If we didn't have any women in the armed services the cases of rape would probably decrease [not cease ..............]
That's probably not true.
Evidently since there would be no servicewomen, no servicewomen would be raped, that is the only conclusion. Since no differentiation is made between rapes in the military and total number of rapes in society (including the military) and since the supposition that the 'probability' of the total number of rapes decreasing, the inference is that servicewomen whilst serving are more likely to be raped than if they were not serving, which could be interpreted as a form of victim blaming, but I think it is just sloppy logic badly phrased.