Those examples were deviations of reality from prediction/model. This is sort of the opposite -- the model predicts negative frequency light but we ignore this

Those examples were deviations of reality from prediction/model. This is sort of the opposite -- the model predicts negative frequency light but we ignore this because it doesn't match what we thought reality was doing. Now we see that there is some tangible real-ness to these solutions. It may have interesting applications, but it wouldn't require us to change the theory (since the theory already predicts it correctly).