I'm currently reading "Causality: Models, Reasoning, and Inference" by Judea Pearl. Early on, he states that the development assumes that there are no certain entailments, no 1 or 0 probabilities -- that every assignment to variables in a causal model has nonzero probability.
I'm interested in applying some of these ideas to program analysis, where perfect entailment is the norm, and most combinations of assignments are impossible. How much of the theory of causality still applies in this scenario?
One interesting note is that he actually violates his own rule at one point when he talks about modelling a STRIPS-like planning language, and mentions in passing assigning 1 or 0 probabilities in that case.