Atheism: Introducing the Type II error.

In my previous post I contended that atheism is a null-hypothesis. In this article I will explore what it means to not reject the null-hypothesis.

A statistical experiment will (or should) clearly state the null-hypothesis — this is the hypothesis which they will attempt to reject.  It has been said that:

“The null-hypothesis is never proved or established, but possibly disproved… Every experiment may be said to exist only in order to give the facts a chance of disproving the null-hypothesis.” – RA Fisher

A failure to reject the null-hypothesis means that there is no reason to believe in the alternative hypothesis (even though it may very well turn out to be true).  And it’s all downhill from there:

  • There is no reason to reject null-hypothesis that “God does not exist”.
  • ∴ there is no reason to claim that “God exists”.
  • ∴ I cannot logically assert that “God exists”.
  • ∴ I do not assert that “God exists”.
  • ∴ by definition, I am an atheist.

I have suggested to you many times before that atheism is a logical conclusion; I have never once claimed to have been able to prove it true, nor is it actually required of me to do so. 

Is this logical? Yes, I think so. Does this mean that the null-hypothesis is necessarily true? No.

A Type II error is one which fails to reject the null-hypothesis although it is in fact false.  In other words, failing to reject “There is no god” while there is in fact one.  The cost of this error?

Enter Pascal.

Just my thoughts so feel free to discuss/debunk it.