I keep a list of all of the scientific evidences that are incompatible with naturalism / materialism / atheism. After our podcast episode with Gunter Bechly, I learned a new argument that he called “the waiting time problem”. I’ve been asking around to see how respected this argument is from all sides, and I found out that even naturalists agree that the problem is a real problem for naturalism.

So, let me tell you what it is first, with this article from Science and Culture.

Let me explain what the waiting time problem is first. The waiting time problem is a challenge to Darwinian evolution based on  paleontology and population genetics. Population genetics explains microevolution, like germs building up resistance to drugs. But the same calculations that support microevolution reveal that geological timeframes from paleontology are too short for the genetic changes needed for major evolutionary transitions.

How many mutations are needed for an advantageous adaptation that natural selection can select for? Complex adaptations require “coordinated mutations”. Coordinated mutations are when two or more genetic changes happen together, each harmless on its own, but together they create a useful trait that natural selection can pass on to future generations. Many adaptations need these mutations, as shown by research and the fact that many traits are controlled by multiple genes.

So the article, which is written by Bechly explains the argument:

So here is how the waiting time problem is derived:

• Evolution is supposed to proceed by random mutation and natural/sexual selection.

• Selection can only work on mutations with a positive or negative adaptive value.

• At least some adaptive advantages require two or more coordinated mutations.

• Any evolutionary adaptation requires mutations that have to arise and spread in a population. Thus, all mutations have two time constraints that mainly depend on mutation rate, population size, and generation time: the waiting time for a mutation to occur and the waiting time for the fixation of this mutation.

• Question: Does the history of life provide sufficient resources for evolution to accommodate these waiting times? The answer can be shown to be negative!

• Conclusion: Neo-Darwinism is mathematically refuted as a viable theory of macroevolution.

Now, what I care about is this. First, what does paleontology say is the window of time available for a complex adaptation? And second, what does population genetics say is a reasonable time to get a coordinate mutation that can get you a complex adaptation?

Now, putting on my Darwinist hat, the first thing I thought of is that the time might be short, but the coordinated mutations would be easier to get with a very very large population. But, since I’m not a biologist, I was wrong – a large population has its own problem. Larger populations have a longer “fixation time”. The fixation time is the time it takes for a new mutation to spread and become permanent in an entire population, so every individual has it.

Bechly explains:

Indeed the waiting time problem is not just a simple problem but rather an inextricable dilemma, because however you change the crucial parameter of population size, it will always increase one of the two waiting times. With large population sizes the waiting time for a mutation to occur decreases, but fixation time increases (the same is true for neutral evolution). With small population sizes the waiting time for a mutation to occur increases, but fixation time decreases. Thus, there is no easy way for evolution to work around the waiting time problem, e.g. by means of genetic drift and the founder effect.

Why is this argument so wonderful? Because unlike many arguments that philosophers make, we can actually do math to prove this argument. This is the equivalent of me writing a unit test with the given, when, then structure. Given a known window of time from the fossil record, when evolving a coordinated mutation, then the time taken must be less than the window of time available.

Bechly says:

The amazing power of the argument from the waiting time problem is that it does not rely on fuzzy concepts but actually allows you to do the math, based on the well-established mathematical apparatus of mainstream population genetics. All required parameters are either known empirically or their range can be reasonably estimated by comparisons with recent relatives (e.g., mutation rates, effective population sizes, generation turnover times, length of binding sites, etc). The fossil record and radiometric dating provide the data for the available windows of time.

So that’s everything in the argument. Now we have to decide whether to add it to my awesome list of atheist-defeating evidence. How to decide? Well, we have to get people on the other side to agree with us that the window of time is too small, and the waiting time for coordinated mutations is too large. And they do! Let’s take a look at the domain of human evolution, to see an example of a change that requires many, many coordinated mutations (if not done directly by an intelligent designer).

Bechly explains:

Michael Behe (2007) in his book The Edge of Evolution made the argument that the waiting time for two coordinated mutations is prohibitive for the Neo-Darwinian mechanism of evolution to work. Behe used empirical data about an actual waiting time for a coordinated mutation that conveyed chloroquine drug resistance in malaria. Applying these empirical data on human evolution, by simply correcting for the much lower population size and much longer generation time, predicted a waiting time of 10^15 years, which is many orders of magnitude longer than the existence of our universe! The mainstream population geneticists Durrett & Schmidt (2008) criticized Behe’s argument with a mathematical model and claimed that his calculated waiting time of 1015 years is unrealistic. However, their own calculations also resulted in a prohibitive waiting time of 216 million years for a single coordinated mutation in human evolution, which vastly exceeds the available window of time of only about 6 million years since the split of the human lineage and the chimp lineage from a common ancestor.

Of course, when results from a theoretical model differ so greatly from experimental real world data, we should rather trust the real world data, because every model necessarily has to make certain simplifications that can introduce errors. Consequently, Behe’s numbers are certainly closer to the true limits of the Darwinian mechanism than those of Durrett and Schmidt. Anyway, both numbers are prohibitive and refute the feasibility of a Darwinian mechanism of macroevolution.

So, even the naturalists agree that evolution doesn’t work, because the 216 million years they calculated is still bigger than the 6 million years that are available for the macroevolution they want to explain.

But, and this is interesting, there is an application here to the recent post I wrote about the chimp – human DNA difference. Often thought to be 1%, the new evidence shows that it is closer to 15%. But the main thing is, Bechly says that you need MILLIONS of coordinated mutations to make all the changes to go from chimp to human:

Using a different model, Sanford et al. (2015) applied a computer simulation to calculate the waiting times in human evolution based on reasonable estimates for an ancestral hominin population of 10,000 individuals and a generation turnover time of 20 years. They arrived at fixation times of 85 million years for a single codependent mutation, and 1.5-15.9 million years for a single specific point mutation. This is very remarkable and prohibitive, considering the fact that the assumed 5% difference in the human vs chimp genome translates to millions of mutations that had to arise and become fixed within 6 million years since the assumed separation of their respective lineages. This comes [in addition to] the waiting time problem for coordinated mutations but is a separate argument (see below).

Sanford, by the way is a brilliant scientist who also happens to be a young Earth creationist. But his paper appears in a mainstream scientific journal.

So, I think it’s reasonable for me to add this argument to my list of scientific arguments against naturalism, and for a Creator / Designer.

Here is the updated list:

1. origin of the universe
2. cosmic fine-tuning
3. origin of life (specified complexity)
4. Cambrian explosion (and other explosions)
5. galactic, stellar and planetary habitability
6. molecular machines
7. non-material mind, e.g. – split brain surgery
8. the waiting time problem

Of course, there are philosophical arguments like “the moral argument” and “St. Thomas Augustine says” and “do you want fries with that?”, but who cares about philosophy, when you can have SCIENCE! Scientific arguments are the best kind of arguments!

Do you know any more evidences that are solid enough to go on my list of evidences?