Computing the Probability of God

Computing the Probability of God April 9, 2014

Probability of GodHave you heard of the Drake equation? It’s a simple product of seven values, and it attempts to compute the number of civilizations in our galaxy with whom radio communication might be possible.

Now that we have found clear evidence of planets around other stars, the equation is slightly more practical than when it was first proposed over a half-century ago, but it still demands reliable figures for factors we can now only guess at: the fraction of planets in the average solar system that could potentially support life, the fraction of those that produce that life, that continue on to develop intelligent life, whose intelligent life develops technology, and so on.

How likely God?

We have a similar problem when we evaluate the claims of Christianity.

Physicist Stephen Unwin wrote The Probability of God (2004) and, yes, he proposes to compute the likelihood that God exists. He uses Bayes theorem (I wrote an introduction to Bayes theorem here). You can take his equation below as a given, or you can see how it is derived from a more conventional form of Bayes theorem in the appendix. You’ll soon see that the interesting part isn’t the math but the assumptions that Unwin makes.

Probability of God

We start with a beginning probability of God’s existence, Pbefore. Use a scaling factor D—Unwin’s “divine indicator,” which is a measure of the likelihood of God given certain evidence—we compute Pafter. Unwin uses values of D from 10 (given a particular bit of evidence, God is much more likely to exist than not) to 0.1 (given this evidence, God is much less likely to exist).

Once he has a new probability Pafter, he uses that value as his new Pbefore and repeats the computation with another value of D, reflecting the likelihood of God given another piece of evidence. The computation is quite simple. The unreliable part, as with the Drake equation, is determining the probabilities.

Unwinding Unwin

We need an initial probability—the likelihood of God given no evidence. Unwin uses Pbefore = 0.5 and calls this “maximum ignorance.”

His first bit of evidence is evidence for goodness. For this, he uses D = 10 (God is much likelier given that human goodness exists). Plug in the numbers, and the equation gives Pafter = 0.91. The equation simply provides a way to merge these different factors into a single probability for God. Here are all six factors with their associated D values:

  • Goodness, such as altruism (D = 10)
  • Existence of moral evil—that is, evil done by humans (D = 0.5)
  • Existence of natural evil such as natural disasters (D = 0.1)
  • Minor miracles such as answered prayers (D = 2)
  • Major miracles that break the rules—a dead person brought back to life, for example (D = 1)
  • Evidence of religious experience such as feelings of awe (D = 2)

And after all that, the probability of God is 0.67. God is likelier to exist than not.

It’s math! How you gonna disagree with that?

(Let me note that I haven’t read Unwin’s book but instead have relied on helpful critiques by Vic Stenger and Hemant Mehta.)

I take exception to Unwin’s assumptions. First, let’s revisit our starting probability about God. Does Zeus exist? Thor? Osiris? Shiva? Quetzalcoatl? If the answer is “Are you serious? Of course not!” then why do we start with a 0.5 probability for Yahweh, especially when he looks like just another Canaanite god?

If Unwin wants to dismiss this information at the starting gate, I can accept that. But then let’s add it in as a new factor:

  • Humans have a passion for inventing supernatural gods. Believers make contradictory claims, so most are obviously false. Yahweh looks like just one more. (D = .001)

Next, let’s reevaluate Unwin’s six factors.

  • Goodness: Altruism exists in humans. This isn’t surprising since we’re social animals. Evolution has selected us with an innate sense of the Golden Rule. The Christian view also explains good traits in humans, so this gives no preference either way. (D = 1)
  • Moral evil: Humans do terrible things sometimes, and the natural explanation has no trouble with this. But Man made in God’s image with an innate sense of God’s existence? The popular free will defense fails. No, this Christian claim maps poorly to the unpleasant reality. (D = 0.01)
  • Natural evil: Indiscriminate killers like natural disasters, disease, and other calamities—things that an omnipotent God could eliminate—are hard for Christianity to explain. Birth defects and other gratuitous evil compound the problem. (D = 0.0001)
  • Miracles: The Bible says, “Ask and ye shall receive,” but prayers aren’t answered the way the Bible promises, not even the selfless ones. Coincidences abound, but we have little besides wishful thinking to imagine that they are the work of God. (D = 0.001)
  • Rule-breaking miracles: Jesus promised, “Whoever believes in me will do the works I have been doing, and they will do even greater things than these,” but science knows of zero amputated limbs that have grown back or dead people supernaturally returned to life. Surely there have been millions of earnest prayers for these, but they have been unanswered. (D = 0.0001)
  • Religious experience: We feel awe in response to both natural realities and supernatural claims. (D = 1)

The probability is now down to 10–16, but we’re just getting started. There are lots more uncomfortable facts about Christianity.

Piling on: more factors to consider

The probability of God is now basically zero (10–36 if you’re keeping score at home).

The apologist might demand equal time for the Transcendental Argument, the Design Argument, the Moral Argument, and so on. I don’t think they get out of the gate (click on the links for more).

The underlying problem with Unwin’s argument is that different people will weigh the factors differently. I’m sure you thought that at least some of my numbers above were off. I do see a benefit to the discussion, however, as it can separate the factors that support one’s belief in God’s existence (or nonexistence).

The subtitle of Unwin’s book is A Simple Calculation That Proves the Ultimate Truth. Yes, it’s a simple calculation, but no, it doesn’t prove God. In fact, the evidence points in the opposite direction.

All Westboro [Baptist Church] was
was evangelical Christianity minus polite behavior.
— Frank Shaeffer interview on Point of Inquiry

Photo credit: Andy Melton



Bayes theorem is easy to understand using a probability tree. See my introductory post for a discussion of that. It’s less easy to understand (for me, anyway) through equations.

Here’s the derivation of the equation used by Unwin, starting with Bayes theorem. We’re computing P(G | E), the probability (P) of God existing (G) given (|) the evidence (E). Bayes theorem says:

Probability of God

where P(~G) is the probability of God not existing. Define D as follows:

Probability of God

D is Unwin’s “divine indicator,” the scaling factor that represents how likely the evidence E would be if God existed rather than God not existing. Now multiply top and bottom by 1/P(E | ~G):

Probability of God

Since P(~G) = 1 – P(G),

Probability of God

Or, using the terminology of Unwin:

Probability of God

"And cisterns full of the remains of dead children who had been in Catholic orphanages. ..."

Stalin Was a Mass Murderer (And ..."
"1) I chose to be on the spot and think for myself. I learn more ..."

Christians: Why You Need an Atheist ..."
""Some scholars have suggested the causal relationship as poor social conditions as the incubator for ..."

Stalin Was a Mass Murderer (And ..."
"Stalin and Mao were Communists first and atheists second, and only because Lenin and Marx ..."

Stalin Was a Mass Murderer (And ..."

Browse Our Archives

Follow Us!

What Are Your Thoughts?leave a comment
  • JohnH2

    Problem with these types of things is that there is no reference for the numbers. They are made up from thin air, meaning the result tells us a lot more about the starting assumptions of the person making up the numbers than they do about whether or not God does or does not exist. GIGO.

    • Which I suppose raises the question, Is this an inherent problem (making this a fool’s errand) or is it possible to assign objectively grounded values to these factors?

      • hector_jones

        It would be possible to assign objective values if there were measurements or mathematical theorems to support them. There isn’t one single mathematical or measurable characteristic known to apply to god.

        I suppose that if Unwin had read all the literature on god, the problem of evil, etc he might have been able to assign somewhat objective numbers, but I got no sense when I first heard about this book that he did that. Unwin just did an ass-pull, i.e. applied christian assumptions, nothing more. Who knows how many times he massaged his input numbers to get a result that he thought was reasonable?

        • JohnH2

          applied christian assumptions, nothing more. Who knows how many times he massaged his input numbers to get a result that he thought was reasonable?

          Applying assumptions for an evil demiurge would probably be more fun.

          Existence of Good = 0.1

          Existence of Evil = 10

          Existence of Natural Evil = 10

          Minor Miracles = 2

          Major Miracles = 1

          Evidence of Religious Experience = 2

          With his same starting probability of 0.5, I get that the probability of their being an Evil Demiurge is >.99. Taking most religious records as being accurate in terms of divine retribution, hell, and historic miraculous disasters the Major Miracles number should probably be much higher, however continuity with Unwin’s numbers was desired. Therefore we can have near certainty that an evil demiurge controls our lives, and only the slightest likelihood that there is a good deity to rescue us.

          Basically, per his own methodology there is a greater probability for the existence of Cthulhu than for Jesus.

        • We see good stuff and bad stuff in the world, which makes a good god and a bad god equally likely. But whoever’s running the show here doesn’t show himself. Now, does that point to a loving god who’s desperate to have a personal relationship with us? Or to a sadistic SOB who wants to give us intermittent reinforcement just to mess with us??

        • JohnH2

          Assuming that a good god and a bad god is equally likely is probably a more rational route to go.

          However that isn’t what Unwin did, He set as 10 the evidence of the existence of Goodness giving evidence for God and as 0.5 that humans commit evil and as 0.1 that there is natural evil being evidence against God. I did bump up the existence of human evil, but even just flipping his numbers on good and evil still gives greater probability for an evil deity, all else being equal, which as I point out it really shouldn’t be given any number of religious texts.

        • hector_jones

          Are you familiar with Stephen Law’s Evil God Challenge? He basically argues that arguments from good and evil serve equally to prove an Evil God or a Good God. Yet Christians instantly reject the idea that God is Evil while insisting that their arguments from good and evil prove a Good God. It demonstrates quite well that God’s Goodness is just wishful thinking.

        • JohnH2

          I just read his paper.

          The Cathar’s actually believed in two principles, and their proof by absurdity is used by the Calvinists as a proof for double predestination (though supposedly per the Calvinists God is still good, mostly by devoiding the word “good” of any actual meaning).

          Translating such arguments to my own faith usually makes them nonsensical, as beliefs about God are very different, (also there is the very strong belief that everything has it’s opposite, so like the Cathar’s, but in a nearly completely different philosophical framework).

        • His evil god challenge is here.

          “The challenge is to explain why the hypothesis that there exists an omnipotent, omniscient and all-good god should be considered significantly more reasonable than the hypothesis that there exists an omnipotent, omniscient and all-evil god. Theists typically dismiss the evil god hypothesis out of hand because of the problem of good – there is surely too much good in the world for it to be the creation of such a being. But then why doesn’t the problem of evil provide equally good grounds for dismissing belief in a good god?”

        • hector_jones

          Yep. I want to make it clear that Law doesn’t think these arguments ‘prove’ an evil God or any God at all, just that the best arguments for a ‘good god’ are equally effective at arguing for an ‘evil god’.

          Nor does he claim that his ‘evil God’ hypothesis disproves god. His claim is that the hypothesis shows that the best ‘good god’ arguments fail to prove anything about the goodness of god.

        • Doomedd

          While there is not any evidence of god, the universe is incompatible with an all good and omnipotent god.

          An evil or indifferent god is just unproven and unnecessary, a good and powerful god is impossible.

        • JohnH2

          You are making an assumption based on the very data point which is in dispute.

        • I’m convinced, and yet believers still believe …

        • wtfwjtd

          I’ve never understood the Christian’s definition of “good” as applied to God. For people, if you are able to prevent something bad from happening to your neighbor, and don’t do it, you wouldn’t be called “good”, in fact, you’d probably be considered quite a bad person. But, God apparently does this all the time, but and yet he’s praised and worshiped and called “good” for it. So how can they call their god “good” in a different sense than the word is used for humans, and then pretend it has meaning when they plug this into a formula?

        • avalpert

          It’s pretty simple really. Anything God is is good, therefore anything that is good is God – if there is something you think is good that is not God you are wrong, or your definition of good is bad.

          Taking this one step further, God doesn’t exist, therefore good doesn’t exist, therefore all there is is evil. I’m pretty sure that solves the whole problem of evil thing.

        • wtfwjtd

          So in other words the word “good” as applied to God has a different meaning than when it’s applied to people. Christianity moving the goal posts again? Now why doesn’t that surprise me?

        • But of course those definitions aren’t in the dictionary. They should properly say that God is unjudgable. You can’t say that he’s bad when he drowns everyone, but you also can’t say that he’s good when he [fill in something good that God did here–nothing comes to mind].

        • Pofarmer

          I think the last Catholic Mass I attended was “god is infinitely good” as the sermon. I thought ” what in the world am I sitting here listening to this nonsense for?” I’m sure most everybody else was eating it up.

        • wtfwjtd

          That just emphasizes the cognitive dissonance many of us have employed over the years and not really realizing it. Once you’ve woke up, it really seems ludicrous. I’m sure several years ago I’d have been eating it up too; now I just cringe and sadly shake my head.

        • hector_jones

          Well done. This is the mathematical version of Stephen Law’s Evil God Challenge.

        • busterggi

          Squid & octopi are real hence made in Cthulhu’s image. No evidence of a god that looks human.

      • JohnH2

        Take what the universe would or would not look like with or without God; What reference do we have? We have our universe and it looks like it does, but whether or not God exist in it is precisely the question we are trying to answer; so any probability or argument that uses anything from our universe being or not being a certain way is inherently tautological, but our universe is the only thing we have so we don’t have anything in the way of hard data. So most “data” points and their probabilities become nonsense just from that.

        • The biggie for me is God getting off his butt to make himself known to us. He’s a smart guy; he could do it if he wanted.

          I guess he just doesn’t much care about a “personal relationship.”

        • hector_jones

          He’s just playing hard to get. Studies show that both men and women find that attractive.

  • RichardSRussell

    The problem with the Drake Equation is that it’s trying to draw a probability line thru the 1 datum point we know of for sure: that life does exist on Earth. Of course, without at least a 2nd point as an anchor, you can draw lines going off in every conceivable direction.

    Unwin’s problem is worse than that. He’s trying to draw a line thru 0 reference points.

    • Nicely put.

    • hector_jones

      Well we know a bit more than just ‘life exists on earth’. We know a lot about the nature of life on earth, what conditions on earth sustain it, etc. So we can make somewhat objective arguments about things like gravity, temperature, or presence of liquid water, atmosphere, etc. that we know are needed to sustain life on earth and apply them to Drake’s Equation. Knowing about life on even one other planet would only increase the probability of life on even more planets, not decrease it.

      But with God you are 100% correct. Zero data.

      • Though on the other hand, we keep discovering how much we didn’t know about life. Consider the undersea hydrothermal vents, the first place we found life that didn’t depend on energy from the sun, at least indirectly. And then extremophiles–we keep finding new ones of those.

        So I’d say that we know about less than one environment that hosts life.

        • hector_jones

          Oh sure, we definitely don’t know enough. It would help a lot, for instance, if we had some workable theory on the abiogenesis of life on earth, but we don’t. So though we might be able to calculate more and more precisely what the probability of other earth-like planets is in the universe, we can’t calculate what the probability of any one of them ever or never developing any life at all is. Is it 1 given enough time? How much time? We don’t know. It’s a pretty small data point as Richard says.

          And without a theory of abiogenesis we also are pretty much at a total loss to calculate the deviation from earth conditions that planets could have and still produce life.

        • Maybe it’s a Law of Large Numbers thing. We can say with confidence what a volume of gas will do or the rate at which a block of radioactive material will decay. The individual elements are trickier.

          Just thinkin’ out loud here …

  • RichardSRussell

    Back in 2005, for a panel at WisCon (world’s leading feminist science-fiction convention), I put together an 8-page semi-scholarly paper entitled “How Rare Is Earth?”, which goes into some depth about Fermi’s Paradox (an antecedent of the Drake Equation). I’d be happy to send a copy to anyone who requests one directly from me.

  • Pupienus

    Oh look, yet another misapplication of Bayes’ theorem. The theorem is applicable only when the events are the result of random processes.

    • hector_jones

      Maybe god is the result of a random process. I could live with that, if I had any reason to believe god existed. Christians aren’t going to like that idea, however. Then again I don’t see any of them running around claiming that Stephen Unwin has proven that God is more likely than not.

      Does Unwin think God is the result of a random process? Enquiring minds want to know.

    • Msironen

      That’s actually exactly wrong. Probabilities are statements about knowledge of things, not things themselves.

      • hector_jones

        I’m not convinced that Bayes theorem only applies to random processes, but even if it did, that’s hardly the biggest flaw in Unwin’s book.

        • Pupienus

          It _may_ be applicable but there’s no way to provide a rigorous mathematical proof. It makes intuitive sense but without proof I am uncomfortable with it. The probability that you have cancer after a positive result from a certain test is a fine application of the theorem (and one which an appalling number of physicians, even oncologists, are unable to compute). It’s a perfect application because the percentage of false positives and false negatives is well known. That’s the prior knowledge. The results are randomly distributed. The probability of you having cancer is determinable because that outcome is the result of the same random processes.

          I don’t rule it out as useful in other applications but I am very uncomfortable with claiming any result thereby derived constitutes _proof_ of that result.

        • hector_jones

          I would agree with this. Well said.

          Perhaps we need to distinguish between ‘Bayes’ theorem’ and ‘Bayesian reasoning’, the former being mathematical and the latter being what historians and textual critics do, and people who argue for god in this fashion.

      • Pupienus

        No. The probability of heads in a coin toss is exactly 0.5. “Heads” is the event or outcome. We can assign the probability because we know that the probability distribution is constant. Coin tossing is a random process. The probability of drawing any particular numbered ball from from a bag of ten uniquely numbered balls is 0.1. It doesn’t matter what numbers are on the balls as long as they are unique and that the drawing is blind so that it is a random process. The probability of you waiting more than N seconds in the bank line is computable from the number of tellers (server processes), the length of a transaction, and the arrival rate, all random processes with a particular probability distribution.

        Probabilities of the type in Bayes’ theorem are the likelihood of a particular outcome (event) based on the probability distribution of a random process. The proof of Bayes’ theorem relies on the outcome or event being the result of a random process. Take that away and there’s no proof.

        Source: someone who has a couple of math degrees,

        • avalpert

          I must not be understanding what you mean by random process. Dr. Lawerence Stone used Bayes theorem to locate Air France flight 447 even though its crash was not itself the result of a random process – at least not by what I understand the definition of random to be.

        • Pupienus

          See my response just infra. And also hector_jones’ very good distinction of Bayes’ theorem versus Bayesian reasoning.

        • Msironen

          Take the classic Monty Hall problem. There’s no randomness as to which door has the main prize and which which two doors have the crappy prize; in fact the game show host knows perfectly well which door holds what. The probabilistic reasoning involves only the knowledge of the contestant.
          Probabilities are intrinsically linked to information and knowledge.

  • MNb

    “It’s math! How are you going to disagree with that?”
    This way:

    “I take exception to Unwin’s assumptions.”
    Exactly. 1 + 1 = 10, you know.

  • kraut2

    “then why do we start with a 0.5 probability for Yahweh”

    I would put the probability for Yahweh at 1/ngods.

    • Msironen

      That’d get you probability of 100% that at least one of them exists.

      • kraut2

        That would only happen if your initial assumption is there has to be one and you sum the results of all individual gods.
        If the question is of the probability of any god existing and using D=goodness in the world as a supporting factor then the probability of any god existing is 1/5000 (let that be all gods known, probably higher) = 0.0002 and plug that in your result of any god existing based on goodness alone (the highest supporting factor) of 0.00199 – pretty low probability in my books of ANY god.

  • zeta

    Hi Bob,
    You said that: “(Let me note that I haven’t read Unwin’s book but instead have relied on helpful critiques by Vic Stenger and Hemant Mehta.)”

    I think that is far more than sufficient to draw the conclusions that you reached in your article, especially if one notices that the subtitle of the book is “A Simple Calculation That Proves the Ultimate Truth“. Anyone with some basic understanding of and experience in mathematical modeling and applicable math would immediately know that the claim in the subtitle is just garbage when dealing with probabilities or parameters that are nothing more than opinions and biases.

    The surprising thing is that, from Amazon’s webpage for this book, there were some prominent people (e.g. Michael Sherman, Paul Davies) who praised it; Sherman’s praise is especially surprising if it has not been taken out of context.

  • Jason Wexler


    The 0.5 initial prior probability makes sense if we assume no evidence either way, either god exists or doesn’t and Zeus, Osiris, Lugh and Quetzalcoatl get the same 0.5 initial probability. I once engaged in a similar exercise by attempting to correct the mathematical errors and inaccurate assumptions that Pascal made in his famous wager, I came up with something slightly more like the Drake equation a product of dependent probabilities where each probability was dependent upon the value of the previous one (moving to the left). So the first question asks god or no god, the second assumes an answer of yes and asks theism or deism, the third question assumes theism (deism suggests that worship doesn’t matter and Pascal was interested in convincing people to worship) and asks transcendental or pantheistic etc… using the same no evidence rule down the line gives a very low probability for god at least a god that anyone would recognize or “want”. Replacing each term with Bayes formula gives a more accurate response but the question of evidence comes back in the problem becomes what are the values for P (E|G) and P (E|~G) how do we provide numerical values for those arguments. I would surmise and I think most atheists would agree that high values for P (E|G) requires assuming an evil god. What is the likelihood that god would be readily and unmistakably visible if god existed? Well he either is or isn’t so 50% or 0.5 however same question but modified for god wants people to believe in god or what is the probability that god who wants to be believed in and worshiped would be readily and unmistakably visible if god existed… well I would think that probability is actually quite high and thus doesn’t conform to reality. I’ve previously written elsewhere that the only options for god based upon the evidence we have is atheism, deism or maltheism, precisely because of this Bayesian reality.

    • My initial reaction to his 0.5% starting probability was incredulity, but after a moment’s reflection, I realized that he was assuming absolutely no prior knowledge. Fair enough, but we have to work in the fact that we do indeed know that humans make up stuff, hence my first factor.

  • KarlUdy

    I don’t put much stock in these probability calculations because the determination of the actual probabilities are inevitably subjective and simplistic.

    However, I must take issue with the factor you added.

    Humans have a passion for inventing supernatural gods. Believers make contradictory claims, so most are obviously false. Yahweh looks like just one more. (D = .001)

    This is like taking the existence of counterfeit money as very strong evidence against the existence of real money.

    • I know that real money exists beyond any reasonable doubt. God? Not so much.

      No, your analogy doesn’t work.

      • hector_jones

        I don’t even know for sure that counterfeit money exists. I’ve heard stories, but I’ve never seen a counterfeit bill. I see real money all the time, although not as often as I would like.

      • KarlUdy

        How does one discern between the real and the fake?

        You argue the existence of the fake as evidence against the real, to the point of making the real 1000 times more unlikely.

        Surely you should at least to be consistent be 1000 times less sure that real money exists.

        • Why should I be consistent across two ideas that have no relation to each other?

          We know that humans make up religions. Most religions must be false since they are mutually exclusive. We have mountains of evidence for the counterfeit and are unsure if any are real. Indeed, the idea that none are real is quite plausible and would nicely explain what we see in society.

          We know that real money exists. I suppose counterfeit money does indeed prove the existence of real money–without real money, there would be no confidence for the counterfeit to ride on.

          Yet again I see no use for this irrelevant comparison.

        • KarlUdy

          Indeed, the idea that none are real is quite plausible and would nicely explain what we see in society.

          Let’s say I give this one to you. And yet the idea that one is real and all the others are counterfeit is also plausible and also nicely explains what we see in society.

          So how can you take evidence that on its own fits two alternative plausible explanations and say it is evidence for one over the other?

        • avalpert

          “And yet the idea that one is real and all the others are counterfeit is also plausible and also nicely explains what we see in society.”

          No it isn’t really plausible – at least not the idea that one of the closely related but mutually exclusive versions is real and all the others aren’t. It requires you to accept a God who want to make himself known but is so impotent that all he could do was confuse humanity as to his identity. That isn’t plausible at all.

          It also doesn’t fit what we see in society well – and certainly no where near as well as the hypothesis that they are all false. What we see in society is, from the moment of the creation of one of these ideas even the initial group cannot agree on what it is they created. That isn’t well explained by one being real and everything else coutnerfeit – it is better explained by each arising out of thin air lending itself to immediate disagreements on the ‘true’ version the same way you find in fan fiction.

        • Let’s say I give this one to you. And yet the idea that one is real and all the others are counterfeit is also plausible and also nicely explains what we see in society.

          One religion could be real, but no, this would be a poorer explanation of what we see in society. If we agree that thousands are fake, you’re going to have a tough time showing how your particular one religion actually differs so markedly in the evidence for it that it is actually true.

        • hector_jones

          How do you discern between the real and the fake when it comes to your god?

          It’s not the existence of the fake that disproves what you call the real. It’s the preponderance of fakes, all of which share numerous attributes with the one that you claim is real, along with the complete absence of even one proven genuine god, that lead one to conclude that yours is just another fake.

          If you really want to stick with this money analogy, you are basically arguing that sure there is a lot of money out there, but it’s all fake except for one bill in your wallet. To know that that is the only real bill would require you to have a standard of what is real to compare it with, which you don’t have when it comes to your god. So the analogy fails. To make matters worse, you won’t even let me see this bill in your wallet you just expect me to trust you that it’s there and it’s real.

          But seriously this counterfeit money analogy is going nowhere. Try something else, like say evidence that your god is real instead of bad analogies. In my experience, analogies may be fun but they convince no one of anything.

        • KarlUdy

          If you really want to stick with this money analogy, you are basically arguing that sure there is a lot of money out there, but it’s all fake except for one bill in your wallet.

          There is one true mint. However many other mints there are out there, the money from them is fake.

        • hector_jones

          Not even that is true. Every country has a mint. Some have more than one. Each country also prints its own bank notes, which aren’t produced in mints. Mints only make coins. So your analogy just keeps failing.

        • Yes, we agree that money exists. No, we don’t agree that any god exists. Sorry–this fails.

        • busterggi

          Funny, the government thinks it has several mints – which do you think aren’t real?

        • To make matters worse, you won’t even let me see
          this bill in your wallet you just expect me to trust you that it’s there and
          it’s real.

          And Karl himself hasn’t even seen it. He senses that his wallet must have that bill in it. I mean, an empty wallet would feel different, right?

        • The Man With The Name Too Long

          The analogy is not very good because counterfeit money is still a real thing. It’s just not the same thing as “real” (non-counterfeit) money. The difference is in the value, not the state of existence, so the analogy fails. If the analogy were accurate, then all those “false” gods would have to exist along with the real one(s), but the false ones just wouldn’t have the same value (maybe they don’t answer prayers or whatever).

        • busterggi

          So now let’s see you prove which gods are fake and which are real – and let’s have solid material evidence not touchy-feely theophilosophy.

      • TheNuszAbides

        the only remarkable part (probably unintended, even by thoughtful ol’ Karl) is the resonance between counterfeit money and [counterfeit] religion, in that: party A gets party B to believe in object X for just long enough that in some (usually unfortunate) sense it ends up not mattering that X has no “ultimately true” value.

    • The Man With The Name Too Long

      I think this also applies to the argument from design. If the universe is so complex that it requires a designer then there must be some kind of Complexity Scale being used in order to reach that conclusion. It would look sort of like this:


      But there’s the problem of arbitrariness. How can we determine where the universe ranks here? How do we know how to calculate the universe’s complexity? If there is a reliable way to do it like a thermometer is for measuring heat then I’d like to know.

      • busterggi

        It also leaves the question of who designed the designer unanswered.

    • Jason Wexler

      Your point here is correct or close to correct, but you picked a poor analogy to demonstrate it which is why you are having difficulty convincing Bob and the others of it. I can’t really come up with a better analogy, but I would make the claim more along the lines that Bob’s choice of numerical value “D” for this piece of evidence is subjective and I am sure he would admit he made it up. The problem with the money analogy is that we have reasonably good evidence that money exists and little or no evidence that god does. There are between 35,000 and 40,000 denominations, groups, cults and religions categorized by demographers right now, although I would argue there are only 127 in which the mutually exclusive claims are big enough to warrant inclusion and the rest are simple variations there in which should give us a D value somewhere between 0.000025 and 0.0079

      Actually after looking over the appendix where D is defined it seems I am mistaken, I calculated P (G|E) above or possibly P (G|E)/P (~G|E) since I would assume P (~G|E) = 1 at least for this evidence/argument. Most of the arguments that Unwin makes aren’t sensitive to the existence of nonexistence of god and so would all have D values of 1 because the probability that those attributes would occur is equal with or without god and so 0.5/0.5 gives us 1. The exception is the miracles and super miracles argument, those would have some sensitivity to god and again it would become greater the more assumptions one makes about god. On the other hand all of rest of Bob’s arguments boil down in the end to the invisibility of god which gets a D value of .5 unless we make assumptions about the nature or character of god. P (E|~G) would be 1 if we are asking about god being invisible, and since D is calculated with that value as a denominator D for god being invisible would be the same as P (E|G) in the absence of assumptions about gods character or attributes we have to assume that god would either be visible or invisible so P (E|G) = D = 0.5, it is the assumptions we make about god which would change that value. What is the probability of a god who wants to be worshiped and have a personal relationship with man being invisible, we would do well to assume (since we can’t measure it directly) that the probability is less and so D is smaller but the actual value is just a subjective guess; conversely if we ask what is the probability of a god who wants to be inscrutable being invisible, we ought to assume that D is greater than 0.5 and in fact probably very close to 1. All in all the problem is probably that Bob isn’t reporting values for “D” but rather assumed values for P (G|E) which is what were supposed to be calculating, of course I suspect that’s also true of the D values given by Professor Unwin as well.

      EDIT: I just noticed that the formula that Unwin came up with runs into problems when P (E|~G) = 0, which in Bayes formula means P(G|E) = 1. Using the Unwin formula D is undefined or infinite when P (E|~G) = 0 which for certain arguments such as the super miracles is by definition necessary. In other words super miracles are events that can not and will not happen under any circumstance without god.

      • TheNuszAbides

        There are between 35,000 and 40,000 denominations, groups, cults and religions categorized by demographers right now, although I would argue there are only 127 in which the mutually exclusive claims are big enough to warrant inclusion and the rest are simple variations

        do you have a list or was 127 a [relatively] arbitrary # for the sake of example?

        • Jason Wexler

 is a growing
          collection of over 43,870 adherent statistics and religious geography
          citations: references to published membership/adherent statistics and
          congregation statistics for over 4,200 religions,
          churches, denominations, religious bodies, faith groups, tribes,
          cultures, movements, ultimate concerns, etc. The religions of the world
          are enumerated here.

          my bad, I just noticed you wanted my list of 127 I pulled this list together from the above sources and I will copy paste it from my document, which seems to be on further inspection only 119:

          African Methodist Episcopalian
          African Pentecostal
          African Traditional Religion
          American Buddhism
          Armenian Apostolic
          Assemblies of God
          Baha’i World Faith
          Children of God
          Chinese Traditional
          Christ Holiness
          Christian Science
          Church of All Worlds
          Church of Christ
          Church of God in Christ
          Church of The Nazarenes
          Community of Christ
          Evangelical Christian
          Falun Dafa
          Falun Gong
          General Neo-Paganism
          Goddess Worship
          Greek Orthodox
          Hare Krishna
          Jehovah’s Witnesses
          Latter Day Saints
          Mahayana Buddhism
          Native American
          New Apostolic
          Non-theistic Cultural Anglicanism
          Non-theistic Cultural Baptist
          Non-theistic Cultural Catholicism
          Non-theistic Cultural Judaism
          Non-theistic Cultural Lutheran
          Non-theistic Cultural Methodist
          Non-theistic Cultural Presbyterian
          Non-theistic Irreligion
          None of the Above
          Reconstructed Paleopaganism
          Reform Neo-Hindu
          Reformed Churches
          Religious Anglicanism
          Religious Baptist
          Religious Catholicism
          Religious Lutheran
          Religious Methodist
          Religious Presbyterian
          Russian Orthodox
          Salvation Army
          Serbian Orthodox
          Syrian Orthodox Church
          The Creativity Movement
          Theistic Cultural Anglicanism
          Theistic Cultural Baptist
          Theistic Cultural Catholicism
          Theistic Cultural Judaism
          Theistic Cultural Lutheran
          Theistic Cultural Methodist
          Theistic Cultural Presbyterian
          Theistic Irreligion
          Theravada Buddhism
          Ukrainian Orthodox
          Unification Church
          Zen Buddhism

        • TheNuszAbides


        • Jason Wexler

          For what it’s worth, it seems your initial question was right, the 127 number was a ball park figure, the list I sent you was culled from a larger list I keep which can be as small as 72 and as much as 295, depending on whether or not I am treating divisions within major denominations as significant (such as between Liberal and Southern Baptists), distinguising Greek and Roman religions, Slavic and Baltic religion (Aesiti), Enlightenment era Pagan reconstructed religion from 20th century reconstructed Pagan religions, and if I am applying the concept I came up with for distinguishing practicing Jews and Catholics from Cultural Jews and Catholics, plus new-age heretics (Non-Theistic Cultural, Religious and Theistic Cultural) to all of the major denominations. So when I came up with 127 two years ago, I am certain I had culled my large list to that number.

        • TheNuszAbides

          i figured it couldn’t be a necessarily rigidly fixed number without at least a few fixed premises filtering the ‘field’ … i just asked because i didn’t know how much work was put into the ‘condensing’. i very much appreciate the effort, because i tend to find one-liners about tens of thousands of mutually-exclusive sects to be overstating the case; while that scale can be edifying as far as the absurdity of crediting any single community/movement among humans as being The True Path towards–well, anything–i’ve occasionally wondered what would be a more ‘realistic’, dare i say ‘relative’ number of divisions, considering the assumption that some of those thousands only seem to change what exact words get bandied about once a week in congregation, not so much the behavior of said congregation during the rest of life. of course even that is an oversimplification, as i’m by no means well-versed in the sectarianism of [>1]theisms.

        • Jason Wexler

          It has occurred to me that the less interested or devout one is towards a particular denominational group, the less seriously you take it’s disagreements. Consider how absurd to many of us in the atheist community, the conflict of transubstantiation and consubstantiation seems when looking at Protestant versus Catholic; or the inability of many protestants to describe what makes their brand of Protestantism different from others; or the apparent absurdity of the initial divide between Sunni and Shia as to who was Muhammad’s successor; and perhaps most poignantly consider the afore mentioned ancient pagan religions, consider how often people today even modern neo-pagans, treat Dodekatheism and Di Selecti as the same religion, particularly given neither would have been recognized at the time they were practiced as a single all consuming religion. Athena worship was not considered the same as Zeus worship by people to whom it really mattered (polytheism is mostly a 16th century misnomer for ancient religious practices, henotheism is probably better),and to make it even more complicated devotees of Zeus Olympia did not view themselves as practitioners of the same faith as followers of Zeus Delos (again sort of like the difference between Liberal Baptist and Southern Baptist). So maybe to some extent the larger numbers have a degree of validity, or at least I think so until I remember the Pentacostalist I knew in college, who was certain that her congregation were the worlds only true Christians, and that Catholics and assorted Protestants, openly denied the existence and divinity of Jesus, which makes it hard to take the differences seriously.

    • busterggi

      No because you can test counterfeit money to determine its reality.

      Now let’s see Yahweh pass a test to show his reality.

    • RichardSRussell

      Gresham’s Law holds that a cheap currency will drive out a dear one. You are postulating a society in which the cheap (fake) gods have completely driven out the dear (real) ones, so that only the fakes are left in circulation. I admire this analogy in that it recognizes that both the production and distribution of the god meme is entirely a human activity.

  • HematitePersuasion

    I loved your opening sentence. I was fortunate enough to take Professor Drake’s class on SETI … very much one of the most interesting classes I had the good fortune to take at UCSC.

  • Brian Westley

    And after all that, the probability of God is 0.67. God is likelier to exist than not.

    It checks out:

  • SparklingMoon

    Physicist Stephen Unwin wrote The Probability of God (2004) and, yes, he proposes to compute the likelihood that God exists.
    As far as the belief in God Almighty is concerned,the approach of philosophers is very different from that of Prophets. The main principle followed by the Prophets is that faith proves fruitful only if the unseen is accepted as unseen, and the self-evident testimony of physical senses and absolute mathematical proof is not insisted upon inasmuch as all spiritual merit and worthiness of nearness to the Divine depends upon righteousness,and he alone possesses true righteousness who safeguards himself against the extremes of investigation, multiple denials, and testing every little detail, and is prepared to accept a way that appears safer and preferable to other ways as the truth, out of a sense of precaution. This is faith, and this is what helps open the door of Divine grace and becomes the means of acquiring good fortune here and in the hereafter.

    When a person establishes himself firmly on faith and then seeks to foster his knowledge through prayer, worship, reflection and observation, God Almighty Himself becomes his Guardian, and, taking him by the hand, leads him from the stage of faith to that of ‘Ain-ul-Yaqin [certainty by sight.]to Certainty by sight. But all this is achieved only through steadfastness, striving, effort and purification of the ego. He who seeks clarification of all details at the very first stage, and is not prepared to abandon his false doctrines and evil ways before such clarification, prevents himself from treading upon the path of righteousness and achieving merit. Faith demands belief in certain matters which are still unseen to some degree, that is to say, they are still in a condition which is not established fully by reason, nor has it been perceived through spiritual vision, but is accepted on the basis of probability.

    This is the true philosophy of the Prophets by following which, millions of God’s creatures have procured heavenly blessings and countless people have arrived at the stage of perfect understanding and many more continue to do so. The perfect certainties which the philosophers sought to achieve hastily and daringly, and failed to achieve, have not only all been achieved with the utmost ease by the faithful ones, but they have reached the stage of that perfect understanding which has not been heard or seen or conceived by any philosopher. (Ruhani Khaza’in, vol. 2, p83)

  • Skepticali InSanDiego

    Aren’t probabilities always between 0 and 1?
    Why is a value greater than 1 allowed in this scheme?
    Something stinks here.

    • That was my initial reaction, but if you look at the definition of D, it’s a scaling factor, not a probability.

  • Charles Boot

    Unwin is just as guilty of scientism as Dawkins, Hitchens, & Harris. They insist on their perspectives being universal and are willing to use science’s & logic’s names in vain to try and prove that they’re right.
    Einstein, Darwin, Jesus, & Buddha were so much humbler.

    • I don’t know that Jesus was all that humble. He did claim to be God, after all.

  • igor

    An easier way to do the calculation might be just to multiply the D numbers. The number you get for the original is 2. That becomes 2/3 or 0.67 when used as a ratio.

    • That might be easier. I felt constrained to do it just like the original author did it. I wanted to use his own algorithm to undercut his argument.

      • igor

        Further to my comments, the very simple way to do this is this:

        Pafter = 2n / n + 1 where n = the Initial Probability.
        So if n=0.5, Pafter=2/3, if n=0.1, Pafter=2/11

        So it depends greatly on the Initial Probability. My problem with this idea of Initial Probability is that any specification of probability requires a (justified) basis or information. The original argument provides neither, but if the argument is to proceed in the specified manner, a value must be provided, or the argument fails to start.

        • The initial probability is an issue, but I thought that I addressed that sufficiently in the post.

  • igor

    I forgot to say this as well – the first five of the original factors could be deleted and the result is the same. So one must ask – on what basis did Stephen Unwin select his six factors?

  • igor

    If we number the six observations we have E1, E2, E3 etc to identify the six evidences or observations. We also have D1, D2, D3 etc to identify the six divine indicatiors.

    In general terms we have D = P(E|G) / P(E|~G)

    but I will replace G with H hence D = P(E|H) / P(E|~H) H represents the value in brackets of Pbefore.

    So for the first and second observations:

    D1 = P(E1|H) / P(E1|~H) where H=G hence D1=(P(E1|G) / P(E1|~G)

    D2 = (PE2|H) / P(E2|~H) where H=G|E1 hence D2 = P(E2|[G|E1]) / P(E2|~[G|E1])

    This is achieved by substitution and consistent meaning of variables – something the original seems to lack – ie the meaning of G (as used in the original) seems to change, and that is against the rules.

    This sort of analysis is beyond my mathematical ability so it needs to be validated by somebody who has the ability/knowledge, as I might be on the wrong track here. What I have shown makes no sense to me, but ot follows. The result may be that the basis for the original argument is bodgy…

    • The use of D came from the source I was critiquing, so I was stuck with that.

      Are you saying that your summary is more correct or just more clear?

  • igor

    deleted duplicate

  • igor

    I have no issue with the use of the divine indicator D. My issue here is a new issue – nothing to do with my summary of a week or so ago. My issue here is that in the original argument the meaning of G changes when it shouldn’t. This can be seen with a closer look at E1 and E2. The same issue applies to E3 – E6, but E2 will illustrate the problem.

    In your Appendix, you show five equations – call them Eq1 – Eq5.

    For Evidence E1 (goodness) working backwards:
    In Eq5, Pbefore = P(G) from the Initial Probability in which P(G) = 0.5.
    All references to G in Eq1 – Eq5 mean “the existence of God”. There are no problems here.

    For Evidence E2 (moral evil) working backwards:
    In Eq5, Pbefore = P(G|E1), because for Evidence E1, Pafter is P(G|E1).
    In Eq4, it follows that P(G) = P(G|E1).

    In Eq2, G means “the existence of God”.
    In Eq4, G means “the existence of God given E1”.

    So for E2 the meaning of P(G) changes through the five equations Eq1 – Eq5. This makes the argument invalid.

    Sorry if I am unclear, this is a bit difficult to explain just in writing. It might be difficult to understand what I am saying here, so it might be worthwhile working through my points systematically and slowly. (sorry about the double posting – Disqus had issues with my copy/paste. Disqus is giving me grief… Maybe because I come from the land down under)

  • matt

    Probability of God = 0

    Yes, I used zero. Get over it!

    • MNb

      Not before you have provided a solid calculation of that probability.

  • Dave

    Excellent. A maths equation that tells us what any reasonable, mature, educated person already figured out. What we need to do now is announce this to the world for the benefit of mankind. Otherwise it would be like discovering a cure for cancer and keeping it to ourselves.
    Or we could just keep quiet and allow the religiously inspired mental and physical violence to continue.
    Puzzling to say the least.

  • Rabidmob

    You’ve seemingly piled on reducing factors of your choosing which purposely unhinges the equation. I feel that if you’re really attempting a subjective point of view, then the most reasonable thing would be to take as my, “for,” variables as, “against,” which should be the largest value options.

    Then it also appears that you’re using the anti-Christian god stand point by adding, “Christianity is just a tradition and can’t be deduced from objective facts about nature. (D = 0.1).” Additionally, under what measure do these factors get their value? That seems arbitrary.

    Alas I stumbled across this trying to find a different proof. This mathematical equation seems like foppery, probability on the cosmic scale doesn’t work that way…

    The house divided against itself shall fall.

    • Yes, I add subjective factors. If you want to argue for different numbers or additional factors, go ahead. And that’s partly my point: the entire thing is subjective.

      You don’t like probabilistic arguments either for or against God? How do you argue for God then?

      • Rabidmob

        A probabilistic argument can be fine, but how can you find non-arbitrary variables and values? Also theoretically there would be an infinite number of variables. We could perhaps take an average of the 5 biggest pros and 5 biggest cons, but how would we event know which?

        The proof I was looking for when I stumbled across your blog was the ontological argument for god, which I’ve borrowed from somewhat in my philosophy.

        How does one argue for or against god? First you must define what god, or a god is. Unfortunately those definitions once again become fairly arbitrary.

        In that case, we get to the idea I was working on when I stopped here. God would then be the, “most powerful, most good,” being to exist, which is to say when I consider this concept I am meaning an being who’s, “Goodness,” and, “powerfulness,” are perhaps made up of infinite factors all moving towards or otherwise being infinite in the positive direction.

        The idea itself I was working on could be said as, “In order for us to compare any quality/quantity with another, it must exist on a 2 dimensional plane. That is to say, that you cannot compare a single point. As such we really can’t have good without evil, and so forth.”

        Well chew on that… It could be foppery itself.

        • how can you find non-arbitrary variables and values?

          I do the best I can. If you don’t like my numbers, have a go yourself. I’m certainly not claiming that I have the final word on the subject.

          The proof I was looking for when I stumbled across your blog was the ontological argument for god, which I’ve borrowed from somewhat in my philosophy.

          I’ve written about this argument here.

        • Rabidmob

          I’ve now skimmed your article, I agree with some, I disagree with other parts.

  • SugarCoated


  • Kenny Newbry

    I haven’t read Unwin’s book either. But your first assertion is that his 50/50 supposition is optimistic because of belief in multiple gods. Now again, reaffirming that neither of us has read the book, are you certain that he is speaking exclusively of the existence of the Christian god, or god in the general sense of there being a creator? That’s an important distinction, for either you or Unwin. As you and others have pointed out, the list of relevant factors could be debated indefinitely. However, there is one factor that should definitely be included. That is, “we’re here.” What does that do to the odds? The likelihood of life is ridiculously low when you really look at it. It’s actually incomprehensibly low. And in this case I only mean life in the sense of self-replication, that’s not even factoring in self-awareness and intelligence, which essentially defies physics completely (at a chemical level, yes, you can explain life, but at an atomic level it falls apart again with as much as we know today). When you think of all the things that needed to align for life to exist, you end up having to lean in the direction that it’s far more likely that there was a creator of some sort – making no statement whatsoever about the nature of that creator – than that we are purely a cosmic accident. It’s basically the infinite monkey theory. Infinite monkeys, infinite typewrites, infinite time, do we eventually get Shakespeare? The answer isn’t “eventually.” The answer is “no.” It’s possible it could happen. But it won’t. Life is much the same. I don’t agree with Unwin’s chosen factors either. But factor in the existence of life at all, and the pendulum has to swing in favor of “god.” Again however that says nothing whatsoever about the nature or motives of that God, only that statistically speaking it’s far more likely that life had a boost than that it did not.

    • kraut2

      When you think of all the things that needed to align for life to exist,

      Explain “life” as a phenomenon in simple terms.
      Is a virus “life”? At what point do you consider an organism to exhibit “life”?

      What kind of a creator do you hypothesize? One it took about 3 billion years to find its way from simple Cyanobacteria to multicellular eukaryotes?

    • The likelihood of life is ridiculously low when you really look at it.

      You imagine an understanding of life, where it can exist, how widespread it is, the way(s) it can develop, and so on that I think is beyond what science claims. Keep in mind that extremophiles keep turning up in bizarre places that continue to amaze biologists.

      Infinite monkeys, infinite typewrites, infinite time, do we eventually get Shakespeare? The answer isn’t “eventually.” The answer is “no.” It’s possible it could happen. But it won’t.

      With 10^22 stars, or thereabouts, plus billions of years, that’s a lot of monkeys. If there’s zero chance of life given this, you need to write this up as a paper, since many biologists disagree.

      Again however that says nothing whatsoever about the nature or motives of that God, only that statistically speaking it’s far more likely that life had a boost than that it did not.

      Statistically speaking? Show your work. If your result is as strong as your confidence, you may have much fame in your future.