delayed recertification, one has to go back to the time right after the second crash. The first question was: How could MCAS rely on one single sensor?
After a few months it became clear that the old flight computer could not handle the input of two sensors. The workaround meant combining the two flight computers for reading both aoa sensors.
While this gives some failsave for the sensors, it eliminates any backup for the flight computer, since there are only two of them and in combined use they count as one without any failsave.
So with the current hardware there just is no way of getting a certifiable MCAS. The Canadians are on the right track, because dumping MCAS and certifying the Max as a new plane is the only way to get it in the air. I know, that this takes time, training and money, but it might save lives and jobs.