Sunday, September 13, 2009


Harris and Hitchens never stopped long enough to define just what they meant by God. You had to guess; most of the time it looked like Yahweh (red in tooth and claw) was meant. Dawkins is more careful--"I am talking only about supernatural gods" (p. 41), although in the next breath he claims that "the most familiar to the majority of my readers will be Yahweh, the God of the Old Testament." Really? Not the Christian God of love and redemption? And the vast majority of his readers, he admits, will be Christians. Why would they be more familiar with the God of the Judaic scriptures than the God of the New Testament?

(Here's a can of worms indeed, the varied contents of which we'll come back to again and again, DV, in these pages.)

Could Dawkins' choice have anything to do with the fact that Yahweh is a more vulnerable target? Let that pass for now. In the next chapter he clarifies what he means by a supernatural god: "a superhuman, supernatural intelligence who deliberately designed and created the universe and everything in it, including us" (p. 52). He doesn't define supernatural. You're supposed to know what it means. For most people, the following equations hold:

natural = what I can understand

supernatural = what I can't understand

Naturally (pun unintended), the advance of science has tended to move one thing after another out of the second category and into the first. Is that process one of infinite reach, so that eventually the second will be totally empty? Dawkins, along with many other scientists, would like to think so. We'll see.

Indeed Dawkins (in contrast to many others of no mean weight yet very different approaches, T.H. Huxley and the anonymous author of the Cloud of Unknowing for two) explicitly states that the existence of God is ultimately provable or disprovable by the exercise of human reason. In fact, he believes (p. 73) that we can already weigh the probability of God (and find it close to zero). He distinguishes seven points on a spectrum of probability, from the 100% of the "strong theist" to the 0% of the "strong atheist" (p. 74). However, there's a problem here if we think back to the definition of "God" on p. 52.

Dawkins is well aware of the difference between theism and deism. But his definition fails to distinguish between the kind of hands-on God who follows us as individuals, rewards believers with eternal life, answers (or at least hears) prayers, works miracles, and all that kind of stuff, and the hidden, mysterious God who created a universe that would eventually spawn intelligent life, for whatever reasons He may have in mind. Clearly the balance of probabilities is very different, depending on whether you are talking about the first kind of God or the second. Dawkins seems to be running probabilities on the first, so what he might take them to be for God 2.0 remains a mystery.

One thing he is adamant on, though, and it's this: people can't hide in the kind of wishy-washy agnosticism that says that since we can't calculate a firm probability either way, we should say the chances of God/no-God are fifty-fifty. Not so, says Dawkins, because if you propose the existence of anything that cannot be empirically proven, the burden of proof is not on skeptics to disprove it, but on you to prove it.

And, on and off through the rest of the book, he repeatedly lines God up with other things that have been (or might be) proposed on the basis of no empirical evidence: Bertrand Russell's earth-orbiting teapot, the tooth fairy, the Flying Spaghetti Monster, and more.

This is bizarre, understandable only in light of one gambit of the New Atheists--to weaken religion by trivializing it, in the same way that picadors weaken the bull so it's less able to avoid the matador's killing stroke. One potent way of trivializing is the same as one way of imputing guilt--by association. Associate God and religion with things like fairies and imaginary tea-pots and you're already half-way to slaying the beasts; at worst you've cut them down to size in people's mind. But such associations, just like "guilt by association", are ridiculous. Nothing turns on the truly imaginary things that Dawkins adduces. Whether or not there are tooth fairies, whether or not there is an imaginary teapot between here and Venus makes not a smidgen of difference to the world or how we think about the world. But whether or not there is a God makes an enormous difference.

Of course, orbiting teapots, fairies with or without teeth, and Spaghetti Monsters flying or otherwise, are all things the probability of which is so low that nobody in his right mind would bother to assess them. However, there is one thing they have in common that they don't share with God. If they existed at all, they would be physical entities in the real world. But God, if he exists, is not. God, as the anonymous president of a New Jersey historical society (quoted in my last post) along with every sensible Christian for centuries, has said: God is a spirit. (In the fourth and fifth centuries there actually were people called Anthropomorphites who not only believed that God was a bearded old man in the sky but were ready to kill those who rejected the notion, but the Church eventually squelched that one.)

Does this make any difference? Yes, as we shall see, it makes quite a lot of difference. Let's start with the question already touched on--whether it is possible, in any meaningful sense, to assess the probability or otherwise of God's existence.

Dawkins thinks it is. Quite apart from the probability scale already mentioned, he considers that with respect to intelligent life on other planets, one can "make a sober assessment of what we would need to know in order to estimate the probability" (p. 95). He adduces the Drake Equation

N = R^{\ast} \times f_p \times n_e \times f_{\ell} \times f_i \times f_c \times L \!

created in 1960 by astronomer Frank Drake, where N is the number of civilizations in our galaxy in which communication might be possible and R* is the average rate of star formation per year in our galaxy, fp is the fraction of those stars that have planets, ne is the average number of planets that can potentially support life per star that has planets, and so on. Dawkins points out that though the numbers to be assigned to each term in the equation are still subject to vast margins of error, some of them have narrowed in the last half-century. For instance, in 1960, the notion that any other star might have planets was pure conjecture, but now we have spectrographic evidence for the existence of several hundred planets, while more (and smaller, potentially more life-friendly ones) continue to be discovered. This, Dawkins claims, should slightly reduce our skepticism about the existence of extraterrestrial intelligence. In the same way, "science can make at least probabilistic inroads into the territory of agnosticism" (p. 96--don't forget that agnostics just as much as fundamentalists are in Dawkins' crosshairs--"He who is not for me is against me" is, as we have seen, a canon in the atheist creed.)

But if N were God, what would the terms of the equation be? Dawkins pours scorn on Stephen Unwin (pp. 132-6) whose book The Probability of God uses Bayesian analysis (a formula slightly different from the Drake Equation, but broadly similar in its aims and results). The scorn is not wholly unmerited. Unwin feeds in figures that are purely subjective guesses for such parameters as the existence of evil or the validity of miracles and even then only comes up with a 67% likelihood for God's existence (which, according to Dawkins, he then ups to 95% by adding his equivalent of the "cosmological constant" Einstein used to make his equation come out right--"faith"!) Obviously, Dawkins and Unwin would assign very different values to the presumed validity of miracles; Bayes' Theorem works when the input consists of objectively valid information, otherwise it's a mental placebo for the distressing fact that there are many things we don't know, and some we can't.

In the final analysis, Dawkins takes neither Drake's nor Bayes's route. He bases his judgment on a single improbability: "A designer God cannot be used to explain organized complexity [the complexity of, say, humans, DB] because any God capable of designing anything would have to be complex enough to demand the same kind of explanation in his own right" (p. 136). Well, how did humans get to be so complex? By natural selection, of course; according to Dawkins, or me if it comes to that, evolution by natural selection is the only way you can progress from simple to complex physical organisms. Dawkins is ready to accept the possibility of godlike aliens who evolved on some other planet (p. 99). They too would be physical organisms. But suppose God is NOT a physical organism? I mean, no sensible person I know of ever suggested He would be. Is there natural selection for spirits? If so, what environments select for different kinds of spirit? (Without variation there's nothing for natural selection to work on.)

Such questions are absurd, of course. Dawkins could easily circumvent them by simply saying, the universe consists entirely of physical organisms and material objects. After all, philosophical dualism is rejected by most scientists, even, nowadays, by most philosophers. So Dawkins could simply win his case by adopting materialistic monism, a healthy tradition from Democritus on. Why doesn't he?

I don't know. But in any case, he's wise not to, as I'll show in my next post. For now, just let's say that Dawkin's killer argument would kill God only if God were a physical organism produced by natural selection, which no-one has ever suggested He was. If He is indeed a spirit, then we have no way of knowing how simple or how complex He would have to be to do whatever it may turn out He does. If He's there.

It is however possible to estimate roughly how complex God would have to be if we consider different possible versions of God. If God indeed produced humans, down to their brain-wiring and internal plumbing, from a handful of dust--like a watchmaker makes a watch--then yes, I guess He'd have to be pretty complex. If on the other hand He simply set the parameters of the Universe so that eventually, something like us--something capable of comprehending the idea of God--would be bound to turn up somewhere, then there's just no way of computing how simple or how complex He would have to be. And if God worked through evolution, as many pretty-orthodox Christians believe, He would not have to be complex at all, on Dawkins' very own assumptions.

Which assumptions a? I'll tell you next time. Nothing like a cliff-hanger to get folks' attention.


  1. I'd argue that "supernatural" doesn't simply mean "things I can't understand". I'd argue it means "things that can't ever be understood".

    There are lots of things that people don't understand that are nevertheless not considered to be supernatural - many people don't understand internal combustion engines or computers, yet they don't seriously believe that they are 'supernatural'. Even if they don't understand them, even if they don't think they could ever understand them, they believe that at least some people can and do understand them.

    But 'supernatural' stuff isn't supposed to be understandable at all. It works in totally different ways and is beyond human ken. Saying something's 'supernatural' isn't saying "I don't understand this yet", it's saying "nobody's ever going to understand this."

    Neil deGrasse Tyson has an essay, "The Perimeter of Ignorance", which details the problems this notion has caused devout scientists before:

  2. BTW, on what basis do you claim that the Flying Spaghetti Monster is not a spirit? What qualifies something as a spirit, anyway? What properties mark something as a spirit?

  3. Good post by Tyson, but again, beside the point. I have never declared a belief that there are "things we cannot know"--we simply doi not yet know whether there are or are not such things--nor that we should ever stop looking and asking questions. Whether or not a human mind can potentially know everything is an empirical question. I am concerned about giving premature answers (e.g., atheism) to questions that are not as easy as they look at first sight. But again, I'll be posting on human reason shortly, I hope, in which case all will (I again hope) become clear.

  4. Let me ask a question. How can you tell the difference between "something we don't know yet" and "something we can never know"?

    In a practical sense, what does "the supernatural" - in the sense I've outlined - add to the conversation?