Wednesday, March 25, 2009

The XKCD Fallacy




I see this objection way too often when I discuss supernaturalism. Here is the problem with this line of argumentation.

Suppose a scientist mixes one gram of salt into a glass of water. A miracle occurs and the amount of salt in the water doubles. The scientist boils the water and finds out that two grams of salt remain. What will the scientist conclude from the experiment? Do you think he will conclude that a miracle occurred, or that, perhaps, he messed up the experiment and forgot to use de-ionized water?

Since science is blind to supernatural processes, using science to argue against the supernatural is reasoning in a circle.

Also, I would define a miracle not as a violation of the laws of physics but as an agent outside our universe (not created at the big bang) exercising causal impact on something in the universe.

Sunday, March 15, 2009

The Teleological Argument


Disclaimer: There is very little that can be known with absolute certainty. Logical and mathematical proofs might be among them, but almost any proposition can be shot down with enough hyperskepticism. I could deny many propositions that virtually everyone else would agree are true; that other minds exist, that the outside world is real, that George Washington was the first US President. While there is no absolute proof that these things I might be trying to deny are true, there is good reason to believe them, because the evidence better supports them than it supports their negations.

Deductive logic operates on deriving a conclusion from premises. If each premise is more plausible than its negation, and the connection between the premises and the conclusion is valid, then the conclusion follows logically and inescapably.

Here are the Big Four arguments in natural theology and what they attempt to demonstrate:

Kalam Cosmological Argument: There cannot be an actually infinite number of past events, therefore, matter cannot be eternal. There was a first event. The first event had to be caused. The cause for the first event had to be uncaused, nonphysical and had to have the properties of volition and causal potency

Teleological Argument
: That from which reality came had to possess intelligence, volition, and causal potency

Ontological Argument: That which is greater than anything that can be conceived actually exists in all possible worlds

Moral Argument: Objective morality exists, and that from which objective morality came had to possess the property of morality

Note before reading: To assert: “Just because we don't understand how all events in our universe could have happened through natural, physical means doesn't mean that we can appeal to the supernatural” begs the question. The existence and causal impact of the supernatural on the natural world is the very issue at stake. To argue:

All effects of the physical world are produced by physical causes.
Therefore, we will eventually find a way to explain all events through physical causes alone.
Therefore, all effects of the physical world are produced by physical causes.
Is circular reasoning.

The conventional wisdom is that the argument from design was destroyed by Darwin and his biological theory of evolution and the critiques of skeptics such as David Hume. In recent years, this argument has come roaring back because of the discovery that the initial conditions in the big bang itself
had to be very finely tuned in order to permit the existence of any sort of intelligent life.

The Teleological Argument:
Premise 1: The fine tuning of the initial conditions of the universe is due
to law, chance, or design
Premise 2: It is not due to law
Premise 3: It is not due to chance
Conclusion: The fine tuning of the universe is due to design

Background for the Teleological Argument
When I use the term "fine tuning" I do not presuppose design. That
would be arguing in a circle. By fine tuning, I mean that there are two
sorts of quantity of the initial conditions of the universe:
1. Certain constants of nature
2. Certain arbitrary quantities

Fine tuning means that if one of these quantities or constants had been altered in a small way, it would be impossible for any sort of intelligent life to exist.

When the laws of nature are expressed as mathematical equations, there are certain constants that are expressed. For example: To calculate the force of gravity between two objects, multiply the gravitational constant of the universe, G, by the product of the masses (m1+m2) and divide it by the square of the distance between them (d^2)

The gravitational constant, G, exists independently of the other laws of nature. There is no need for the value of G to be 6.67300 x 10^-11 It could be any other value and the laws of nature would still hold.

The arbitrary quantities include the amount of entropy in the universe, the expansion rate of the universe or the ratio between matter and antimatter.

What scientists have been stunned to discover is that many of these quantities have to be fine-tuned to an incomprehensibly precise range of values in order for the universe to permit the existence of any sort of intelligent life, carbon-based or otherwise. In fact, some of these constants have to be finely tuned for matter itself to exist!

If the weak force had been altered in one part in 10^100, that's a 1 with 100 zeroes after it, the very elements necessary for life could not exist. To give you a conceptual idea of those kinds of odds, a 1 in 10^60 chance would be like randomly launching a dart from some empty space just outside our galaxy and hitting a target one inch in diameter...at the edge of the known universe (20 billion light years).

And that's just one constant out of about a dozen fundamental constants.

How about a quantity? If the expansion rate of the universe was increased by one billionth of one percent, no stars or planets would form. The universe would be a dark void of cosmic dust. If the expansion rate was reduced by a billionth of one percent, it would have collapsed into a singularity long before any stars had a chance to form.

The other way is in which we can imagine the possible range that these quantities and constants might take. Think of a radio dial. For you to dial in the station of life-permitting ranges, imagine a dial the size of the known universe. You would have to dial into an area of 2.5 centimeters.

What about a universe that was governed by totally different laws of nature? Those are not relevant to this argument, because we are only talking about universes that are governed by our laws of nature. To illustrate: Imagine a blank piece of paper with a little red dot on it. That red dot represents our universe. Let's change some of these constants and that would give you a new universe. If it is life-permitting, give it a red dot. If it is life-prohibiting, give it a blue dot. Do it again and again and again, until the paper is filled with dots. You will have a sea of blue with only a few pinpricks of red. Any randomly thrown dart that hits the paper will almost certainly hit a blue dot.

Imagine a fly resting on a large blank area of the wall. If a bullet is fired and hits the fly, it would be reasonable to conclude the bullet was aimed, even if outside the large blank area, the wall was covered with flies.

To conclude that the universe is balanced on a razor's edge is a gross understatement, it's more like being balanced on a single line of quarks!

Premise 1: The fine tuning of the initial conditions of the universe is due
to law, chance, or design

As far as I know, all causes fall into one of these three categories, or a combination of the three. Why is the sun round? Law (also known as necessity). Why did the lottery turn up a certain number yesterday? Chance. Why does the device on which you are reading this appear the way it does? Design.

Premise 2: It is not due to law


The idea that these quantities and constants appeared out of necessity is extraordinarily implausible, because it says that it is physically impossible for the universe to be life-prohibiting. However, it seems
plausible that the expansion rate of the universe could have been a little more or less, that the ratio of matter to antimatter could have been different, or that the entropy had been slightly different. The difference in any one of these quantities would have created a life-prohibiting universe. This objection bears an extraordinary burden of proof of why these quantities necessarily have to be as they are. Of course you could resort to pure speculation that these quantities might be fixed, and I could speculate
that the universe began five minutes ago with the appearance of age, or that I am a brain in a vat. There
needs to be some sort of evidence that every single one of these fundamental constants and quantities could not have been different, and there simply isn't any.

And even if every single one of the constants, such as the strength of gravity, electromagnetism, the strong force, the weak force, and the cosmological constant could be explained by the theory of everything, it still does not explain the quantities such as the ratio of matter to antimatter, the expansion rate of the universe, and the low entropy.

To quote PCW Davies: "Even if the laws of physics were unique, it does not follow that the physical universe itself is unique. The laws of physics must be augmented by some cosmic initial conditions. There is nothing in present ideas about laws of initial conditions remotely to suggest that their consistency with the laws of physics would imply uniqueness, far from it. It seems, then, that the physical universe does not have to be the way it is. It could have been otherwise"

It is important to understand what the term "theory of everything" means. It is not literally a theory that would explain everything, it is a catchy way of speaking about a unified theory of physics that would explain that the four fundamental forces (gravity, electromagnetism, atomic strong force, weak force) are all manifestations of a single force. Even if such a theory were discovered, it would not explain the cosmological constant, or the arbitrary quantities, such as the expansion rate of the universe. The best
candidate would come from string theory, an attempt to unify quantum physics with general relativity, reducing all matter to vibrating strings of energy. String theory itself involves fine-tuning, as string theory can only be formulated in 11 dimensions (m-theory). There is no explanation in the theory why reality would need to have exactly 11 dimensions. The fine-tuning of the four forces would be replaced with geometrical fine-tuning.

And it still does not explain the initial quantities.

Premise 3: It is not due to chance
The problem is that the odds of this happening are so incomprehensibly great that they cannot reasonably be explained this way. No reasonable person would ever adopt this alternative to explain some other improbability. No forensic investigator would look at a fingerprint in the victim's blood and
conclude "look how random chance caused a fingerprint that happens to match the fingerprint of our main suspect" It is true that fantastically improbable events happen. Therefore, it is not improbability alone that it significant. The key lies in specified improbability, that the event is improbable but
also conforms to an independently given pattern.

If you are playing cards, any deal of the cards is equally improbable, yet if someone continually dealt himself a perfect hand every time he dealt, you would know he was cheating, even though a perfect hand is just as improbable as any other hand.

What tips you off to design is when improbability conforms to an independently given pattern.

This shows the fallacy of responding to the argument with "every possible set of initial conditions and quantities is equally improbable. It's like winning the lottery. Any one person's winning the lottery is equally improbable, but somebody has to win it, so you shouldn't assume the lottery was rigged."

I'm not talking about the probability of any universe existing, or even this particular universe existing. I am talking about the probability of any life-permitting universe existing.

To use a more accurate lottery analogy, it would be like a lottery where 10^27 (one billion x one billion x one billion) black balls are mixed together with one white ball, and you are commanded to reach in and pull out a single ball. If the ball is black, you will be shot. If the ball is white, you will be allowed to live. While each ball is equally improbable, it will be overwhelmingly more probable that you will be dead. If you reach in and pull out a white ball, you would suspect that this lottery is rigged. What
if you had to pull out the white ball 5 times in a row?

When I say that an event arrives by chance, I am not saying that there is this force called "chance" that causes the event, or that the event is not determined by anything. Instead, it means that there may be two or more independent causal chains leading to the production of the event, and there isn't any relationship between the chains of events. In that sense, it means that the two independent causal chains happened to produce the same result. If you and I happened to meet by accident at the local Starbucks, and I say it was by chance. I am not saying that there is this thing called chance that determined us to meet there. I am saying that one set of events caused me to go there at such and such a time, and an unrelated set of events caused you to go there at the same time.

Anthropic Principle Objection
The Anthropic Principle states that we should not be surprised to find ourselves in such a finely tuned universe, because if the universe were not finely tuned, we would not be here to be surprised about it. Therefore, there is nothing to be explained. This appears to be Richard Dawkins' objection to this argument, and it is fallacious. The statement "we should not be surprised that we do not observe conditions in the universe that are incompatible with our existence" is a true statement. If conditions were incompatible with our existence, we probably would not be around to observe them. It does not follow from that statement that: we should not be surprised that we DO observe conditions compatible with our existence in light of the incredible improbability of those conditions existing.

Another way of saying it is "Why should these constants and conditions even exist in the first place?"

The Firing Squad Analogy
Imagine that you are visiting a banana republic country. You anger a relative of the dictator. The next day you are arrested and are sentenced to be executed. You are blindfolded and dragged before a firing squad. You hear the command: "Ready, aim, FIRE!" and then hear a series of clicks. The squad captain takes off your blindfold and you notice that of the 20 members of the firing squad, all of their guns jammed. Frightened, the dictator lets you go. When you return to America, you tell the news to a reporter who says "That's not surprising. If they killed you, you wouldn't be around to tell
me about it."

The Many Worlds Hypothesis
According to this hypothesis, our universe is just one member of a greater collection or ensemble of universes, which are real, actually existing, parallel (causally independent) universes. These are not just possible worlds, but actual universes. In order to ensure that our universe will appear, by chance, in this ensemble of universes, there has to be either:

1. An extraordinarily large number of these universes (like, Graham's number large) or
2. An actually infinite number of universes.

In either case, their constants and values must be randomly assigned, in order to make sure that life-permitting universes show up.

Here are four problems with both models:
1. Since these universes are unobservable, this hypothesis is neither less metaphysical nor more scientific than the hypothesis of a divine designer. John Polkinghorne said "People try to trick out a many worlds hypothesis in pseudo-scientific terms, but that's what it is: pseudo-science. It is a
metaphysical guess that there may be other universes with different laws and circumstances."

2. It is the ultimate violation of Occam's Razor, which states that causes should not be multiplied beyond necessity. It is simpler to posit one cosmic intelligence rather than many universes.

3. There is no explanation of how or why such a collection of universes should exist. Nobody knows where this world ensemble is supposed to have come from, why it exists, or why each universe would have random variables. Some scientists have tried to avoid the problem with an inflationary model
of the universe, like Andre Linde's chaotic inflationary model. In order to work, however, this inflationary model requires an extraordinary amount of fine-tuning itself. To quote Robert Brandenberger: "Linde's scenario does not address a crucial problem, namely, the cosmological constant problem. The field which drives inflation in Linde's scenario is expected to generate an unacceptably large cosmological constant which must be tuned to 0 by hand. This is a problem which plagues all inflationary universe models."

4. There is no evidence for the existence of a world ensemble. There isn't any independent reason to postulate it. By contrast, I believe we do have independent evidence for the existence of God. For example, the Kalam Cosmological Argument, the Moral Argument, and the Ontological Argument.

Many Worlds (large but finite) hypothesis
If there is an extremely large but finite number of universes, then what should we expect to observe? Obviously, all the life-prohibiting universes are ruled out, so that leaves us with the universes that produce intelligent life that can ask questions and observe what is going on. Other than that, the universe should be pretty ordinary. To use an example, suppose there was a party held for all the winners of the state lottery. Everyone who wins any money in the state lottery is invited to the party. Now you, being inexplicably star-struck by lottery winners, decide to walk up to a random individual and ask for an autograph. What are the odds that you will meet a big money winner? Very small indeed.

Back in the 19th century, there was a scientist in Germany named Ludwig Boltzmann, who was troubled by the problem of why we do not find the universe now existing in a state of heat death. According to the second law of thermodynamics, the universe will gradually use up all its energy until
it becomes a big hunk of inert, homogenized stuff. No stars, no chemical reactions. No activity. Why isn't the universe in this state already, if it is infinitely old? Boltzmann got around this by postulating a many worlds hypothesis. He believed that the universe as a whole does exist in a state of equilibrium, but by chance alone, there will appear fluctuations in this overall state of equilibrium which will produce pockets of disequilibrium. Boltzmann referred to these pockets as the many worlds. If our universe is just a fluctuation is this sea of equilibrium, then it is overwhelmingly more probable that we will be observing a much smaller area of disequalibrium than we do, namely, it would contain our star system and nothing else, or it might only contain a Boltzmann brain.

A parallel problem attends the many worlds hypothesis for explaining away the fine-tuning of the universe. According to the prevailing Darwinian model of evolution, intelligent life, if it will evolve at all, will do so as late as possible within the lifetime of its star. The less the amount of time available for the mechanisms of random mutation and natural selection to produce life, the more improbable life becomes. Given the complexity of the human organism, we should evolve as late as possible within the life cycle of our sun. John Barrow and Frank Tipler have listed in their book The Anthropic Cosmological Principle, ten steps (p 562-564) in the evolution of homosapiens each of which is so improbable that before it would occur, the sun would have ceased to be a main sequence yellow star and would have incinerated the Earth.

The Ten Steps

1. The development of DNA-based genetic code
2. The invention of aerobic respiration
3. The invention of glucose fermentation to pyruvic acid
4. The origin of autotropic photosynthesis
5. The origin of mitochondria
6. The formation of the centriole/kinetosome/undulipodia complex
7. The evolution of an eye precursor
8. The development of an endoskeleton
9. The development of chordates
10. The evolution of homo sapiens in the chordate lineage

Many Worlds (actually infinite) hypothesis
This is an utterly different scenario than the previous one. Infinity isn't just a large number, it is completely different. In an infinite number of universes, there are an infinite number of universes like this one. There are an infinite number of universes almost like this one, where perhaps, there is an extra oxygen molecule on Saturn. There would be an infinite number of universes that have greater differences, like a universe where you finish reading this post, walk out of this room, and get mauled by velociraptors.

Further, in an infinite multiverse, probabilities don't matter, because all possible events happen in some universe. There would be an infinite number of universes that do not contain the necessary conditions for intelligent life, but nonetheless contain it! Such observers are called "freak observers." Consider a random phenomenon, such as an evaporating black hole emitting Hawking radiation. When a black hole evaporates, there is an astronomically small but technically nonzero chance that it will emit an object such as a pair of shoes, a rock, chewing gum, or a human body, or a brain.

If all mental states are reducible to physical states, then there is a finite probability that any given brain will produce a certain mental state. One might be possessing the thought "I am a spontaneously materialized Boltzmann brain in an otherwise dead universe" while others might be experiencing illusory states "I am Spock" "I am Kaiser Sose" or "I am reading a blog on the Teleological Argument" Under such a hypothesis, it is logically consistent and rather likely that we are such freak observers.

It is also possible that there is no set of conditions which allows some orderly progression of a universe that leads to life. If this is the case, then all observers are freak observers. While an infinite multiverse must contain some freak observers, it need not contain any real observers. Perhaps there are no laws or constants at all. Perhaps there is only chaos, and out of that chaos, a brain pops out for a very brief period of time. Electricity, atoms, stars, planets, all you think you know could be an illusion.

Therein lies the problem with this line of thinking. Humans are extraordinarily capable of rational thinking, but once we abandon the first principles of metaphysics and fall for the brain in a vat theory, all rationality goes out the window. Once you follow the lure of hyperskepticism, it leads right into the trap of solipsism. Under the infinite multiverse scenario, we have to reason to expect or even think
about anything.

The Winn Brothers made a great video about this infinite universes hypothesis: http://www.youtube.com/watch?v=Wx4GZJpL8W0 For further reading on the infinite multiverse scenario, I'd recommend Anthropic Bias by Nick Bostrom

Ultimately, the many worlds hypothesis commits the inflationary fallacy of multiplying probabilistic resources without independent evidence. If there are many universes, then why are we not experiencing the absolute minimum conditions for intelligent life? If there is an infinite number of universes, then we fall into the trap of solipsism and abandon rationality entirely.

Who Designed the Designer?
This is a misunderstanding of the nature of explanation. In order to have a best explanation, you don't need an explanation of the explanation. Imagine a bunch of archaeologists are digging in the soil and find a bunch of arrowheads and tomahawks. They decide that the best explanation is that they have uncovered a village of a lost tribe. In order for that to be the best explanation, do you have to have an explanation of the tribe, who they were, how they got there? What if arrowheads were found on Mars? Same thing? What if some sort of machinery was found on the far side of the moon? If every explanation, in order to be valid, required an explanation of its own, then it would lead to an infinite regress of explanations. You'd need an explanation for the explanation for the explanation... and everything would be inexplicable.

Wouldn't the designer have to be more complex?
A common objection I see to this argument is that if the universe is so complex that it needs explanation, then how much more complex and in need of explanation is the designer?

First, this objection is making the claim that the designer is necessarily more complex than anything it designs. This isn't just a claim that designers are usually more complex than the things that they design, but that it is logically impossible for the designer to be less complex. In this case, the objector bears all the burden of proof, needing to show where the contradiction lies. On the face of it, there does not appear to be anything logically incoherent about the design being more complex than the designer.

Second, remember that the definition of God is: the greatest possible being. Such a being would be an unembodied mind, and such an entity does not seem that complex. It would have some complexity, such as memory, consciousness, and volition, but a mind would have to possess that in order to be a mind. In other words, it would possess that small amount of complexity by necessity.

Third, if divine simplicity is even possible, then there might not be any complexity to explain.

Finally, even if God is more complex than the universe, remember that there are three ways of explaining complexity: necessity, chance, and design. If God needs to be complex in order to be the greatest possible being, then His complexity would be due to necessity. In such a case, anything that would lack the necessary complexity would not be God.

Hence, it is reasonable to believe that from which reality came had to possess intelligence, volition, and causal potency.

For further reading on this subject, the Blackwell Companion to Natural Theology will be released in May of this year.
Image Hosted by ImageShack.us