Sacco and Vanzetti
Ferdinando Nicola Sacco (April 22, 1891–August 23, 1927) and Bartolomeo Vanzetti (June 11, 1888–August 23, 1927) were anarchists who were convicted of murdering two men during a 1920 armed robbery in South Braintree, Massachusetts. After a controversial trial and a series of appeals, the two Italian immigrants were executed on August 23, 1927.
There is a highly politicised dispute over their guilt or innocence, as well as whether or not the trials were fair.
In Hepler, A. B., A. P. Dawid and V. Leucari (2007). "Object-oriented graphical representations of complex patterns of evidence." Law, Probability & Risk 6(1-4): 275-293, Hepler et al apply their BN framework to the Sacco and Vanzetti case as described in Kadane, J. B. and Schum, D. A. (1996). A Probabilistic Analysis of the Sacco and Vanzetti Evidence. We apply our idioms to the case and believe that the resulting structure is more elegant and accurate.
We focus on the same part of the case covered in Hepler et al. Figure 4 (with further breakdown in Figures 9 and 10), namely the case for Sacco being at the scene of the crime. Our model is presented in Figure 1.
Figure 1 BN for Sacco and Vanzetti case
We have used the same shading conventions as before (evidence nodes in dark shading, accuracy nodes light shading, and hypothesis nodes unshaded), but this time we have distinguished prosecution and defence hypotheses and evidence nodes by making the former bold.
The ultimate hypothesis here is “Sacco at scene”. The whole model is built only from instances of the evidence accuracy idiom. The improvements/corrections over the model in Hepler are:
- Some of Wade’s testimony is directly about the “Sacco at scene” hypothesis and some is about “Man similar to Sacco at scene” ( Hepler . wrongly assumes only the latter). Figures 9 and 10 of Hepler. are wrong structurally with respect to Wade.
- The handling of the Frantello testimony in Hepler. is structurally wrong.
- The model in Fig 4 of Hepler. did not include the crucial Pelser component
In wider discussions of the case Figure 7 of Hepler . introduces an explicit node called ‘motive’ and an ‘opportunity’ node (“At Scene”) as we already mentioned in Section 3.3. But Hepler et al treat both of these the same as any other evidence about the guilty hypothesis, i.e. the links are from guilty to motive and guilty to opportunity rather than the other way round.