In just the same way you can get away from taxes by lying vehemently… he lost his job and reputation in less than three years.
Since the paper itself was okay, but the data was falsified, obviously it was hard to prove the data was false until other studies not only showed it, but also his reputation was discredited and (presumably) investigations finished.
Incorrect data can happen even to a good paper in good faith due to instrument error.
The paper in question, again, was lots of “maybes” and no direct conclusions. The earth shattering conclusions were reached in press conferences where the guy lied vehemently, and the journalists ate it up like coke.
The goal posts were not moved at any point. It was a discussion of the situation, as it is.
Please look at the paper you refer to: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(10)60175-4/abstract It was only retracted because of “In particular, the claims in the original paper that children were “consecutively referred” and that investigations were “approved” by the local ethics committee have been proven to be false. Therefore we fully retract this paper from the published record.” It was retracted due to fraud. I don’t think it’s in any way wise to blame the possibility of fraud on the peer review process. Just as fraud can happen in any field because some people decide to pathologically lie.
However, besides the fraudulent ethics, the paper is fine, and as always previously reiterated multiple times. All it says are a bunch of maybes. It makes no extraordinary claims, it holds no conclusive proof, just a lot of “this maybe hints to something”. The paper is publishable.
The actual scandal was caused by the Wakefield lying profusely in media.
These are two different things: what Wakefield said in media, and what Wakefield said in the paper. You should separate them.