That's a much stronger claim, stronger than the one made in Kennedy's criticism. If you or anyone can prove that they actually lied or fabricated data then yeah

That's a much stronger claim, stronger than the one made in Kennedy's criticism. If you or anyone can prove that they actually lied or fabricated data then yeah, nuke the paper from orbit. But if the criticism is that they're wrong—and that is the criticism here, then you tear it apart in the open. And if you want people to believe the opposite conclusion rather than simply ignoring the results then you need some data proving your new conclusion.