For one thing, it's very hard to detect a "flopping around in the breeze" situation using either analogue or digital means. For another, you're talking about a single failure mode - it might work in an AF447-type scenario, but it won't work in other scenarios - say, if the vane itself fails structurally, or if the cable connecting the vane to the transducer unit fails for whatever reason. This is one of the oldest conundrums in engineering - making sure that an alteration that solves one issue doesn't inadvertently create more problems (as Boeing have been experiencing with their battery issues).
Additionally, this isn't an issue of 'some boffin think[ing] "that can't happen"'. The AoA vanes were supplied with a specification that said they will not function correctly below 60kts IAS - therefore the *aero engineers* wrote the systems specifications to disregard output below that value. According to A33Zab, Boeing's engineers wrote the same specification below 50kts IAS.