It’s just an unfortunate fact pattern. Compensatory damages are by design supposed to be a quantifiable number that you can directly point to on a ledger somewhere. Mann claims to have lost grants, and indisputably lost reputation, but has remained employed at a seemingly proper level throughout his career. Scientists don’t get kickbacks from grants, so no quantifiable losses _to Mann_ there. Loss of reputation is hard to quantify, though it is sometimes done by having a PR expert testify “this is how much it would cost to restore a person’s reputation to its previous state”, but considering Mann’s previous state was “unknown to the public”, that’s also tough to do. IIRC, Mann has had to have private security at various points, but that was paid for by his employer, so Mann can’t be compensated for it. It seems like in the absence of hard numbers like that, the jury decided to compensate using punitive rather than trying to make up a number, and no party is allowed to tell the jury that that will get struck down.