Well I guess we just have different standards of what makes something proven. To me there is hierarchy 1. Mathematical or logical proof 2. Proof by controlled

Well I guess we just have different standards of what makes something proven. To me there is hierarchy 1. Mathematical or logical proof 2. Proof by controlled, reproducible experiment 3. Proof by predictions bourne out reliably by observations We certainly don't have the first two here, and for the third, given the time frames, we find that predictions (from computer models) from the past have been wrong compared to realised events. But I suppose if backfitting a model is proof enough for you then you already have all you need.