Saturday, March 10, 2007

Calculating God

Welcome one, welcome all, to my 100th post! I know it's been a bit slow here, but I've been spending the time working on this one, making it as good as possible. What I wanted to do was to take on the hardest piece of woo to argue with I could find and end up tearing it to shreds.

Oh, and speaking of important numbers, it seems that there's this blogger who goes by the name of "PZ Myers," on a blog called "Pharyngula," who recently turned 50. While the squid fascination is a bit worrying (last guy I knew who acted like that ended up fervently worshipping Bel-Shamharoth out of fear that if he didn't, he'd be sucked into the Octavo), the guy does seem to be something of a decent skeptic. In honor of this event, I am officially adding Pharyngula to my blogroll. You're welcome.

Anyways, onto the woo! In my last big post, I talked about the Principle of Charity. Today I'm going to put that into effect and take on one of the best arguments in existence against my personal worldview. This would the argument from fine-tuning used by the religious to argue for the existence of a (more commonly "their") god. To put it frankly, this is the best argument I've heard from them, though it's by no means sufficient.

Constructing the Argument

First, to be as charitable as possible, I'm going to go in with my knowledge of physics to construct and refine the best fine-tuning argument I can (taking apart some of their actual arguments would be too easy, as they rely on strawmen, misinterpretations, special pleading, and sloppy logic which are all easily debunked). Let's start this by simply listing all the possible degrees of freedom in the universe; we'll get to narrowing them down later. For the sake of sanity, we'll be keeping within the regime of universes with the same fundamental forces as ours, which exhibit quantum effects, and with same number and orientation of directions (3 spatial and 1 time-like. The time-like nature actually comes straight from the spacetime metric, where a time-like dimension will have a negative measure. Why? It's complicated, and I'll do it in a later post if you guys want. Note that this also implies relativity). The reason for making this limitation is that without it, the total number of degrees of freedom will be a degree of freedom itself, possibly extending to infinity.

Anyways, the list. The number in parentheses after each entry is the number of degrees of freedom it potentially has. Yes, some of these are tied together; we'll prune later.

  • Speed of light (1)
  • Planck's constant (1)
  • Gravitational constant (1)
  • Permittivity of free space (electric constant) (1)
  • Permeability of free space (magnetic constant) (1)
  • Strength of weak nuclear force (2)
  • Strength of strong nuclear force (1)
  • Mass of the Higgs particle (1)
  • Fundamental charge (1)
  • Fine structure constant (1)
  • Curvature of space (1)
  • Composition of the energy in the universe (3, for matter, radiation, and dark energy)
  • Spacetime metric (4)
  • Independent components of the CKM matrix (4)
  • Independent components of the Maki-Nakagawa-Sakata matrix (4)
  • Particle masses (up to 12, for the 6 leptons and 6 quarks, all the known elementary particles aside from photons, which are by definition massless)
Total: 39

Oddly enough, Creationists have only been able to come up with around 26 constants. They likely performed some pruning on it themselves, but as you'll see, I'll go further. The first problem is that a lot of these overlap and actually represent the same thing. The second problem is one of scaling. This means that with certain constants representing conversions between units, you could scale the whole unit system throughout the universe and have no change in behavior. In addition to that, you can throw on one more scaling factor to everything with no harm done.

Okay, a piece-by-piece pruning of the unnecessary degrees of freedom:
  1. The speed of light, Planck's constant, and the strength of the electrostatic force all work as conversion factors, so we can set them to 1 without loss of generality.
  2. The time element of the spacetime metric is simply the speed of light squared, so we can get rid of that.
  3. Let's get rid of another spatial element of the spacetime metric with our freebee overall scaling factor.
  4. With only the other two components of the metric changed, it's just a scaling of Cartesian space (as if the universe were compressed or expanded along one axis). This wouldn't actually be noticeable to anyone within the universe, however, so these are out.
  5. The fundamental charge is part of the formula for the fine structure constant, so that's out.
  6. The magnetic constant can be determined solely from the electrostatic constant and the speed of light.
  7. The strong and weak nuclear forces have been pretty much combined (or at least, it's been shown to be likely they can be combined) into a force with electromagnetism, so those are likely dependant on the other factors.
  8. Of the elementary particles, only four actually play a role in the formation of life: The electron, electron neutrino, up quark, and down quark. The up and down quarks also have the same mass, so only 3 relevant degrees of freedom here. (Note that the electron neutrino is virtually massless and its mass plays little role, so it might seem like it should be out. But consider that if it were instead extremely massive, things might get screwed up. So, it is relevant.)
  9. Under the model of early exponential inflation, the curvature of space will be normalized to almost 1 very early on in the universe, so it's not relevant.
  10. The CKM and Maki-Nakagawa-Sakata matrices tie in with the electromagnetic force and the strong nuclear force (they describe oscillations between particles that exhibit these forces). It's likely a future theory will be able to predict them.
  11. We don't know that the Higgs particle actually exists at all, and if it does, it likely will have little effect on the formation of life.
So, here's our pruned list, using only constants that we know (or are very sure of at least) are independent and relevant. If we use constants that we don't have a good reason to believe are independent or relevant, then we're quite simply creating a bad argument, giving people one extra venue of attack. That's why we're only using the constants that we know matter.
  • Gravitational constant (1)
  • Fine structure constant (1)
  • Composition of energy in the universe (3)
  • Particle masses (3)
Total: 8

Okay, so with the best argument we can make, we have 8 relevant degrees of freedom for the universe which might need to be tuned quite precisely for life to evolve. It's not nearly as bad as the 26 Creationists usually claim, but it's sufficient that there's indeed some problem. It's at this point that the Creationist making the argument would argue that the fine-tuning necessary makes production of a universe that can support life like ours by random chance extremely unlikely. However, an omnipotent god could easily create this.

Problems with the Argument

The Scientific Model

The argument from fine-tuning rests on the assumption that the alternative model is one in which a single universe is created randomly. By relying only the choice between this model and the claim of a god, they're committing the fallacy of Bifurcation. While this is a possible model, science never claims this. In fact, modern science makes no claims as to what actually did happen. There are, however, possible models that would explain this (and yes, the existence of a god is one such model). However, none of these are (as yet) testable, so science can progress nowhere beyond the generation of these models. A few example models, aside from the aforementioned Goddidit and One Random Universe models:

The Flying Spaghetti Monster - Based on a parody religion, one model of the universe is that instead of being created by a god, it was created by a flying spaghetti monster. Yes, it seems ridiculous, but it's not falsifiable. Therefore, it's just as valid a model as the Goddidit model.

The Farting Raccoon - A parody of a strawman of evolution created by Ann Coulter. Despite those strikes against it, it does qualify as a valid creation model.

Okay, that's just joke models so far, but we can't rule them out. Most people do anyways, because they seem ridiculous. It's at this point that I'll point out that atheists see the Christian Goddidit model to be equally ridiculous.

Think about it: A Magical Sky Daddy poofs the universe into existence (twice, in different ways, if you believe the Bible). Then he creates man. Then he creates woman as company for man, and he declares sex between them - the only way to create more humans - a sin. He gives them no real intelligence or moral guidance, and then plants a tree with a big Neon sign saying "Don't Eat the Apples" and creates a snake saying "Eat the apples." His creations, which he had complete control over, then eat the apples, making them intelligent (obviously he didn't want us to be intelligent), and he punishes them. Yeah, that makes sense.

Onto more models:

The Deist Model - Some god or something (say, a godly kid for a science fair project) poofs this universe into existence, then sits back and does nothing with it. The whole noninterference part immunizes this to testing and being falsifiable.

Other Religious Models - Pretty much every religion has their own creation story, and most of them can be tweaked to be unfalsifiable, so these fit in here.

The Multiverse Model - This is the model used by many atheists. In it, there's not just one, but many (possibly infinitely many) universes, and different ones are created with different physical constants and initial conditions. This model is as yet unfalsifiable, but it does circumvent the problem of a universe like ours being unlikely - with many universes, it suddenly becomes likely or even a certainty that at least one will be like ours.

So, there we are with many alternative models. On the face of it, there's no evidentiary reason to believe in any of them over any of the others. At this point, the argument from fine-tuning has now failed to provide any reason why any of the models which involve a god is superior to the others, so there's no reason to accept it as proving anything.

There is one thing we can do, however, to prune down these models: Occam's Razor. Essentially, the more complicated and bizarre models are less likely to happen randomly at the beginning of everything, so they're less likely to be true. Almost all of these models assume something complicated preexisting, be it a Flying Spaghetti Monster, Farting Raccoon, or a god. Assuming something complicated like this at the beginning seems pretty odd.

However, the multiverse model doesn't necessarily have this problem. Sure, having an infinite number of universes is large and expansive, but it's not necessarily "complicated." In fact, you could argue for its simplicity by the fact that you can sum it up in a single word: "Everything." So, what seems to be more likely to simply exist at the beginning (or infinitely backwards if there is no beginning): An omniscient, omnipotent god capable of knowing precisely what parameters will result in human life and capable of creating a universe with precisely those parameters, or a maelstrom of possibility where everything that can happen does happen? To me, it's obviously the latter, which is why I prefer that explanation.

The objection most theists have to such a multiverse model is that the multiverse itself would have to be extraordinarily fine-tuned in order to create ton of other universes. They come up with an extensive list of mechanisms necessary for this universe production to go on. Despite the fact that these lists are generally loaded with assumptions of how this would have to work (why bother with extensive mechanisms at all, when we can just say the natural laws just let it spontaneously happen?), they don't seem to realize that all these mechanisms could just as easily be enforced on their god. If they don't have to be enforced on their god as well, then there's no good reason they should have to be forced on this model either.

Calculating God

Now, what happens once you apply the argument from fine-tuning to the hypothetical (well, let's pretend they don't have a specific god in mind) tuner? The same thing that happens with ID arguments from complexity, interestingly enough. We break down into three possibilities for the tuner:
  1. The tuner is more fine-tuned than our universe.
  2. The tuner is exactly as fine-tuned as our universe.
  3. The tuner is less fine-tuned than our universe.
If case 1 or 2 is true, then postulating a tuner solves nothing; we have just as much, if not more fine-tuning as we had before, so the same arguments would have to apply to this tuner (that is, if the argument from fine-tuning were valid). If the third case is true, then we've established that low fine-tuning can lead to higher fine-tuning, so what's the point of a fine-tuner at all? Why not postulate natural processes that lead to increasing fine-tuning?

(It's worth noting that different theologies put the complexity of their god at different levels, and there are indeed both gods more fine-tuned than us and gods less fine-tuned than us.)

If that sounds suspiciously like the same argument I've used against ID complexity arguments, that's because it essentially is. Oddly, these two Creationist arguments stem from different sources, but they both feature the problem of what happens when you apply them to their god. This is the point at which the Creationists just argue it away with special pleading: "Well, God is made of a material unknown to us, so we can't apply these arguments." However, the arguments are mathematical and independent of the structure of the universe, so whatever their god is made of, the arguments can still be applied. The only way out of it is if they postulate an illogical god (some do, but that's a whole other can of worms).

Alternate Life

There's one big assumption that underlies the fine-tuning argument: They say that the universe must be fine-tuned for life as we know it to exist. But what about some other bizarre form of life? Is there some society off in another universe with their version of IDiots claiming that the universe must be fine-tuned because without gravity being the strongest force and the fine structure constant precisely equaling 1/pi, life as they know it couldn't exist. There may be a myriad of islands of stability in the range of these constants that allow life to form. It might even be a continuum with a different version of life at almost every step along the way.

The common objection to this argument is that they're only comparing our universe to adjacent universes that differ only in small amounts from ours, and that these adjacent universes seem to be incapable of sustaining life. The problem with this argument is that we don't actually know that these universes are incapable of sustaining life.

For instance, let's take one example of how the universe would be different if we change one constant: Increasing the strong nuclear force by about 2%. Doing this, "diprotons" suddenly become stable (the residual strong nuclear force is enough to bind two protons together), and hydrogen would likely fuse into these instead of deuterium and helium. This then leads star formation off on a different path.

To this I respond, "Yes, and...?" What's to say this won't lead to some alternate form of life, which instead of using elements with nuclei made from a mixture of protons and neutrons, are made from multiple protons. There's no obvious way this will prevent life, so why assume it would? (Note: I'm using this as an example because it's the one they most often give. If someone reading this has a better example, please leave a comment.)


Although the fine-tuning argument does indeed raise some good points about the formation of the universe, and does force us to consider what led to it, where it really fails is in connecting this to the existence of any god, much less some particular god. The question is worth asking, though, is this argument why they believe in their god themselves? I doubt you'll find many who say that it is (though there are a few Deists who claim this, but no one who believes in a specific god and none of the big proponents of ID). If it isn't, why aren't they using the arguments that convinced them of God to convince us?

Unless... what convinced them of God were the biases and flaws in human reasoning (that are in fact expected results of evolution) which skeptics have recognized and work to overcome, and they're now trying to mold facts to fit their predetermined theories. It would seem a lot easier if they would just present some scientifically reproducible evidence of their god instead of conjecture based on scientific findings, but evidence seems to be beyond their means. I wonder why...


Akusai said...

This is probably the best takedown of the fine tuning argument I've ever seen. Dawkins does a fair job in The God Delusion, but his major field of study is not physics and so it felt somewhat lacking. Yours really tackles the problem head-on and exposes it as wanting. Great 100th post. Here's to 100 more!

Though I'm a bit surprised you didn't mention the Very Strong Anthropic Principle. You know, the one that states that the entire purpose of creation is ensure that there is a man in England who spends his time writing Discworld novels, and the fact that Terry Pratchett does just that proves it all true.

john said...

I agree, this is a fantastic article. It systematically takes down every valid argument for believing Deism reasonable through the fine-tuning argument. Although the anthropic principle would have applied here, it is often kind of a universal argument that is overapplied, and it is more impressive to be able to make the point without using it.

This argument concludes that there is a higher probability of no god than of God. You invited Deists to explain why they are convinced of the existence of God, so I will explain why I practice Christianity (although in a form very different from the average Christian).
Some parts of the Bible are reported with a historical nature in the same manner as wars are reported. The wars in our history books are from recorded history that we accept because we have little reason to believe the alternative that people recorded a bunch of lies just to fool future readers of their documents. Another main factor in determining the past is unity. If a mass set of historical documents support one conclusion, we accept it. In the Bible's case, some of the books are not anywhere near this standard. For example, no one knows who wrote Hebrews, bringing the question, "How can so many Christians accept that Hebrews is a divine book if we don't know who wrote it?" You mentioned Genesis, with its two creation stories, as particularly questionable. But, for example, it is generally accepted that Mark was alive at the time of Jesus, and wrote his account of Jesus' life later. Various accounts of his life existed, and scholars at the time picked out the books that they believed were most accurate, discarding others. While in many cases I do not feel they were stringent enough (anyone ever read Revelation?), Mark is an example of a book I follow the teachings of.

At this point you might say, why believe something that seems obviously more likely to be false than true? Believing in this provides eternal heaven if you are right, and being kind of stupid if you are wrong. Rejecting this provides a little satisfaction from laughing at the dumb people and not "wasting" your time on a falsehood if you are right, but you have no idea what happens if you are wrong. After looking at various belief systems, I have concluded that most other religions are dumber and their conflicts and ridiculousness make them not worth doing, as you appear to have done with all religions.
Various psychological studies have shown positive correlations between religiousness ( and good health. This is not to say that God smiting the atheists, but that our beliefs do not need to always coincide with "What is more likely?" Occasionally, our beliefs best coincide with "What will benefit me?" By attacking a problem with personal research, knocking out the Flying Spaghetti Monsters and Islams of the religious spectrum, finding something that seemed the most valid, and understanding the potential benefits, I feel I have taken an actively beneficial approach. I don't generally criticize other people's views, and when it comes to the creation of the universe I would rather believe that a Big Bang occurred then that one of the conflicting Biblical accounts with no basis are correct.
The problem comes when attackers of well-reasoned valid arguments like "Calculating God" decide to resort to arguments like "evolution is only a theory" and banking on people's inability to grab a dictionary.

"The question is worth asking, though, is this argument why they believe in their god themselves? I doubt you'll find many who say that it is (though there are a few Deists who claim this, but no one who believes in a specific god and none of the big proponents of ID). If it isn't, why aren't they using the arguments that convinced them of God to convince us?
Unless... what convinced them of God were the biases and flaws in human reasoning (that are in fact expected results of evolution) which skeptics have recognized and work to overcome, and they're now trying to mold facts to fit their predetermined theories." Another thing to add to this is, why do so many people emulate their parents in their religious affiliation? If Christianity really is so smart, then why are more people in countries with freedom of religion switching from it than to it? For many it becomes a source of identity, like the sports team you like, which also generally stays constant regardless of whether the New York Yankees winning a baseball game really is a rational thought. Still, many people accept the Bible as truth over more established things such as global warming.

An overwhelming amount of Deists regard atheism with contempt when nearly any outside observer (say, someone of a differing form of Deism) would given the two choices, say atheism is more likely to be true. Using a "What is more likely?" approach will lead people to an atheist belief system. However, rational people could differ regarding the "What will benefit me?" question. And in that sense, my presumably unique doctrine works for me, and yours works for you.