Popper or Bayes?

I read a book review the other day in which the author said he wasn’t a Popperian who believed in falsification, but a Bayesian who believed in updating. In my view, this is a false dichotomy, on par with saying, “I don’t believe in math, I believe in engineering.” Just as math is the preconditional method of engineering, so Popperian falsification is the preconditional method of Bayesian updating. We can only update beliefs if we have submitted them for falsification. Those people who don’t make their beliefs falsifiable (Freudians, astrologers, etc.) always find ways to make their theory “confirmed” by any outcome, regardless of what happens. They never get to the point where they can update a belief because an unfalsifiable belief never needs updating. Updating beliefs as we find outcomes contrary to our expectations is the essence of rationality and the only weapon we humans have against confirmation (and other) bias, and the only way we can truly be Bayesian. Falsification is not opposed to updating, falsification is what makes updating possible.