[extropy-chat] Be[ing] or Not Be[ing]

[David Lubkin]

Harvey wrote:

>None of these are good reasons to believe in a theory.  They are reasons 
>that crackpots use to believe in untestable theories.  First, there are 
>reasons they would want the theories to occur.  Second, it is hard to 
>disprove their theories (or prove them).  Third, they reject Occam's razor 
>and invent more complicated explanations that add no value to the simpler 
>obvious explanations.
>
>An explanations does not have to be the simplest, *IF* the more 
>complicated theory works better.  But it *DOES* have to be the simplest if 
>the more complicated theory does not add any predictive or explanative 
>value. You can call this "science" all you want, but if you do not proceed 
>from observed phenomena, explain future observations better, test your 
>results, and have falsifiability/testability, it is NOT a science.

Part of the question is why do we do "science." I see two goals, that 
appear compatible with each other, but need not be:

1. Understand reality.
2. Predict effects, in order to
         (a) believe we understand reality and
         (b) materially enhance our lives.

The scientific method, including Occam's razor, is focused on #2. The more 
reliably our predictions match subsequent observations, the more confidence 
we have that we have hit on the truth of the matter.

This, however, does not mean that our theory is true, even if it has 
complete predictive value.

Perhaps the pyramids *were* constructed by space visitors, the world is a 
simulation, or your cat is actually a time-travelling graduate student 
researching its thesis. As a practical matter, we discard these theories, 
not because they are false but because they are undecidable and unnecessary.


-- David Lubkin.