Figure 5 Comparison of alleged violations, press versus cases, 2000 to 2006 45 Percentage of total 40 35 30 25 press 20 cases 15 10 5 Ar so n U nl Ass aw U a nl ul aw ful t fu arr l e de st Pr De ten In te o a t rfe pe th io n re rty t hr nc de ea e t w st ru s ith ct fre ion ed Po Ki om s d l it ic na al pp in i tim ng id at io n R ap e D Mu is pl rde ac r em At en te m To t r pt ed ture m ur de r 0 Violation It seems clear, therefore, that there is substantial agreement between the two data sets, and hence the case reports are likely to have good validity. In both data sets, there is strong agreement over the geographical distribution of cases, the frequency of the alleged perpetrators, and the frequency of the alleged violations. The only difference is in the temporal distribution, but, whilst this was a negative correlation, it can be seen that there a degree of agreement for three of the seven years. The contrast between the two sets of data regarding the temporal distribution is interesting. As will be recalled, the Daily News and the Daily News on Sunday were banned in September 2003, and this seems to be reflected in the sharp decline in reports of human rights violations in the press thereafter. The Daily News was assiduous in following the political violence that emerged after the referendum in 2000, and this unjustly earned the paper a reputation that it was supporting the MDC. The newspaper reported on a large number of cases of political violence during its short existence, and many of these were confirmed by other sources. With the newspaper’s closure in 2003, this type of reporting diminished dramatically. Thus, it can be seen in Figure 3 that there is a sharp decline in the number of reports per year from the closure of the Daily News onwards. However, this does not affect the more general findings, as stated above, Who guards the guards? Violations by the Zimbabwe Republic Police, 2000 to 2006