What about Noah and Population Genetics?

That is not true. The opposite, really.

Your skin, hair, blood cells, and internal organs all have dividing cells. They are the first to go. They die, and you die in days, weeks or months. It is a very horrible death. You loose your hair, you vomit and loose your appetite, and you stop producing red blood cells and white blood cells, becoming unable to fight off infections. We actually administer lethal doses of radiation in the hospital to treat many cancers, but we have to do a bone marrow transplant to keep patients alive.

Cancer is not what kills you. That is just the long term risk if you somehow survive.

As for radiation as the mechanism of this (as proposed by RTB), it would cause primarily deletions and rearrangements, both of which are much more dangerous than point mutations (which is most of what we see). Rearrangements can delete genes or break them totally, and so can deletions if they cause a frameshift.


The lethal dose of radiation for humans is ~400 Rads, which kills 50% of people. At 1kRads, everyone is dead. The LD50 for uniform low LET irradiation of man - PubMed

Roughly 1 rem, or 1 Rad, is approximately 3 years of natural radiation exposure. So we get about 10 rem, on average, before having kids. Increase radiation by about 40x, and we are up to the point where 50% of the population is dying every generation. More accurately, there would be even more dying if you take cumulative effects into account. Of course, keep in mind:

  1. Most of these mutations are not the right type any ways, and the germline cells will be among the the least likely to be affected (because they are buried deep in the case of eggs, and protected if they are testes).

  2. Infertility would take place before that point too, as rearrangements cause genetic interference, and many of the mutations that won’t kill us will still injure development irrecoverably.

  3. We do not see this many deletions in the human genome. We see mainly point variations, which are less likely.

You could look at alpha and beta radiation, but that type of radiation does not penetrate deeply.

He does not say that any more. He thinks that TMR4A does test the hypothesis, and rules it out before 500 kya.

That is not an ethical study. We know that we have just never observed anything approaching that rate in large mammals (or probably even small mammals). It has never been observed.

I think we have.

I’m all for that, but the arguments are not determined to be improper or proper based on use of a “bayes factor.”

The evidence is really clear. The YEC model of a genetic bottleneck of 5 people for Noah does not fit the evidence. To make it fit, we would need to invoke some large, ongoing miracles, where God is adding mutations to human populations for thousands of years in the past, and then stops for thousands of years. We also need to either (1) believe God was miraculously adding mutations before the flood too, or (2) posit that Adam and Eve were Genetic mosaics. Those options might be the only way for this to work.

Of course, ongoing miracles of that sort are unattested to in Scripture, and do not have any reason, it seems, except to fit the data. There does not seem to be an independent reason God, which should substantially reduce our confidence in it.

1 Like

Thank you for the clear explanation!

This is a clear numerical limit, and it is clearly sufficient to show that

This is true. The mutagen should produce point mutations for there to be any sense in postulating it. For the sake of completion we will want to calculate the limits for point mutation inducing mutagens. It seems ENU (N-ethyl-N-nitrosourea), which is known to produce point mutations, and has been studied with mice, should give us a good reference for the whole class of point mutation inducers (we don’t have to care for the infinite number of possible causes if we can rule out the whole category):

The mutation frequency induced by a 400 mg/kg dosage of ethylnitrosourea is 12 times the maximal mutation frequency achievable with a single exposure to x-rays and 36 times that reported for procarbazine, the most effective chemical mutagen previously known for mouse stem-cell spermatogonia. Ethylnitrosourea is already the mutagen of choice in deliberate attempts to create mouse models for human disease and in any experiments in which a maximal mutation rate is desired. Repeated-dose regimens similar to the ones reported here should increase the efficiency of such studies. [emphasis added]
(Dose-repetition increases the mutagenic effectiveness of N-ethyl-N-nitrosourea in mouse spermatogonia)

A fractionated dose of ENU, an alkylating agent, can produce a mutation rate as high as 1.5 × 10^−3 in male mouse spermatogonial stem cells. (Mouse Mutagenesis Using N-Ethyl-N-Nitrosourea (ENU) )

The spontaneous mutation rate in mice seems to be similar to the human rate:

Thus, the total mutation rate was about 1.1 × 10−5 per locus per generation. Assuming 103 b per locus, we obtain 1.1 × 10−8 mutations per b per generation. (Rates of Spontaneous Mutation)

The rate reported for ENU must be in per locus units, but it would still be 100x the spontaneous mutation rate of 10^-5. [edited] This would be in line with the “12x better than x-rays”, but I might have misunderstood something. That is quite a difference to the radiation induced data, but as you said, radiation causes “primarily deletions and rearrangements, both of which are much more dangerous than point mutations”.

Assuming the above, and if we can use the 12x as some kind of scale factor in comparison with the radiation, we would get the equivalent of 40 * 12 = 480 times mutation rate as a preliminary answer to the question: “what is the maximal mutagenesis compatible with population lethality/sterility?”

Am I completely wrong with that one? Feel free to correct me. :slight_smile:

Exactly: he was right at the time when he said it.

There are certainly unethical ways to perform the study, I must agree I seem to have messed things up again with the small mammals. :sweat_smile:

I hope you don’t feel irritated by this, but it seems to me like we still have some work to do. But we are making good progress!

I fully agree with that, once we can rule out the numbers related to point mutations. I must add that the ongoing miracles is really an intellectual suicide even by YEC standards!

Help me out with this too?

I’m not any more.

I always had the idea that if I would abandon the YEC position, it would be because of arguments that I had really understood and gone through properly.

Once I had adopted a bayesian epistemology, I couldn’t avoid seeing priors and likelihoods everywhere, and although I had previously adopted quite a nihilistic attitude to scientific certainty, some things can still be crunched into huge bayes factors (I had been pretty much a scientific antirealist, following the underdetermination argument of Bas van Fraassen, especially when it comes to natural history).

I also saw that even if there was no shortage of intellectually lazy arguments on the YEC side, the intellectual honesty wasn’t admirably higher on the other side either (I didn’t feel an emotional pull, just the push from both sides): quite often the arguments were faulty of circular reasoning, and didn’t really give the YEC side any benefit of doubt, even for the sake of the argument. I felt like few scientists really deal seriously with YEC arguments, and those who do, too often do it arrogantly and with poor arguments.

I want to mention that I wasn’t really actively searching for the best old earth argument I could find: it was more like “I have plenty to do with the ones coming at me, and I’ll take them one at a time to see what the are made of”.

At the time I was open to questioning any of my core arguments from Scripture, at least in principle (It becomes pretty obvious within the bayesian framework that its hard to justify a posterior of one for any hermeneutic conclusion, especially when we are talking about a dead [or resurrected] language and a large cultural distance). But for quite long, simply the amount of seemingly strong and unanswered (for me anyways) biblical arguments was something that made me doubt that even if any single one of them could turn out to be flawed, what is the likelihood that they all are?

It seemed like abandoning a YEC theological paradigm was more like ad hoc theology than an intellectually honest way of looking at the scriptures. I had seen some fine points from old earthers that I could agree with and even accommodate in my YEC theology, but the cornerstones seemed to be untouched.

When I found satisfying answers for a couple of the core arguments, it kind of made the crack in the paradigm: I started to systematically re-evaluate and analyze the biblical core arguments. I pointed the same skepticism and bayesian analysis into seeing if I could poke holes into the rest of them while sticking to a strict methodology. It appeared to me that most of the arguments could be attacked quite powerfully, and there were some strong biblical arguments that went against the AiG catastrophism. Even the remaining unanswered questions didn’t seem so strong anymore.

I hope to go into more detail some time later. :slight_smile:

1 Like

I hope this Joshua’s recent comment shows why this is a relevant discussion:

As long as the YEC model has the room for speculation by some ad hoc science fiction (beyond the cartesian demon level skepticism, which would make God a deceiver by their own words), it is not as bankrupt as it can be. To most powerfully make the scientific case against the YEC model, one should start with assumptions that they are committed to, show that it does not fit the evidence, and rule out all the logically possible categories of explanations.

The best way to do this seems to be to take things that must have a known value after the global flood (such as the amount of people, or the amount of standing trees), and have a steady rate of change that can be empirically measured, which can be independently calibrated by known historical dates 2000-3000 years back, and where any way of increasing the rate in question would have a limit or would not go unnoticed.

It seems credible that one could make the argument from those datasets, but it isn’t yet clear to me how exactly I would do that, and how powerful the argument would be. But I’m open for any suggestions!

1 Like

ENU is not a good either. Keep in mind that we see far more transitions than transversions:

ENU is an alkylating agent and has preference for A->T base transversions and also for AT->GC transitions.[5] However it is also shown to cause GC->AT transitions.[6]
ENU - Wikipedia

We would see that in the data, but we do not. Instead, we see this:

From: Common Descent: Humans and Chimps / Mice and Rats - #28 by swamidass

So before we do any more work to work out math or plumb the literature, it might be worth trying to find a mutagen or mechanism that can increase mutations in precisely these ratios. I’m not sure if one exists.

1 Like


A conventionally dated global flood is contravenes, easily,by the uninterrupted continuity of Egyptian dynasties… also, the same flood which created myriad fossils should have a million human corpses to fossilized.

However, a global flood 10,000 to 100,000 years ago makes rebuttal against a flood in that period more cumbersome.

So the only way to have a flood sometime during the Egyptian historical period is for it to be regional.

Note the correction…

I agree, sort of.

In the end, they can still decide…

In the end, if we just posit miracles everywhere, we can’t really say anything with confidence. That ultimately may be where YEC leaves us.

That is really true. I hope you have found something different here. Also look at @Joel_Duff’s excellent work:

That was true for me too. YEC is difficult scientifically, but eventually I couldn’t understand that certainty in such a idiosyncratic interpretation.

Yes, I saw strong biblical arguments against YEC too, ironically.

Yes, and that is a clear reason for why I wouldn’t use ENU as an actual hypothesis. The point I was making here is that ENU is the best known mutagen for producing a maximal amount of point mutations, and it can be used to measure the partial problem of “how much mutations could a human population bear”.

It seems you could make a strong argument from the mutation fingerprint, but at least two questions must be settled before It can be appreciated for its true strength:

  1. Do we have the data for similar mutation fingerprints for most known mutagens? (Otherwise this seems like, to use the bayesian analysis, an argument from a low prior probability: how credible is it that there would be an unknown mutagen, which would amplify the natural mutation pattern, without making measurable differences in the pattern? And if such a mutagen exists, how easily could it affect the whole population, or even the whole biosphere?)
  2. Is the pattern of “natural mutations” caused by one single cause (in which case a simple amplification would help remarkably), or is it a sum of several independent mechanisms, which would need to be independently amplified in the right proportions to give the pattern (in which case the likelihood is significantly smaller)

Otherwise that data seems to be suitable for another kind of strong argument for an old earth, but lets keep that discussion somewhere else for the moment.

I don’t like to leave my arguments in the state of “I’m not sure if a suitable mechanism exists”, if we can do better. We should give the devil his due by assuming for the sake of the argument, that such a mechanism could be found and applied, and see if we can still derive a baffling ratio of likelihoods.

So we can analyze the probabilities P(“unknown magical mutagen”|YEC) and P(Evidence|YEC & “unknown magical mutagen”) separately and in the end we can show how many and how severe ad hoc assumptions have to be made to save the hypothesis.

I would call that beating up the superman. :smiley:

Sorry @Otto, that is not how science works.

It is an issue of resources. It is reality that this is often how things are left. Though, this gives a way forward for people who want to scientifically challenge this mutation rate. A research program to make the scientific case would:

  1. Identify a mutagen that produces mutations at these ratios.
  2. Demonstrate this mutagen can mutate mice or rats at 50x to 100x their natural rate, without killing them or reducing mutation rates.
  3. Produced a plausible explanation (or evidence) why everyone across the entire globe was exposed to this mutagen.
  4. Demonstrate that ancient genomes do not contradict the mutation rate.

If that can be done, that full cohort, then perhaps even that might become part of mainstream science. As you can see, that is a very difficult thing to do, even if it is what happened in the past.

You two questions fine, however what you are running into a “fine-tuning” problem. Such a fine-tuning is highly non-parsimonious. We also have an independent way of testing mutation rates in the distant past (ancient DNA).

For such a reason, such a position is high vulnerable to challenge.

That is not how science works. It is invitational.

We invite those who think this is a real possibility to show us how it is coherent with the evidence. We always maintain the humility that we missed something. We hold results provisionally.

I see what you are trying to do @Otto. It just does not work that way. Science does not bring us to the type of confidence for which you are looking. It is invitational.

Yes, one could make a strong case from Egypt, assuming the YEC is strongly committed to the MT timeline. But there are practical problems and therefore I prefer other arguments.

First the YEC might try something like Velikovskian timelines, where the dynasties overlap and some co-exist (Two short articles here and here).

When pushed on the facts, the YEC might back down on a particular timeline, but demonstrating that the timeline is solid at each point includes investigating a mountain of evidence from several disciplines (see a lengthy introduction to the issues here). In the end of the day, if you can engage the YEC long enough, and you have the patience to go through it all, he will probably decide to go with the LXX dates.

This seems to be happening with AiG, at least that last article from 2014 includes the mention of collaboration with ABR, quoting: “This new collaboration is aimed at using the respective strengths of each ministry to help each other through the difficult areas of science and archaeology, including all things Egypt. Although we might not agree on specific dates or timing of events, as young Earth creationist organizations our agreed aim is to preserve the integrity of Scripture at all times.”

It just happens to be the case that the article linked previously in support of LXX chronology is a 2017 paper by an ABR member in AiG journal.

The first Egyptian dynasty begins after the LXX flood date, so one could make the case from pre-dynastic Egypt or ancient Mesopotamia, but those timelines are calibratid by carbon dating, so you would have to first make the case for C-14 (which happens to be the scientific case that really made me convinced that the YEC can’t work).

This can also be answered relatively easily: the Bible tells us that people were very violent and corrupt, limiting population growth, and one might also make a hypothesis that people didn’t reproduce with a maximal rate (the high begetting ages being some food for thought on this), but more importantly: the people seemed to live in cities and the Babel incident seems to imply they had a “bad habbit” of not scattering around the world (Gen. 11:4), so extrapolating this as part of the corruption before flood is not such a stretch. So if they all lived in a relatively small area, they could be in the bottom of the ocean (or below).

A more powerful case could be made from apes in general. The YEC would have to invoke something like regional ecosystems corresponding to the different geological layers.

Dear Joshua,

I don’t think you know what I’m trying to do. You don’t seem to graps the type of confidence for which I am looking for. I’m working from the framework of subjective bayesian epistemology (which for I’m convinced is a set of limitations for minimally rational non-deductive formal inference (in the same way and with similar precautions as deductive inference), and it happens to be also in line with the epistemic principles mandated by the Torah). I am happy to explain later why your criticism simply misses the target.

Right now this makes me frustrated, so I’ll write you back in after a couple of days once I’ve calmed down.

No hard feelings,

What I was saying there, was not that an argument is improper if it is not explicitly in the form of a “bayes factor”.

I categorize arguments in the following way:

Arguments can be (1) informal or (2) formal, and formal arguments can be either (2.a) deductive or (2.b) non-deductive.

Informal arguments are not necessarily bad or improper, and they can be very convincing if presented with good rhetorics. But when we want to make a maximally rational analysis, we might want to extract the logos component and see how far it goes, no matter who says it, or whether we feel the conclusion appealing or not. Especially when dealing with a person who has antipathy to the conclusion and distrust in most established facts, this might be necessary (assuming you can convince him to discuss the argument, and that he actually cares for the truth). Not all informal arguments are easily formalized, but it often brings strength and clarity to the argument if it can be presented formally.

The proper way of formulating a deductive argument is by using a syllogism.

The proper way of formulating a non-deductive argument is by using epistemic probabilities.

Non-deductive arguments can be (2.b.I) evidence based or (2.b.II) not based on evidence

Here I categorize all a-priori arguments into 2.b.II, even if it applies “background knowledge” which includes evidence, as long as the estimated credibility is based, not on evaluating how likely some evidence would be if the claim was true, but rather on estimating the probability of the claim before testing it with evidence.

With evidence based non-deductive arguments (2.b.I) I mean arguments that can be stated in the form “X is more credible once we know the evidence E”. The Bayes formula shows us that E is evidence for X if P(E|X) is higher than P(E|not-X), and this we can call the likelihood principle. (These turn into syllogisms when the nominator is zero, so actually deductive arguments are just a special case.)

The ratio of the likelihoods = P(E|X)/P(E|not-X), tells us how strongly the evidence supports X (or not-X), and how high an intellectual cost one has to bear, i.e. how much a-priori credibility not-X (or X) must have, so that one could still maintain a belief in not-X (or X) after observing the evidence E. Often with large datasets we can override even unreasonable a-priori biases, so in general a good evidence based argument is preferable to a-priori arguments.

This ratio of likelihoods is what we call the “bayes factor”.

To apply the likelihood principle one doesn’t need to be a bayesian: it is analogous with what a frequentist would call Maximum Likelihood Estimation. To deny the validity of the likelihood principle one would have to go with a wholesale denial of statistics and probability calculus (which I am not recommending).

So when I’m saying “we should build the argument properly into the form of a bayes factor,” I assume we appreciate the power of formal arguments, that we prefer evidence based arguments when good ones are available, and that we are committed to the likelihood principle.

No legit scientist or Evangelical I know are willing to accept the baggage that goes with Velikovsky.


The @swamidass Model allows the option of using LXX dating… and since the model has a miraculous
de novo that doesn’t interest the pro-evolution wing… they shouldn’t care where the de novo faction puts it.

I’ll take your word on that, but as a matter of fact some have tried similar timelines, and seem to have been influenced by Velikovsky. To quote from one the last link in my previous answer to you:

“For example, some of the revised chronology schemes popular with Christians also have serious problems because they follow the Russian psychiatrist Immanuel Velikovsky in promoting misidentifications, such as that of Hatshepsut with the biblical Queen of Sheba.”


See my post again (immediately above) on the flexible nature of where a de novo creation is placed. As long as Common Descent is recognized…it won’t matter to the Evolutionist Christian (who by definition is already someone who allows for strange miracles).

This is @anon46279830’s dilemma. He uses LXX dating (correct me if needed, Revealed). But he Really Really wants limited evolution… which will stand in the way of mutual acceptance.

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.