[snystrom]

You're correct that there's never been an experiment on the scale of the current pandemic, but there is a long history of demonstrating that viruses rapidly evolve to escape antibodies, and that this effect is enhanced under direct selective pressure.

Late 2020 there was a paper published where they demonstrate this effect for SARS-Cov-2 in human cells, then they look for those variants in humans and find that they were already infecting humans, demonstrating this selection activity happens in the wild [1]. I totally hear you about the 1 does vaccine argument. I think the kicker is two-fold. First, we do not know whether vaccinated individuals can still be infected with and spread COVID. Second is the rollout is currently very slow. You could imagine a scenario where enough folks are vaccinated to add pressure, but not reduce widespread transmission. If the data comes out that you can't transmit covid after the first dose (and I believe that Moderna has this data, but not ready to publish), I think the answer is clear to delay the second dose. But I think it is the absence of the transmission data, not whether evolution will happen to the virus, is part of the reasoning for the current strategy of completing the vaccination course in the ~3 week timeline.

[1] https://elifesciences.org/articles/61312