election, and the 2008 Presidential rerun. Straus and Taylor do not include the 2008 elections as their analysis takes the period between 1990 to 2007 only; however, they do include the 2005 Parliamentary elections as an example of a poll in which there was a high degree of political violence, which many Zimbabwean commentators might dispute4. Thus, our understanding about political violence becomes more nuanced. There is violence that takes place in countries that falls short, and is more short-lived than so-called “armed conflict”, but can have very profound effects on citizens and a country as a whole. Recent elections in both Kenya and Zimbabwe have been violent, although the degree of violence has varied between elections, but the violence as a whole caused huge social disruption, deaths and injuries, destruction of property, amongst many various problems. 2 3 4 For an analysis of how this affects Zimbabwe, see RAU (2012), Bucking the Trend: Africa, Zimbabwe, Demand for Democracy, and Elections. May 2012. HARARE: RESEARCH & ADVOCACY UNIT. Straus, S., & Taylor, C (2009), Democratization and Electoral Violence in Sub-Saharan Africa, 1990- 2007. Prepared for delivery at the 2009 Annual Meeting of the American Political Science Association, September 3-6, 2009. There is little dispute that the 2005 elections were seriously flawed, and that there was significant intimidation, but it is doubtful that there was much violence in the sense in which the other Zimbabwean elections since 2000 were violent. See Reeler, A.P., & Chitsike, K.C (2005), Trick or Treat? The effects of the pre-election climate on the poll in the 2005 Zimbabwe Parliamentary Elections. June 2005. PRETORIA: IDASA. 5