2007-06-04

Most of the time, before trusting a new method, we look for some evidence that the new method is providing accurate information. We look for some alternative method to corroborate the results of the new method being considered. If the new method doesn't agree with previous methods, it is not trusted.

There is no alternative method to corroborate observation. The only way to get corroborating information is through another observation. How can we know that both observations are not mistaken or incorrect?

Thus, there is no way to prove that observation is a valid method.

0 Comments:

Post a Comment

<< Home