Wednesday, January 20, 2010

What Is the Probability of God?

When I hear arguments from theists about how improbable evolution is (often they mean how improbable abiogenesis is), I wonder if it's at all appropriate to consider God's existence in terms of probability and improbability. Besides, to anti-atheists evolution is like God in the atheist "religion," and they talk all the time about probability.

Of course, to the philosophers the God everyone believes in is a necessary being. That is, if God exists, he must exist. There's no possible world where God exists but where he didn't need to.

But I don't think this is the end of the discussion because God's necessity is hypothetical. It's a convention of thought used to make it possible to go on with the discussion. So, if we can table the necessity idea I'd like to present a syllogism that is very - if not fatally - flawed, but that can perhaps serve as a thought-starter for more and better ideas.
1. According to traditional Jewish thought, God is utterly unique.*

2. Something unique is rare, unusual, and/or distinctive.

3. Something rare, unusual, and/or distinctive is improbable.**

4. God is the single most rare, unusual and distinctive being possible.

5. Therefore, God is more improbable than any other being or occurrence.
Why is the reasoning suspect? Because to assign a probability to some event or being, there needs to be some real data that establishes base rates. Without actual data and base rates, we're just guessing and assigning the probability in terms of what subjectively seems surprising.

The more I look at the syllogism, the less I like it. (1) needs to be unpacked, and (4) gets into the realm of ontological arguments. But I am not really interested in constructing a perfectly solid formal proof so much as trying to support the idea that physical processes leading to the emergence of life on Earth should be no less surprising or believable than the existence of God.

Perhaps another approach is to ask why it should at all be believable that the universe comes from a god or "intelligent designer."

* Comment on (1): This means that there is only one God and that there is nothing like God. Whereas we like to think of individuals as being unique from one another - me, you, Barack Obama, Senator-elect Scott Brown, and so on - these individuals belong to the class "human beings," so other people are like them and belong to their kind. My understanding of the God idea is that nothing is like God and nothing is a kind to which God would belong.

** Comment on (3): I'm not sure that this premise holds up. The lack of real vales hurts because something could be infrequent but at least predictable. 


  1. How would you go about calculating the probability of God?

    Anyway, probabilities don't mean what most people think they mean.

  2. "How would you go about calculating the probability of God?"

    I wish I could say I had an idea about how to do this...but, alas.

    Probabilities depend on prior conditions, but the idea of God short-circuits this because there is no priority to be had.

    As I try to say in the main post - flailing, struggling - if there is a probabilistic argument to be made for or against the existence of God, I would think it would have the strongest relation to an ontological argument.

    What do you think?

  3. Ontological arguments aren’t really proof of God. At best, they’re proofs that, given a set of conditions, if God exists He must be X. I suppose you might be able to calculate the probability that some of the conditions are true.

    Maybe we can assign a probability of a given thing existing by seeing how many things we know exist that are similar to it. We could use reports of its existence as another measure of probability, but those should be contingent upon the mental state of the person reporting, so that, for example, the report of someone who was sober when he sighted the thing in question would carry more weight than someone who was drunk.

    So a unicorn would have a fairly high probability of existing on the first measure, as it is similar to horses and other equines; but it would have a very low probability on the second measure as there are few (no?) reports of unicorn sightings.

    God would have a very low probability on the first measure, as by definition there is nothing else like Him, but a fairly high probability on the second measure, as there are many reports of interaction with God. The probability on the second measure would have to be adjusted, though, for the reporters’ mental states during their “sightings” and for the social indoctrination that lead them to interpret an experience in a given way.

    Given all that, I would estimate the probability of God to be greater than 0, but much less than would be needed to actually worry about.

    This is all gross guesswork though, and I’m biased. It would take a lot of work to quantify any of these, and I’m not sure it could be done. It’s not like flipping a coin.

    Maybe we can think about it like this. If you hear hoof beats, and never see what was making the noise, what is the probability that it’s a unicorn?

    If you have a divine experience, what is the probability it was caused by God?

    If a coincidence occurred, what is the probability it was caused by God?

    Does God pick bridge hands?


Feel free to comment if you have something substantial and substantiated to say.