Thursday, February 28, 2013

Agnosticism and Epistemic Caution

Nick Bostrom has some very interesting things to say about reasonable agnosticism here.  I agree with a lot of it, but I'd add a few points about theistic belief at critical junctures that, I think, tip the scales in the overall probability estimation (in favor of atheism.)  He's being too epistemically cautious, as I see it, and not acknowledge some of the most salient features of the epistemology, sociology, history, and psychology of religious belief.  The vast majority of human religious belief is much more adequately explained by a number of natural, not supernatural accounts.  So the fact that so many people believe, and that disagreement persists, should not pull our probability assessment so far into the agnostic zone.

Transhumanist Nick Bostrom on Agnosticism

2 comments:

Brad Lencioni said...

I think Nick Bostrom is brilliant (his website is full of good stuff). In this context, though, my guess is that if you really pressed him on the existence of religious gods, he would fall in the atheists category.

All of Bostrom's work concerns posthumanism and AI; his most popular paper is the "simulation argument", in which he concludes that it is rational to be "agnostic" about the proposition that we are living in a computer simulation.

And I think this is his primary motivation for claiming to be "agnostic" about the existence of a "godly" super-intelligence responsible for our universe. (I dont think it is that he thinks Plantinga or Craig, etc., are doing any legitimate philosophy.)

Bob H said...

Richard Carrier’s new book, Proving History, Bayes’s Theorem and the Quest for the Historical Jesus, New York. Prometheus Books. 2012 Print, makes a strong argument that there is always a non-zero probability of any hypothesis. Using Bayes analysis we can determine what the epistemic probability that a hypothesis is true. Carrier’s argument is the only methodology of inquiry should be based on the proven theorem of Bayesian method of analysis. Carrier provides us with the ‘canon of probabilities’, P(h|e.b) ≤ .0001% is virtually impossible. This method gets us away from being agnostic, we can say with certainty that the God hypothesis is virtually impossible, therefore there is a clear case for atheism.

The Impossibility of GOD edited by Martin and Monnier, provides sufficient background knowledge to set P(e|~h.b) = 1. In other words the evidence we should expect from historical revelation if the evidence is made up is 100% not true.