If one removes red while the other removes green, then if something looks dark through the lens that blocks green, and looks bright in the one that doesn't block green (but blocks red), then one could tell that it is green. (and visa versa for something red)

Not that this would give the same subjective experience of a person w/o colorblindness seeing red vs seeing green, but I wouldn't be surprised if it allows one to pass r/g colorblindness tests fairly well?