Prove that God exists

[GameOver22]

WereKitten said:

GameOver22 said:

It seems like everybody ignored your post, but I think you raise a number of good points. The one thing I always found ironic about the scientific method is that it actually utilizes an invalid argument form (affirming the consequent). The way the scientific method is structured, it says,

1. If hypothesis A is true, B will be observed.

2. B is observed.

3. Therefore, hypothesis A is true. 

This is an invalid argument form. As a counterexample:

1. If there is fire, then there is oxygen.

2. There is oxygen

3. Therefore, there is fire.

Obviously, this is false because the presence of oxygen does not mean there is fire. In the same way, the first argument is false because the presence of B does not mean hypothesis A is true. This is why there is such an emphasis on repeatability in science, and scientists are always quick to point out they don't prove theories. The best they can do is repeatably confirm them, and they argue that repeated confirmation allows theories to become knowledge.

Overall, I agree with you that the scientific method is the best system we have for gaining knowledge about the empirical world, but it is still a flawed system.

Sorry, but that's simply not how the scientific method works. Your A-> B example would be a logical fallacy and no scientist would ever commit that, but the point is that the method works by temptative falsification, not by absolute confirmation.

In your terms:

1 If hypothesis A is true, B will be observed.

2 B is observed.

therefore

3 A is currently compatible with observations, ie currently not falsified

if NOT(B) were positively observed (maybe at a past point in time with better experimental apparata) then strict logic would imply NOT(B)-> NOT(A) and A would be a falsified hypothesis

Obviously, you can make up a lot of theories that are hardly falsifiable, but they are simply not good scientific theories: a good theory should both 1) have explainative/predictive power 2) be falsifiable.

String theory seems to have a lot of (1), because it seems to explain mathematically a lot of what emerges in several other more mundane physics theory. And yet it has, currently, basically zero in the way of experimental falsification. That is why most theoretical physics don't consider it a "good" theory notwithstanding all the explaining and semplification it seems to bring.

On the other hand special and general relativity or quantum mechanics not only did explain a lot, but have been tested against, literally millions of times in thousands of different experiments over the last century.

A scientist will never tell you that they are "true" in an absolute sense, but will rely on the fact that they survived falsification in those thousands of experiments and thus that they are likely to correctly predict the behaviour of the world in real cases similar to those tested. If one day a single experiment will - say - falsify general relativity, then scientists will look for a theory that a) predicts world behaviour at least as well as relativity in the previous N cases b) is compatible with that last (N+1)th result.

The method provides a ladder of increasingly accurately predictive theories, not logic absolutes based on faulty induction. Google for Popper.

Good response. Ideally, if science operated as Popper outlined, I would agree with you. However, science often operates more along the lines of verification rather than falsification. I'm not saying that falsification is not used because it is, but verification is used much more, especially in regards to  how experiments are designed as well as conducted.