What about Noah and Population Genetics?

I just finished reading the Biologos discussion on the bottlenecks, and I have a couple of questions related to the TMR4A analysis:

Would it be easy to define the TMRNA values for growing numbers of N? If you would plot the medians for each N, would this help in ruling out radical changes in mutation rate, or perhaps find other anomalies? I suppose you could use the same method for calculating the minimum bottleneck within the last 10 000 years allowed by the genetic data?

A related question: how about modeling the anthropologically universal flood bottleneck? That would be TMR10A (Noah and his wife + the three wives of their three sons), and even front-loading Adam and Eve with a diversity of stem cells wouldn’t help with this challenge.

When I wear my YEC hat, it seems my best bet would be to resort to higher mutation rates after the flood. There are already multiple lines of reason for why this might be postulated within the YEC paradigm:

  • the declining lifespans need an environmental cause that affects everyone to the same degree (An increased mutation rate would tend to prune any traits that are not essential for the survival of the population, such as health long after fertile age)
  • there is the need for ultra-fast speciation after the flood
  • perhaps accelerated nuclear decay during the flood or other environmental effects of the flood could make the post flood environment toxic.

So an increased mutation rate wouldn’t be so ad hoc, given the other postulates of the paradigm. But would this kind of a scenario leave a trace that could be seen (or ruled out) in the TMRNA values?


@otto, welcome to the forums!

In addition to your link, a key place to look is here: Heliocentric Certainty Against a Bottleneck of Two?

These are great questions. There are a lot of people interested in these questions, and they require some time to fully address. I’ll try and separate them all out first; let me know if I missed anything.

Yes, we can do that. I’ll caveat the reasons why this will be a large underestimate of the bottleneck limit 10 kya later, but I’ll compute a range of them and get back to you.

That is correct. Looking at the TMR10A is a great thing to plot. I’ll be doing that.

You are right. Before we look at the data, however, think through what that would require. What would the date be that would be needed for TMR10A for this to be sensible? What should be the expected ratio between TMR4A and TMR10A under this hypothesis?

TMR4A per se does not rule this out, but other summaries of the data readily available to us do rule it out. Engage my earlier question to start to make sense of that.

We can and should look at other data too, to see what mutation rates we actually observe in the past, and how they line up with past events. There is an important body of work that presents a serious problem for the YEC paradigm, undermining any hypothesis of dramatically increased mutation rates in the last 10 kya.


They build this website to display the results, which is really important to look at closely.


Notice this key and substantiated claim from the abstract:

We used this to produce an atlas of worldwide human admixture history, constructed by using genetic data alone and encompassing over 100 events occurring over the past 4000 years.

These are historical events that we know from historical records. Things like the Mongol invasion of China, and slave trade, and the Silk Road. They are identifiable in the genetic data, leaving an imprint they can detect, if and only if mutation rates are largely constant over the last 4000 years. You can see the data yourself on the website.

This is not TMR#A data, but it severely undermines any notion that rates were substantially different over the last 10 kya.

The other option is to just allow for people “outside the garden,” as is often wondered about in traditional theology. I’m not sure if you’ve seen this yet, but it turns out that:

Entirely consistent with the genetic evidence, it is possible Adam was created out of dust, and Eve out of his rib, less than 10,000 years ago in a divinely created garden where God might dwell with them, the first beings with opportunity to be in a relationship with Him. Perhaps their fall brought accountability for sin to all their descendants. Leaving the Garden, their offspring blended with their neighbors1 in the surrounding towns. In this way, they became genealogical ancestors of all those in recorded history. Adam and Eve, here, are the single-couple progenitors2 of all mankind. Even if this scenario is false or unnecessary, nothing in evolutionary science unsettles this story. So, evolution presses in a very limited way on our understanding of Adam and Eve, only suggesting (alongside Scripture) that their lineage was not pure.

This seems like the easiest way to make sense of a literal reading of Genesis. Many YECs are open to this. Those that are not, usually, it is nothing to do with Scripture, and everything to do with anti-evolution bias.


I also welcome you @Otto, and yours is the kind of question I have often said must be addressed no matter what happens with Genealogical Adam.

The Good Doctor has sent you to a thread showing the scientific case that there was no bottleneck of two in any time which reasonably comports with Adam and Eve, and the same kind of data will also rule out a more recent bottleneck of a single Y-Chromosome from Noah. None of this contradicts scripture, but it contradicts what a lot of people thought was in the scriptures. The Christ-Centered framework for Early Genesis shows there is no conflict between what the genetic data says and what the scriptures say.

1 Like

It seems the date is proportional to the mutation rate: Assuming we get a TMR10A median of 200kya (TMRCA of 2mya divided by 10) with a steady mutation rate, and we try to make this fit within ~1800 years from 3200BC to 1400BC (From the Flood to Moses, using the LXX date for the flood and the longer sojourn in Egypt), this would need on average 100 times (= 200 000 / 2000) mutation rate during this nearly two millennia.

Another option would be to try to inflate the genealogies, but I see a trade off here: the timescale would have to be multiplied by a large factor for this to make any difference, but maintaining a higher mutation rate for more than ten thousand years seems to bring new problems.

Well, is this sensible? My intuitions would say two orders of magnitude is huge. I guess we could use some radiation induced data for a sanity check here? But this number is so large that such a signal shouldn’t go unnoticed even if it was possible.

Unless we assume a higher mutation rate also before the flood (this would be more ad hoc than post flood, but I suppose one could try blaming it on the fall), one might go with “special biology” in Adam and Eve (i.e. created variation in the reproductive stem cells). As long as they had 5 or more children (Kain, Seth and their wives would make already four) whose alleles were distributed in the population, this would be filtered through the flood bottleneck as “very old”.

Most of the events under the menu “Historical Event” seem to be within the last 2000 years. Some go back 3000 years, such as “Shared ancient events and diverse modern admixture in Central Asia”, which mentions:
“In the original (Full) analysis, the Kalash possess our oldest estimated date of most recent admixture, of 600BCE (990-210BCE)”

Wearing my YEC hat I would place the increased mutation rates to the time of lowering lifespans, which seems to end around 1400BC, or 3400 years ago. It seems to me (I’ll try to take a deeper look later) that this data set has least coverage back during 2000-1500BC, so I’m not sure how well it is suited to refute my hypothesis.

Yes, this is an option if we accept people “outside the ark” surviving the flood. But if we take the anthropologically universal flood to be here the hypothesis to be tested, I don’t see how this would change the picture.

1 Like

Exactly. We would see that in the data, especially because radiation induced mutations have a specific signature. More importantly, we do not know of any cases where humans can survive that much mutation. We have never observed a case of humans mutations mutating 100x faster than the average rate. Mutating about 2x faster is enough to substantially increase risk of autism. It is not plausible to imagine mutations at that high a rate. Such a high rate of mutation just kills us.

That is right. TMR10A should be close to TMR4A if that is the case. Actually, it should be nearly equal. TMR10A should be about TMR4A - 2,000 years.

Actually, using ancient DNA, we can easily get back as far as 10 kya. Admittedly that is other datasets, but it is the same basic point.

Also, keep in mind, for every 100 years we can validate mutation rates farther back in the past, that is 100 years least for hypermutation. That increases the among of mutation we’d need to make sense of the data, I’d guess, well over 500x or even up to 1000x. Once again, we have never observed mutation at this rate ever in humans, even after radiation exposure. That rate of mutation, and you’d just die.

Yes, there would be people outside the ark too. The passage itself does not say that all people were killed, just that all Sons of Adam in the region were killed. So that is hardly a problem. You’d have to do a lot of work to demonstrate “anthropologically universal” includes people who did not descend from Adam. (@anon46279830, help me out here?)

1 Like

Indeed. And if one reads the opening pericope of Genesis 6 to be an explanation of why God brought judgment through the flood, that leads to the interpretation that the Noahic Flood was meant to judge and purge the mixture of non-Adamic and Adamic tribes which had brought much evil on the ERETZ (land).

[I’m not trying to advocate for that position per se, but just mentioning it for those readers who may be new to that point of view.]

1 Like


I assume the reason we are using the LXX time frame is because it is the only way to accommodate (just barely) a global flood? The Masoretic timeline puts a global flood right in the middle of the Egyptian culture … which is notably not interrupted by any massive flooding, let alone total destruction.

In the interests of accommodating YEC biases, I can accept LXX dating … but how do we get YECs to accept it? They are rather attached to the King James translation - - and that was based on the Masoretic version.

NOTE: moving the Flood moves the de novo creation of Adam/Eve as well. So I can see I will have to modify my fixation on Adam and Eve’s special creation being just 6000 years ago.

1 Like

Hi @Otto

I am afraid I am bushed after many hours on the road but let me start with this…if you take a Christ-centered and truly Trinitarian view of the text of Early Genesis the account looks very different from the traditional view.

The whole case is laid out in detail in a book I wrote, but you might be able to get the gist of it from just watching a few videos. Let me start with the after-flood part first, because it is actually easier to see here that the Bible assumes there were other people already around when the clans of Noah came out of the hills…

I think you will also be surprised at what the text really says in the Hebrew about the animals which were on the ark…and how poorly translated our English text is…

And crucially, this last one sorting out who is saying what in Genesis 6 and 7. This is about 250 pages in the book condensed into one 20 minute video so I think to get it is going to require one’s full attention and perhaps a few stops to go look something up in the concordance et cetera…

Between these three I think it pretty much covers the most confusing verses in the flood account. I invite you to be a Berean and scrutinize these things closely to see if they are so, and of course hold me accountable with any questions or places you think I might have mis-handled the Word.

1 Like

There is a way to look at the Genealogies with no gaps in time (and a reasonable explanation for a small gap in generations) so that they are literally true in a way which should accommodate even most YECs and still have the flood more than 6K ago…


I know that in the past I had some vigorous objections to your work with the genealogies.

But it was in the posting above that I realized I would rather have YECs insist on a Regional Flood before the rise of Pharaonic Egypt… than to tell YECs they couldn’t have a flood at all.

@Otto, I haven’t re-read your posting above, but I assume you are okay with a regional flood instead of a global one, yes?

A global flood (whenever it might be placed) means all of our beautiful “pre-Adam adams” get wiped out … and that’s not really workable.


That sounds even worse than I had expected, but I’d like to see some evidence based baseline of how high a mutation rate makes a mammalian species infertile. After all, even Chernobyl is inhabited by wildlife that seem to cope quite well with the disaster.

It seems tumor cells can have an increased mutation rate of four orders of magnitude, so at the cellular level even a 10 000x larger rate doesn’t lead to a total meltdown (yes, cancer cells are seriously malfunctioning, but they are still functional enough to reproduce wildly):
“The values of the baseline and raised intrinsic mutation rates are 10^−8 and 10^−4, respectively.”

Then there is the background that the starting population would have been extremely healthy to start with: You don’t live to be a methuselah if you don’t have something going on for your biology. Not to mention Adam begetting Seth at a golden age of 230 years. And like with cancer cells, one might expect the problems to arise when multiple safety systems fall simultaneously. So we would expect these “demigods” to be at least somewhat more resistant to the mutation load.

There is also a natural mechanism for weeding off some lethal mutations, at least for sperm cells: a dead sperm doesn’t swim too well. For eggs it isn’t as easy, but some problems may be compensated by heterozygosity, and in the end of the day an increased mutation rate would mean we would see a high level of infertility. As long as fertility isn’t too close 0%, the population should survive. (It actually seems like the patriarchs had some problems with reproduction: Abraham famously had only two boys, one per wife, and Isaac being a miracle; Isaac had twin boys; Jacob had 12 sons, but with four women, and half of the sons with Leah).

So I won’t close that door just yet, but I’m sure you have some hidden gems waiting for me. :slight_smile:

Maybe you can help me with this one: it seems to me we could have, under the “special biology” option, an apparent bottleneck of way more than 10 alleles for the second generation from creation (Even without taking into account their longevity, Adam and Eve could have easily had 4-20 kids, giving a bottleneck of 8-40 alleles).
If we have only a vague guess of how different each created chromosome would be, it is hard to say how big a TMR4A or TMRCA we should expect. This would be “virtual history”, so it seems your equation of TMR10A ~= TMR4A - 2,000 wouldn’t apply here.

The problem then would be that any classical YEC would question the timing of the historical admixture events before 2000BC, so the use of this for independent calibration would need some pretty robust backing.

1 Like

Abraham had a lot of boys from his third wife. But it seems to me your appeal to cancer cells as a way to find a way for a higher mutational load is just as objectionable. They are not “functional” in any good sense.

PS- Rather than spending 100% of your energy looking for a way to make a flood that wiped out humanity work with the science, why not spend some of that energy considering that the scriptures when rightly handled do not teach that such a flood occurred? Why not give those vids a chance and we can talk about scripture and not just science…

1 Like

A YEC with strong MT inclinations would have to question the timeline of Egyptian history (the usual approach seems to be to make some dynasties coexist), and somehow squeeze everything to the time after 2200BC of Babel. Which obviously brings some serious problems with it.

This might play a part of why YEC archaeologists seem to like LXX. But there is also the fact that the LXX genealogies seem to have way better exegetical support (as documented by Henry B. Smith in the article linked above).

I think the mentioned article does some good ground work there (It was published in the journal of AiG after all). I’m not too familiar with the King James movement (having become a YEC before ever touching a KJV), but it might help if you note that some times Jesus quotes the OT in ways that are supported by LXX but not MT.

Sure, I have many hats, and I attack each of them at turns. It seems my remaining scriptural questions with a local flood are smaller than what I have with the cataclysmic YEC scenario of completely overturned geography.

Exactly, and that is why I’d really like to see how the population genetics data will treat the global flood.


Thanks, I had missed that one!

I thought that goes without saying. :slight_smile:

My point here is to find a well documented limit for how large mutation rates a human population could tolerate without extinction. The cancer cells mutate at a 100x higher rate than what is required for this YEC scenario, so it is clearly not to be taken as an example of victorious survival and heathy functionality, but rather the extreme at the border of lethality. A complete human body will not suffer as much without dying, but I hope to see an empirically determined limit for how much is too much.

Perhaps this will help you see where I’m aiming at: I’d rather see @swamidass beat up a superman than a strawman.

For a fairly simple reason: the title of this topic is “What about Noah and Population Genetics”, and I really like taking things one at a time.

But thanks for the videos, I had a quick listen and you seem to have good stuff there!

1 Like

In fact, @Otto, the great majority of OT references in the New Testament do derive from the LXX.
But aside for the eccentrics (even me) on this blog, I’ve never encountered any YECs who have said,
“… and that’s what the LXX says!”

I’m not following you about population genetics treating the global flood. There’s nothing to consider.

A global flood eliminates the pre-Adamites. This blog is introducing the pre-Adamites in a serious vein.

There was some considerable back-and-forth discussion between Venema (BioLogos) and Dr. Buggs (I.D.) on how far back do we have to go to the “Wall of Genetic Noise” which makes it theoretically possible to hide a 2-person bottleneck (which is an Adam/Eve hypothetical). If my memory serves correctly, the current Human Genome Diversity is so pronounced, we have to go back 700,000 years (which takes us well into the pre-sapiens timeframe) before we can “HIDE” a hypothetical bottleneck of 2 humans.

A similar effort could be done for a 6 person bottleneck (Noah’s children with wives). And it seems obvious that it would be easier to hide a 6 person bottleneck than a 2 person bottleneck… but even if it takes us back to as recently as 300,000, it not only doesn’t conform to the usual timelines for a global flood - - - but it’s not going to support @swamidass’s work.

PS: The reason I use the term “hide” in conjunction with the bottleneck analysis is that human genome diversity is so pronounced, and a 2 person bottleneck so drastic, the data clearly excludes the possibility of such a bottleneck for any time frame significantly more recent than 700,000 years ago. The best we can expect is to allow for the possibility of a bottleneck beyond the “clear zone”, and past the wall of “genetic noise”.

1 Like

Well if you want to go at it from the other direction we now have DNA samples for hundreds if not thousands of individuals from 13,000 to 5,000 years ago. NOTE: Even if a YEC would not accept those dates, they are still people who lived in the same areas in the distant past. Even if it was 3,000- 6000 years ago what I am saying about amount of genetic difference holds. …If DNA mutated significantly faster than it does now during that period then we should not be able to build plots like the one in this link…

Or build webs of relationships like the one in this link…

And this one is only mtDNA but we should not be able to have mtDNA with only one letter difference after 9000 years…

In short, even if much higher mutation rates could have happened in theory without proving fatal we have strong evidence that they did not happen…

Therefore, a humanity-ending global flood didn’t happen. But the scriptures don’t really say that is what happened anyway. What they really say happened is in the videos/book.

1 Like

I totally agree. Take the rest on another thread.

I 100% endorse that. I have no use for strawmen arguments. If you can find a way to make it work, I’ll certainly make it known too. I have no motivation to mispresent the data.

Some Preliminaries

Yes, but it is not likely that these animals were getting 100x times the normal rate of mutation in their germline. Remember, that somatic cells are going to get far more mutations than germline cells (because they are more exposed and dividing), so you are far more likely to die, than somehow live long enough to produce viable offspring.

Also keep in mind that every child is born with about 100 to 200 de novo mutations, and as the father ages, this can approximate double. However, that increase in mutation rate makes it much more likely that the child will have autism.

What you are proposing here is that there are 10,000 or more mutations per generation for several generations. We know of no mechanism that can increase mutations rates that high, and then magically return to the levels we’ve seen currently. It is hard to imagine a process that would mutate at this high a rate with out just leaving us dead. I know of no examples of any mammal that can survive that amount of mutation, producing viable offspring. Can you find any papers that do show anything like that?

That is different. Cancer cells are not producing viable human offspring. That is an example of somatic mutation, which is far higher than germline mutation.

It is just entering the realm of science fiction to imagine mutation rates this high. I’m pretty sure we do not even see survivable mutation rates that high in insects and C elegans. I’m a medical doctor, by the way, and weeding out mutations in sperm is not going to work. What ever process is increasing mutations rates 100x in germline is doing much more in somatic cells. I’m just no sure how any individuals survive that degree of mutational load, let alone have viable offspring. Maybe I am missing something here, but that is a rate of mutation that we have just never observed.

The TMR10A Distribution

Here it is. It 181,000 years ago. So, this predicts a mutation rate about 45x higher than we directly observe in humans (if the flood was 4,000 years ago). That is really high. If we try and squeeze all the mutation into a window in the past. That means you are looking at >100x mutation rate than ever observed for >1000 years. That just boggles my mind. I’m very open to finding any models that might work, but this totally stumps me. I just do not see what could work.

So yes, if Adam and Eve do not have our biology, then yes, TMR10A does not have to be close to TMR4A. However, for them to be far appart (180 kya vs 500 kya), that would require them to be genetic mosaics, it seems.

That option, however, is not available for Noah. I do not think that trick for them is playing fair. I’d hope you agree. This seems to strongly falsify the notion that there was a Noah bottleneck of 5 individuals within the last 10 kya. Do you agree?


@Otto are you a YEC? Just FYI, so is @J.E.S, one of our moderators. We will treat you with kindness here.

TMR10A does seem like pretty clear evidence against the AIG view. I’d remind you, however, that the AIG understanding of Genesis is not the traditional view. It arises very recently in history, first in the mid 1800s, and then largely dropped. And then again in the 1960’s. It is fine if you feel the need to take that view, but the AIG position is not the traditional view of Genesis.

It would be interesting to see how you understand this this data, and how it interacts with your understanding of Scripture.

I’ll now just give a couple of brief comments, and hopefully I’ll soon have time to answer the other questions.

I agree, this seems to make the problem even worse. We might want to talk more about it later. But at this point I’ll just raise two points that may make it harder to give this effect a numerical estimate: the exposure effect would be related to the type of radiation, so that would be very much true for Alpha radiation, to a smaller effect for beta radiation but not so much for gamma rays. And correct me if I’m wrong, but the cells that divide often are also more disposable. Se the big problem are not the dead cells, but the uncontrolled cells: we are mostly talking here about the growing risk of cancer.

I know, it is hard to imagine for me too, but if it is possible, I’d use either evidence or evidence based calculations for this, rather than my intuitions. I’m ok with using intuitions for determining priors, but it is harder to crunch out bayes factors out of them. The most interesting bayes factors come out of clearly defined likelihoods.

I expected we had seen something more than the autism data by this time, but we have not, so I’ll see how far that goes.

The autism rate (not counting other autism spectrum disorders) seems to be 0.2%. The data about a doubled autism rate with a doubled rate of mutations would sit well with the assumption that we get the combined probability by multiplying the probability of the independent events. The correct way to calculate the probability is by multiplying the inverse, and for a 100x dose we have:
1- (1-0.002)^100 = 18.1%

An autism rate of 18% is certainly very high, but if autism is the canary in the coal mine for rising mutation rates, this seems to leave some room for speculation. A population could survive this, if autism was the only factor. But obviously it is not the only factor, so we need to take into account more data.

Correct me if I’ve understood this wrong, but I made a quick search for C elegans mutation rates and this seems to imply that the limit would be around 10 000 to 100 000 times for C. elegans:


I’m actually quite surprised with this number (that is speaking about per gene mutation rate, but also the spontaneous mutation rate is with the same units). If I didn’t get that completely wrong, we have here a multicellular organism which is able to stand up to 1000 times the amount of mutations needed for H. sapiens in the hypothesis we are trying to test here. We are no more talking about cancer.

So far this still seems like an open question to me, at least based on the data presented here.

Dr. Buggs was right to point out that the bottleneck hypothesis had not been tested. I would propose a similar claim, that we seem not to have tested the claim: In H. sapiens, what is the maximal mutagenesis compatible with population lethality/sterility?

And I would be happy to see this tested (Of course I’m not an expert in the subject, so probably I’ve not found the relevant studies yet, but I have more faith in your expertise, and I’ll try my best playing the devils advocate :smiley:).

Thank you Joshua! That is beautiful! So this would be pretty much in line with the 100x I calculated before.

Yes, this is exactly why we are talking about Noah: we don’t need to wrestle with the question of how fair the “starlight created in transit” category of explanations is (I would say the mosaic biology view is certainly ad hoc (=low prior), and if not in this category, it is close enough), because even if you had as much of them as you like, they won’t apply here.

If you can properly test the increased mutation rate hypothesis, this will also make any discussion of mosaic biology irrelevant.

I’m not saying the hypothesis is likely, but I’m saying we should build the argument properly into the form of a bayes factor.