Modern creationism makes no sense

Post Reply New Topic

Men and women have the same number of ribs.

So?

Suddenly, I'm hungry for some Baby Back Ribs.

So how can it be literally true that a rib was taken from Adam? I've heard some creationists say that men have one less rib than women.

Abraham was commanded to circumcise the males in his household, as were the Jewish people later on for each generation. If the Bible were promoting the notion that removing an anatomical unit of a parent removes it from the offspring, then the command regarding circumcision would not have been given. However, since the command of circumcision was given for each generation, the Bible, thus, does not teach the false notion of Lamarckism. It does not matter what "some creationists claim" if what they say contradicts both natural and historical revelation.

Correct, and totally irrelevant.


Let's calculate. A Bayesian approach lets us compare hypotheses instead of trying to falsify each in isolation. The comparison shows why propositional logic is not up to the job of hypothesis testing where empirical data are concerned.

I will slant my assumption heavily against evolution, just to make the creationists happy.

The scenario is that you have been home schooled without any knowledge of biology or creationism. You do know your probability theory. You want to know whether the accurate prediction made by evolutionary theory favours that theory over possible competitors. Because you have no prior knowledge, you start of with these prior probabilities:
p(E) = p(evolutionary theory is true) = 1/3
p(C) = p(creationism is true) = 1/3
p(U) = p(as yet undiscovered theory, unrelated to either evolution or creationism is true) = 1/3

We want the posterior probabilities of each theory given the data: p(E|D), p(C|D) and p(U|D). We need the probabilities of the data given that a theory is either true or not true: p(D|E) and p(D|not E), p(D|C) and p(D|not C), p(D|U) and p(D|not U). These conditional probabilities tell us how strong the predictions are. The larger p(Data|theory is true) the stronger the prediction of that theory. But p(Data|some alternative is true) makes a big difference to the posterior probabilities, and that depends on what the alternatives are and how strong their predictions are.

If evolutionary theory is true and if you have enough data, the only reasons for not finding the predicted correlations would be measurement error or a very improbable coincidence. Therefore p(D|E) should be quite close to 1. I set it it 0.999. p(D|not E) depends on what else could generate the same data. Pure chance should have a very low probability. Depending on how much data you collect, you can get it arbitrarily close to 0. The probability that an as yet undiscovered theory that has nothing at all to do with evolution makes the same prediction as evolution should also be extremely low, but I promised to stack the deck against evolution, so I set this to 0.5, including in there the possibility that the result predicted by evolution happened just by chance. That is ridiculously high, but will make things more difficult for evolution.

What is p(D|C)? It should be very low. An omnipotent, omniscient and honest creator should not mislead by just happening to make species look as if they evolved. Re-use of design also doesn't explain things, because if the best designs are used, we can't account for all the Unintelligent Design and lack of re-use that can be found in nature. But someone who doesn't know anything about creationism might set the probability as high as 0.5.

Now we can calculate:
p(E|D) = p(D|E)*p(E)/(p(D|E)*p(E)+p(D|not E)*p(not E))
substitute the alternatives for not E:
p(E|D) = p(D|E)*p(E)/(p(D|E)*p(E)+p(D|C)*p(C)+p(D|U)*p(U))
p(E|D) = 0.999/3 /( 0.999/3 + 0.5/3 + 0.5/3) = 0,49975

Do the same for p(C|D) and p(U|D). The numbers are the same, so I need to calculate only once.
p(C|D) = p(D|C)*p(C)/(p(D|C)*p(C)+p(D|not C)*p(notC))
p(C|D) = p(D|C)*p(C)/(p(D|C)*p(C)+p(D|E)*p(E) +p(D|U)*p(U))
p(C|D) = 0.5/3 /( 0.5/3 + 0.999/3 + 0.5/3) = 0.250125 = p(C|U)

The prior probabilities I used were the same. The posterior probability of evolution, given that its prediction came true, is nearly twice as high as for either creationism or the as yet undiscovered theory. The Bayesian calculation takes into account the possibility that there might be other causes than evolution that could make the prediction come true, avoiding the fallacy iamnotaparakeet accused me of. The Bayesian calculation quantifies the possibilities and finds quantitative support for the theory that makes the strongest predictions. Creationism makes no strong prediction in this case. The re-use of design is an ad-hoc addition intended to rescue creationism from being falsified. That makes creationism wishy-washy (using AngelRho's words), as expressed in the middling value for p(d|C). That weaker prediction weakens the theory in comparison to a theory that makes a stronger prediction, a theory that has to give a higher conditional probability to the data.

We can then use the posterior probabilities from this calculation as the prior probabilities for the next calculation. For example during Darwin's day a physicist objected that there couldn't have been enough time for evolution because all known sources for the sun's energy (at the time that was chemical reactions and the afterglow from the potential energy of the infalling matter being converted into heat) could only bring the sun to its present temperature if it was no older than about 300000 years. The obvious answer is that evolutionary theory must predict an alternative source of energy. When nuclear fusion was discovered, the physicist withdrew his objection.

p(alternative source for sun's energy|evolution is true) must be high. Set it to 0.999 again to account for measurement errors.

Young Earth Creationism had no need to predict that other energy source because 300000 years is plenty. If we are generous, we can set p(nuclear fusion|creationism is true) = 0.5. Do the same for the unspecified theory. Repeat the calculations and you get:
p(E|alternative source for sun's energy) = 0.666222
p(C|alternative source for sun's energy) = 0.166889
p(U|alternative source for sun's energy) = 0.166889

After two strong prediction from evolutionary theory against two wishy-washy predictions from creationism, the posterior probability of evolution is nearly 4 times as high as that of creationism. That's after stacking the deck against evolution by making the predictions from creationism and some as yet undiscovered theory a lot stronger than I think they should be.

I could repeat the calculations using other cases of strong predictions from evolutionary theory, or more generally from orthodox science, compared to wishy-washy predictions from creationism.

Empirical data can constrain the probabilities of events only to values smaller than 1 and larger than 0. Propositional logic is not an adequate tool for hypothesis testing where empirical data are concerned. Applying some basic probability theories shows that the theory that makes the stronger predictions is favoured over more wishy-washy theories when the predictions come true. Theories that make stronger predictions also would have lower posterior probabilities if the data contradict the prediction. Theories making stronger predictions are more falsifiable, but falsifiability is a quantitative phenomenon. I leave the calculations as an exercise for the reader.

@iamnotaparakeet: Do you see why propositional logic is irrelevant to what I wrote and why you were wrong to think I had committed a fallacy?

@AngelRho: Do you see why ad-hoc assumptions, leading to wishy-washy predictions, weaken a theory? Do you see why a "stuff happens" theory is just irrelevant?

Creationism, based on a literal interpretation of Genesis is not only wrong and false, it is absurd. The probability of its truth is 0.

ruveyn

The probability can only become 0 if creationism is internally inconsistent. If your reason for objecting to creationism is based on conflict with empirical data, the probability of creationism being true can only approach 0, but if you have a finite amount of contradictory data there will always be some finite probability that creationism might be true, however close to 0 that probability may be. That's just the maths.

If you can show an internal contradiction, that's another matter. You appear to know the Bible fairly well. Can you demonstrate internal inconsistency?

It is inconsistent.. In Genesis there are two distinct creation myths which mutually contradict each other.

As a description of the physical world the Bible is Nonsense. It gets the age of the Earth wrong, it puts the creation of the Earth before the creation of the Sun and Stars. It teaches that the Earth does not move. All of this is wrong, wrong, wrong. It is nonsense, it is balderdash, it is humbug.

As Galileo once said: The Bible does not tell us how the Heavens go but how to go to Heaven.

The problem I would have to argue with using inconsistency is a measure is that we don't have a set of logical propositions, but rather we have a list of interpretations. Unless you can argue that it is impossible to interpret this in a logically consistent manner, we don't have


All of those issues are empirical.

Which is why they make sense. With the exception of some theoretical doings in math and physics the only thing that makes sense are issues that can be decided empirically.

Empirical - Good, Philosophy (especially metaphysics and theology) -Bad.

ruveyn

Well, ok, but the issue is going back to Gromit's statement about Bayesian methods. Only contradictions can be 0, but empirically, nothing can be absolutely proven as 0, only approach it, simply because some convoluted set of possibilities can be constructed for any believed empirical issue.

*clears throat*


So what if creationism makes no sense? You don't whine about Tolkien using magic, why hold the Bible to a different standard? Give it a couple hundred years and people will hold Lord Of The Rings with a similar reverence (some people do already). Debating the scientific accuracy of a book presumably written by hippies two thousand years ago is just ridiculous. Seriously.

Because people actually believe the creation account, and want it taught to children in lieu of science. No one actually believes in Orcs, or advocates teaching classes about hobbits.

Speak for yourself. We need to start preparing our children TODAY for the orc threat from the lands of Mordor!!

Tolkien never claimed his trilogy or the Silmarilion was true. Religious Christians assert that the Bible is True and the very word of God. Tolkien knew he was writing fiction and he never claimed otherwise. The people who push the Bible claim the bible is factual.

ruveyn