Birds & dinosaurs, prehistoric beardogs, the astonishing Chinese raccoon dog, proof of speciation, replies to comments on my "Probability Zero" post, and more
That's a well-informed argument, Dennis. Nicely done. As before, I'll link to this today and give everyone a day to comment upon this here before I provide my response.
And once more, I am extremely grateful to you for presenting this argument, as your beach mouse study has opened a door that should prove fascinating for everyone on both sides of this discussion.
First few paragraphs already show Dennis still hasn't understood the real scientific question. He opens with yet more strawmen fallacies, confusing evolution with neodarwinism, skeptics of neodarwinism with skeptics of evolution, species with subspecies, and more.
Last week we saw Dennis didn't know that the original lactose tolerance and peppered moth stories fell apart years ago. He didn't know most of his examples weren't even about new species. And of course he misrepresented Vox's argument. It was a complete and highly embarrassing disaster.
Neodarwinists and creationists are equally naive and unscientific. Neodarwinism has become a cult defending a naked emperor, and Dennis cannot escape this cult because a large part of his own identity, life and work are tied up with this cult. It's sad, but we cannot change it.
Since you exerted so much argumentative energy in the canid direction, I’m surprised that you didn’t include an extensive discussion of the thylacine — morphologically so similar to a placental canid that it’s difficult to tell its skull from that of a wolf, but genetically more closely related to a numbat, a possum, or even a kangaroo than to any placental mammal. Just sayin’.
I don't understand why McCarthy even bothers to debate you lot.
He said right above: "50 more mutations from the 2nd generation should fix around the 40,001st generation. And so on. "
Then you deliberately misrepresent this claim as it happening immediately
Vox Day and his audience are filled with bad-faith debaters constantly making straw-man arguments and outright lies because their actual arguments are just that bad.
Before fixation, just looking at frequencies, the "fixed" mutation looked like a nearly-universally shared version. You would have called it the "baseline", and the few holdout individual "mutants". From a time-myopic perspective, if you just look at previous and next population gene count, you saw a 0.0000..001% variant going extinct (which is textbook Kimura BTW). Not surprising.
The surprising part is that if look back in time and plot generation vs frequency, then 10 generations ago its frequency was (random numbers, only for illustration purposes, perhaps our host could give us more accurate ones) 99.8%. 20.000 generations ago, ~50%. 50.000 generations ago ? 1/20000, it appeared as a de novo mutation in an individual in some ancestral population.
It’s a bit less surprising if you consider that the frequency of any such mutation will follow a random walk (because : we’re considering neutral ones) and there are million of them. Some of them will get lucky.
It also explains why it’s very hard to answer : "point me to anything that got fixed in the human genome recently. If there’s 50 per generation, it should be easy, right ?"
What you’re looking at from "today" point of view : a very rare "mutation" (but which is not a de novo mutation : it was inherited, and inherited from a very distant ancestor), happening in 0.0000...0001% of individuals (so good luck finding those) that also happen to be the "main" variant in the ancestral population 50.000 generations ago (do we even have the data to identify that ? Perhaps our host can shine some light here too). It does not seems something easy to show.
50 mutations fixed every single generation? Read the fucking book, you retard. That’s not reality. That is not happening. That has never happened in any species. That is several orders of magnitude away from anything ever measured.
McCarthy is just making shit up. If he had actual data on this, he’d do so. If it happened every generation, we have at least three living generations to draw on. He could point to this in real data. He didn’t because IT’S NOT HAPPENING.
Not novel genes, no. Out of many hundreds of billions of tiny, harmless mutational differences that have entered the population of 10,000 individuals over many thousands or millions of years, roughly 50 of those hundreds of billions will reach fixation in a generation.
So you’re just making things up. That’s all this is. You’re playing pretend because facing the problem makes you feel bad.
By your fairy-tale math, 100 new genes have fixed all across the human species since I was born. How could that possibly happen? It makes no sense. We don’t reproduce in a way that would cause that.
Yes, fixation of new mutations, which is what the word “fixation” refers to in evolutionary biology, and which you misunderstood as referring to new genes.
You say the average is 50 and “there have been essentially no dramatic jumps”, which means each generation can’t deviate from that average by whatever constitutes a dramatic jump (a vague term that can mean whatever you want it to mean). There can be some 0-5 fixation generations but not too many otherwise a dramatic jump will be necessary in order to maintain the average.
By dramatic jumps, I mean a bird turning into a horse or a dog giving birth to a cat or any other such obvious impossibilities that some evolution-skeptics believe is part of the theory.
50 fixed mutations per generation? A fixed mutation is that which the entire species or kind has. So you are saying, for four hundred thousand generations every single generation, every single one, 50 mutations fixed in every member of the population, and these exact mutations passed on to every other member of the species.
Come on bro. That's impossible, as we know that in no generation is everyone descended from the same lineage, except in tiny, tiny, one family based populations.
So either every single human today is related to one family stretching back 400,000 generations, or that mutation spread through every member of the species, while humans were living short, brutal lives, where few women even survived childhood, and where there were many fathers and many mothers.
Now, we know from modern genetics that it takes well over a thousand generations to fix a gene. That is based on data that comes from non-sexually reproducing species. And that includes parallel fixation. (So it is a massive under-estimate).
Hence, You are arguing that evolution was not just a thousand times faster in the past (that is for one gene fixing in the entire population per generation), but more like a factor of 50,000 times faster (though someone should check my math). This absolutely destroys your argument that it is the product of consistent slow change. Your own argument undermines your premise.
You've literally decided to argue for the hardest possible naturalistic position (consistent, slow change). And the math, by your own arguments, massively disagrees with your slow gradual premise.
That’s only fifty fixing out of the hundreds of billions of harmless mutations that have entered the population of 10,000 individuals over millions of years. Most of those tiny mutational differences disappeared. A small fraction drifted around at low or moderate frequency for many thousands of years, and only rarely—roughly 50 of the hundreds of billions—reach 100% in a generation.
Rev. Littlefield writes: Come on bro. That's impossible, as we know that in no generation is everyone descended from the same lineage, except in tiny, tiny, one family based populations.
Dennis: Actually, while this seems counter-intuitive, everyone shares ancestry with each other over remarkably small time frames. A famous example of this is that every European descends from Charlemagne. Why? Well, to quote a Guardian article on the subject: “If you’re vaguely of European extraction, you are also the fruits of Charlemagne’s prodigious loins. A fecund ruler, he sired at least 18 children by motley wives and concubines, including Charles the Younger, Pippin the Hunchback, Drogo of Metz, Hruodrud, Ruodhaid, and not forgetting Hugh.
“This is merely a numbers game. You have two parents, four grandparents, eight great-grandparents, and so on. But this ancestral expansion is not borne back ceaselessly into the past. If it were, your family tree when Charlemagne was Le Grand Fromage would harbour more than a billion ancestors – more people than were alive then. What this means is that pedigrees begin to fold in on themselves a few generations back, and become less arboreal, and more web-like. In 2013, geneticists Peter Ralph and Graham Coop showed that http://www.nature.com/news/most-europeans-share-recent-ancestors-1.12950. Basically, everyone alive in the ninth century who left descendants is the ancestor of every living European today, including Charlemagne, Drogo, Pippin and Hugh. Quel dommage.” https://www.theguardian.com/science/commentisfree/2015/may/24/business-genetic-ancestry-charlemagne-adam-rutherford
How could a hypothetical genetic mutation who's origin is in Asia fix to my European grandchild in 50 years if none of my grandchild's parents or grandparents have any sexual selection exposure to anyone with that genetic mutation.
Exactly! That's why the expectation for mutations to fix among the entire human race in the way humans have been distributed all across the Earth for the last 50,000 years is absurd. And evolution doesn't predict it. This falsifies the rest of the examples in "Probability Zero" regarding Genghis Kahn, Black Death, and Lactose intolerance. Now you're getting it.
But for everyone to have everyone else's fixed genes, everyone has to be descended from the same line. Hence, if all Europeans are descended from Charlemagne, which is not correct, as there were other powerful men in Europe at the time with large families, but even if it were, then they are not descended from the Slavic equivalent, or African equivalent, etc, etc. My friends kids don't have my genes, and my kids don't have his genes. And this reality has been true for countless generations in history. So, you go back 50,000 generations to find a small population where these genes could fix in a smaller population, but still not everyone was descended from the same people even in those smaller populations. Everyone had their own lineages. So 50 generations fixing in every generation could still not happen. Either way, your math does not work.
So your slow gradual change model fails on the data, and how asexually reproducing species create offspring.
To make up the short fall you'd need massive jumps, massive jumps.
Let me reword my point. Every generation has its own mutations. Hence Seth did not have the same mutations as Cain. The lineages branch, exponentially, as eazh new generationis born, and older lineages keep reproducing in many cases, as well. So we may all have Adam's genes, but evolution does not stop there, if it is, as you say, a gradual ongoing process, each lineages branches. Of course through intermarriage some lineages will interconnect. But many wont, either way, 50 mutations per person, per generation may be easy, but not 50 fixations. Because it is not just based on the probabilities of how how many mutations are happening, but how much the branches are diverging. Observed fixation rates in asexually reproducing species are too slow, even in smaller populations they are not fast enough to make up 50 genes fixed per generation.
Yes, Seth did share those mutations with Cain that were fixed at that time--and that Cain shared with his wife (where did Cain's wife come from again?) And once the mutation is fixed among the small population, it will remain in the population no matter how large it becomes. You get half your genes from dad, half from mom, so if both have the same mutation, you get it. If everyone has it, all their descendants will have it, forever.
But as the branches increase, the lineages further increase. And the fixation across population falls of a cliff. Hence, its not just a case of arguing we are all descended from Adam. Its a case of saying we are all descended from Adams, and his tight knit family group thay stayed interacting for tens of thousands of generations, until the recently era. But we know that's not how the story goes, Cain was exiled far away, lineages diverge very quickly. Fixation then takes many generations.
The entire article is yet another case of the motte-and-bailey fallacy: it pretends to defend neodarwinism (evolution by random variation and natural selection) but then quickly retreats to defending evolution in general, without providing any evidence for neodarwinism.
And again, almost all of these examples aren't even species but merely subspecies. They were incorrectly classified as different species only because scientists at the time didn't know they could still interbreed, and now scientists don't want to change their textbooks.
That's the real irony: Darwin based his theory of evolution on the dog breeding example, but dog breeding doesn't involve natural selection and doesn't produce new species.
He then thought the Galapagos finches confirmed his theory, but it turned out they weren't different species, may not have evolved through random variation and natural selection, and weren't even finches...
The emperor is naked. Some people have noticed this, others probably never will.
It's fascinating that all evolutionists seem to argue with this same motte-and-bailey. Everyone knows their claim: that humans formed themselves through random processes out of primordial goop. When pressed, they start arguing things that no one ever disputed even pre-Darwin, like that dogs can be bred or that species change over time.
I have to laugh at the OP's AI-generated images, which seem seem to tell where his mind is stuck. Last time we had an AI picture of a mantis with flowers sprouting from his back, to illustrate the wonders of evolution. The one this week is especially revealing about the word-games that evolutionists like to play. The propaganda is that "birds evolved from dinosaurs". Shocking, right? The massive, scaly land reptiles that we've all seen pictures of evolved over quombatillions of years into cute little tiny feathered creatures? Wow! Evolution! And then when pressed: oh, you thought I meant that? Oh no, haha, by "dinosaurs" I really meant birds, please see this fake picture for reference.
Yet nother identically worded “motte-and-bailey” accusation. The problem is only Creationists ever imagined the bailey—sudden, shocking mutational leaps in which a fish births a monkey and a monkey births Scarlett Johansson. Evolution has always ever been about microevolution, in which changes over generations are barely noticeable, and nothing else.
Very interesting, although it seems a lot of the discussion in general, and I mean beyond Dennis McCarthy and Vox, looks like its about evolution happening at all, not the mechanism
If I understand correctly, Vox has simply proven that this cannot possibly happen through natural selection. Not that evolution cannot happen through ANY means. Perhaps MITTENS could be considered the continuation of evolution, through other means...
For me, seeing the astronomical chances of a single protein forming randomly was enough, seeing how transitional fossils don't exist was another, seeing paleontologists making up BS like Punctuated Equilibriums to justify the lack of transitional fossils and daring to call that philosophically and intellectually dishonest idea science, was yet another. Too many operate with the assumption that TENS cannot possibly be wrong, so if anyone finds a "problem" then the problem is apparent, and a series of explanations are quickly cooked up as if its infallible. Its looking like a Frankenstein monster with the doctor quickly rushing to replace the parts that are the most rotten in a desperate attempt to keep the corpse alive, but its too late. The way journals keep rejecting Vox's most important papers show interest beyond science are at play. And what we think of science and what it actually is are nothing alike
Anyway, I am enjoying this exchange. It has been one of the most interesting reads in the longest time for me, can't wait for Vox's reply, and the subsequent (probably) by Dennis McCarthy 👍
Wade Wilson: If I understand correctly, Vox has simply proven that this cannot possibly happen randomly and by chance.
Dennis: No, as Vox Day has explicitly admitted, his attack on neutral theory has nothing whatever to do with evolution by natural selection. And by the way, when I start proving cases of evolution by natural selection, many creationists will say those examples are so mind-numbingly obvious that they are trivially true--and no one denies them. I have to point out: Nope, people still deny evolution by natural selection.
Wade Wilson: "seeing how transitional fossils don't exist was another"
A couple of animals that happen to look alike aren't evidence of accumulation of small changes through natural selection, the differences between them aren't small, saying you imagine one eventually became another isn't proof of anything, as some paleontologists noted, time and time again, their imagined assumptions tend to be proven wrong when they look at the evidence. There is no evidence of small incremental changes in the past, there is nothing but entire groups of extinct animals that remain unchanged for long periods of times and then suddenly they cease to exist and are replaced by others. Which is why some of them retreat to intellectually dishonest and borderline insulting "theories" like PE to explain what TENS cannot
This was described in greater detail by Richard Milton over 30 years ago, but he wasn't the first to notice this
Well, what do you expect to find for transitional fossils except a steady progression of fossils from smaller ancestral, three toed forms to the more modern forms today? If those aren't transitional fossils, what would be?
But to be clear, prior to 30 million years ago, we find no horse fossils at all. None. We find no modern mammals at all, in fact. No giraffes. No elephants. No wolves. No tigers. No human beings. All we find are primitive, obviously ancestral versions of these mammals. Around 10 mya, we still find no modern mammals, but the fossils are clearly much closer to the modern forms than ones from 30 mya. We find this across the world with all mammals that have left a fossil record. Saying this is evidence of nothing is just to put blinders on.
Regardless, you claim there are no transitional fossils. Well, what could paleontologists have possibly found to meet your definition of a transitional fossil in, say, horse evolution?
What is it evidence of? God first filled the earth with primitive mammals? Then he wiped them out and filled the Earth with slightly more modern looking mammals? Then he wiped them all out and filled the Earth with even more modern looking mammals? etc?
Okay, but even considering the source of the following: who still needs to catch up? Most horse lovers, myself included, long ago read up and followed this closely. And let's REALLY not get me starts on my fav, the beautiful brontosaurus!! NOT (and never WAS) an Apatosaurus!!
==================
"The acceptance of Eohippus (now more accurately referred to as Hyracotherium) as a transitional fossil in horse evolution is highly contested and no longer widely accepted in mainstream paleontology as the direct ancestor of modern horses.
Eohippus/Hyracotherium was once presented in textbooks as the earliest and most primitive ancestor in a linear evolutionary series leading to modern Equus. However, modern research shows this view is outdated.
Fossils of Hyracotherium are now classified as part of a broader group of extinct odd-toed ungulates called paleotheres, not true ancestors of horses.
The supposed "transitional sequence" from Eohippus to Equus is now recognized as a misrepresentation. Many of the species once thought to form a clear lineage (e.g., Mesohippus, Merychippus, Parahippus) lived at the same time, not sequentially, and often overlap in the fossil record.
Major scientists, including Colin Patterson (British Museum) and David Raup (Field Museum), have criticized the museum displays of the horse series as misleading, even removing Eohippus from displays due to its inaccuracy.
Today, the horse lineage is understood as a complex radiation of species rather than a simple, linear progression. The morphological changes seen—such as increasing size, reduction in toe number, and dental adaptations—are better explained by adaptive radiation and microevolution within a single lineage, not a direct ancestral-descendant sequence from Eohippus.
In short, while Eohippus remains a fossil of an early horse-like animal, it is not considered a transitional fossil in the direct evolutionary lineage of modern horses by current scientific consensus. The traditional "horse evolution" story is now seen as a simplification that no longer reflects the complexity of the fossil record.
Smiling. Btw, what do you think all those early-primitive horse types were? God's first drafts?
Anyways, Eohippus is considered a transitional fossil--and obviously the evolutionary branch of horses (as with essentially all animals) is messy and not a thin straight line. Here's a full picture and explanation of the multi-branching tree: https://www.britannica.com/animal/horse/Evolution-of-the-horse
(And I have no 'grip' on an actual source; first we need to untangle the "they look alike and so must be related." Solve that and then we can look for Who-dun-it!)
We REALLY NEED a good time machine -- go back and get some genetic material from all these lineages.. Might make things worse; might clear some things up. Just don't let anyone step on any butterflies!
And in the meantime, going back-and-forth on this or that animal and "I want to classify it as one of THESE" and "no no! It's one of these OTHERS!" {shrug} I may have different view once I've read Vox's two books.
"But we should not expect these 50 mutations to fix immediately, but after 40,000 generations. 50 more mutations from the 2nd generation should fix around the 40,001st generation. And so on."
This is just a retreat to parallel fixation. By this logic after 20.000 generations you will have 50*20.000 = 1.000.000 mutations somehow "being fixed" in parallel. This is already addressed in PZ and TFG via the Bernoulli Barrier, and contradicts the Law of Large Numbers.
No, once a mutation reaches 100% then everyone who mates shares it. So all descendants inherit it by default— unless something else removes or alters it (eg. through gene loss). So after 20,000 generations, one million will have become fixed indefinitely. But many wold have been fixed for over 10,000 generations, etc. Once it’s fixed, it’s fixed.
No, after 20k generations one million mutations will not "have fixed", as per initial claim it takes 40k generations to fix even the 50 from the first generation. In fact 50 of them will only be one generation old and won't fix for another 40k -1 generations, as per provided math.
But they are on their way to fixing, eventually, so as I initially said, you have one million mutations on their way to fixing, in parallel. That has been mathematically proved impossible both for natural selection and for drift, if only one would read the books in question.
"once a mutation reaches 100% then everyone who mates shares it"
So a mutation that gets 'fixed' amongst the 10,000 humans around the fellow what birthed that miraculous gene-bit... somehow managed to reached 100% of humanity across the entire planet? They say Genghis Khan spread his personal mutation(s) very VERY far-and-wide... but how did he get it to cross the Atlantic into America and South America, and how into Australia and Africa? If his beneficial mutation reached every human who had sex and made a baby across the ENTIRE planet, what then of the OTHER 49 fixed genes from the original 10,000 humans in the set?
Again, I find it odd that you are not discussing the proofs of speciation and evolution of birds from dinosaurs--and are now focusing on Kimura's neutral theory which has nothing whatever to do with speciation or natural selection. Also I explained this in the post. Quoting:
Vox Day defines fixed as occurring in 100% of the human race everywhere around the globe and now refers to beneficial mutations appearing in humans within the last thousand years. But evolutionary theory predicts that no mutation, whether neutral or beneficial, that has arisen in the last 50,000 years or so can reach and spread throughout all populations on the planet. The reason is that over that time—and especially over the last 10,000 years—human populations have become fragmented and geographically isolated in places such as New Guinea, Australia, Tasmania, the Andaman Islands, the Pacific islands, and the Americas.
Most people of these regions have had effectively zero genetic contact with the rest of the world until very recently, if at all. Under such conditions, it has been impossible for any single mutation—whether neutral or beneficial—to reach fixation across the entire human species. The genes that helped some Europeans survive the Black Death, for example, could never have also raced across the Americas, let alone manage to have spread to the Hewa people of New Guinea, who had not seen a white person until 1975.
Instead, the roughly 20 million fixed genetic differences between humans and chimpanzees accumulated during the millions of years when ancestral hominid populations were relatively small, geographically concentrated, and tightly interconnected by gene flow.
Given this, the rapid rise of lactase persistence in Northern Europe between roughly 8,000 and 3,000 years ago is especially revealing. Although Northern Europeans already numbered in the millions, they still constituted a far more cohesive, interbreeding population than humanity as a whole. This makes them a much better proxy for the small, interconnected ancestral populations in which essentially all human fixations occurred. The near-fixation of lactase persistence across this region within a few thousand years demonstrates just how quickly a strongly beneficial mutation can spread through such a population.
Your arguement for 50 fixations 40 occuring 40k generations later is rediculous if you are arguing for natural selection.
Are you trying to argue constant selection pressure across all humans across 40k generations despite geographical movements and dietary changes? Why would a mutation to adapt to stimuli even 20k generations prior be beneficial enough to fix that much later? If that is your arguement again it is you who abandoned natural selection. Which mathmatically you have to. As is the entire point of the discussion.
You are right this has nothing to do with natural selection. It’s based on the science of neutral mutations (which have no physical consequence on the individual). These appear in and spread throughout populations at rates that are predictable. And I am not the one who focuses on neutral theory; it’s Probability Zero that does. This explains the science in more detail: https://en.wikipedia.org/wiki/Neutral_theory_of_molecular_evolution
Dear Dennis, thank you so much for engaging with Vox in this way as it's clearly going to lead to some interesting results and a part of me is somewhat loathe to intervene. It's taken me this long to see where the impasse lies as it's such a simple and fundamental one. Your quoted figure for "fixation" is for an individual to carry the gene, not the for it to have propagated across an entire species. Fixation means being fixed across the entire species, not the pick up of the gene in an individual. I hope this is helpful to both sides as there are clearly some inflamed emotions among some. I'm quite sure that Dennis is not at all being disingenuous here. May we all come to a greater and more accurate understanding by our discourse.
Hey Nemesis, thank you for the question. No, that number (50) is the quantity of mutations (out of hundreds of billions), on average, that after hundreds of thousands of years have finally been passed on to 100% of the 10,000 babies born in a particular generation. Most of the hundreds of billions of neutral mutations actually disappear, but a handful (roughly 1 out of 20,000) beat the odds and spread to the entire population.
No, Dennis is NOT saying that fixation is just for "an individual to carry the gene, not the for it to have propagated across an entire species". He is saying fixation is what it is, that it has propagated across the entire population.
"50 more mutations from the 2nd generation should fix around the 40,001st generation. And so on."
This is a mistake in your reasoning. There are many mistakes in your reasoning. But this one is something you need to understand or continued dialogue is pointless.
This image depicts the 12 petri dishes of e. coli and the allele frequency from DNA samples every 500 generations. Fixation occurs when a line hits the top of the graph. As was noted in the follow up paper, there is a constant rate of fixation. However, many alleles that fixate for a time later go extinct while no single allele replaces it.
For some reason you assume that it takes 40,000 generations for the "engine" of fixation to get started and then 50 fixations occur every generation. But that's not what the LTEE experiment shows. Fixation is rare. Stable fixation is even rarer. However, it continues at a very stable pace of one fixation per 1,600 generations on average in e. coli.
Edit: Another fundamental error that the LTEE shows isn't true.
"When there are fewer mutations available to fix, then you end up with fewer fixed mutations"
No, from the second paper "The difference between the appearance and fixation times of each successful mutation (the “transit time”) is a proxy for the strength of selection acting on a lineage. Despite the declining rate of fitness gain (Fig. 2a), we observe a broad distribution of transit times throughout the experiment (Fig. 4b). Even after 50,000 generations, some mutations appear to fix nearly as rapidly as those that occurred in the first 5,000 generations of evolution. This observation suggests that fitness differences between cohorts of mutations can remain high, with selection coefficients at least ~2log|1-Δf|/Δt~1%, even after many beneficial mutations have fixed."
Thinking skeptics don't claim birds popped out of dinosaur eggs ex nihilo. That's a strawman designed to inflame and distract from the central issue. Skeptics understand the claim is gradual change.
The problem is demonstrating how gradual changes, even neutral ones, can accumulate into functional wings, complex eyes, or new body plans within the available generations. You can't just assert "small steps accumulate" without addressing the combinatorial mathematics.
The actual debate is this: Skeptics say "Even granting gradualism, the math doesn't work. There aren't enough generations, reproductive events, or selection opportunities to coordinate the necessary changes." Evolutionists say "We've observed small changes accumulate, so given enough time they can accumulate into any structure," without actually addressing the combinatorial constraints.
Yes, this equation is correct: 400,000 gen × 50 fixed/gen = 20 million substitutions. But this assumes 50 mutations can fix (not just appear, but fix) per generation, that they fix simultaneously without interference, and that a new batch of 50 appears and fixes every subsequent generation.
Under neutral theory (where the 1/20,000 comes from Kimura's neutral fixation probability, 1/(2Ne)), mutations are selectively neutral (neither beneficial nor harmful), fixation happens by random genetic drift, and the 1/20,000 represents the probability that a neutral mutation randomly drifts to 100% frequency.
Yet for evolution to explain complex adaptation, you need beneficial mutations to fix at higher rates than neutral (which requires selection), deleterious mutations to be eliminated (which also requires selection), and coordinated changes that build functional systems (which requires selection). So using Kimura's neutral fixation is a bait and switch.
You can't have it both ways. If you invoke neutral fixation rates, you're describing random drift of inconsequential changes. If you want functionally significant evolution, you need selection, which means selection coefficients, fitness landscapes, and far more restrictive probability calculations that account for combinatorial constraints. The math in this response shows that neutral mutations can accumulate randomly, but it doesn't show they can build the coordinated, functional complexity evolution requires.
Icky writes: "If you invoke neutral fixation rates, you're describing random drift of inconsequential changes. If you want functionally significant evolution, you need selection, which means selection coefficients, fitness landscapes, and far more restrictive probability calculations that account for combinatorial constraints."
Dennis: Most mutations are neutral, a few beneficial. Day's mathematical argument was all about neutral fixations, and I showed why it was wrong, even though it has nothing to do with natural selection. Regarding beneficial mutations the math is much simpler--and *obviously* a beneficial mutation can and will proliferate in a population in severe circumstances, often reaching fixation. This is the point that everyone--including Creationists--agree is trivially true. Do you want me to provide the math for that?
I'm happy to look at the math, but while Day's post the other day may have been addressing the neutral fixations point, the overall argument(s) from PZ and TFG need to be taken in totality. For instance, unless the math addresses the population constraint issue, it is nothing more than (admittedly compelling) sci-fi story telling.
There is no bait and switch. Neutral alleles drift to fixation and beneficial alleles are selected while deleterious ones are eliminated. All of this happens simultaneously which is what the modern synthesis describes.
Just to be clear, the Modern Synthesis does not include Neutral Theory. Huxley and Mayr predated Kimura by 26 years. You're presumably referring to the ad hoc Post-Modern Synthesis, which isn't really a coherent theory per se, but rather two very distinct theories running in parallel.
To the best of my knowledge, no one has tried to synthesize the Modern Synthesis and the Neutral Theory in any mathematically serious way.
If 50 neutral alleles drift to fixation, how many of them are beneficial vs deleterious? For the math in the article to work, you have to assume all 50 are beneficial each time, every generation, simultaneously across the entire population. If there is assumption that "pops up out of an egg" analogy in play, it is the assumption that 50 beneficial alleles pop up miraculously in the ENTIRE POPULATION every generation with none of them being deleterious. If you want to claim that there is a net 50 positive out of some greater number of total alleles drifting each time (two steps forward, one step back type of situation) every generation, you need more than 50 shifts per generation. Which makes the math explode.
If they are beneficial, then they are not neutral by definition and they will fix faster than neutral alleles. Fixation does NOT require or imply that the fixes happen instantaneously and simultaneously across the population, I don't know where you're getting that from. And you DO have more than 50 mutations every generation. You have 50*N where N is the amount of new individuals.
E.coli have one single new mutation in one single individual every few hundred generations, it can literally be over 500 generations before a single new mutation happens in one offspring. Animals like humans have several dozen mutations, even up to a hundred, in every single individual that's born. There are literally orders and orders of magnitude more mutations per generation in humans than there in e.coli. The statement that "e.coli have the fastest fixation rate per generation ever observed" is a blatant lie and false.
Precisely, if they are --beneficial--, you cannot call upon Neutral Drift mathematics to make your case. Fixation requires that the beneficial mutation, spreads throughout (enough of) the ENTIRE population. How do you propose that happens across the ENTIRE population every single generation? The math on the 1/20,000 is assuming the SAME 50 FIXES every generation. That's how it becomes fixed, everyone now has it.
A neutral allele that fixes in a deme in some generation g did not start the previous generation, it started thousands of generations ago and has been drifting towards fixation for thousands and thousands of generations. And no one claims that all the differences between species are only due to neutral alleles, so that is a strawman.
Every single generation experiences neutral fixations that started thousands of years ago. The overwhelming majority of mutations are eliminated as the generations go on (and during this time there are tens of millions of new mutations happening every generation as well), and a small handful fix after ~4Ne generations. Every single generation has neutral fixations (at the rate of average neutral mutation) where those mutations that fix started thousands and thousands of generations ago.
Thus, the relevant number is "mutations per generation".
E.coli literally do not even have a single mutation per generation on average. Humans have 60*N mutations per generation on average, where N is the number of new births. The claim that e.coli has the "fastest fixation rate ever observed" is a straight-up blatant lie.
You are correct. Dennis is saying the same 50 fixations, not different ones, are occurring in every individual at the same time, every generation. No study has shown this happening.
People are comically expressing surprise and horror at the fact that
20 million/400,000 = 50.
Yes, 20 million fixed mutations in the human race acquired over 400,000 generations equals an average of 50 fixed mutations per generation. (Again, this is out of 400 billion mutations.) And it's not me who says it, but geneticists, evolutionary theorists, etc.
I’ve never understood why naturalists say that dog breeding isn’t an example of natural evolution. Humans are natural biological beings that operate according to natural biological and physical laws. The genomes are altered in micro evolutionary fashion according to those natural laws. It’s all natural. We’re told that you can’t include intelligence or design in any theory that explains the process in an attempt to differentiate it from natural evolution, so that’s what it must be.
No. All the fossils, the salamander ring, the results of the moth study, the DNA studies, the animal and plant distributions, taxonomy, etc., etc., etc.-- those are all facts and data. I just don't shut my eyes to them.
That's a well-informed argument, Dennis. Nicely done. As before, I'll link to this today and give everyone a day to comment upon this here before I provide my response.
And once more, I am extremely grateful to you for presenting this argument, as your beach mouse study has opened a door that should prove fascinating for everyone on both sides of this discussion.
First few paragraphs already show Dennis still hasn't understood the real scientific question. He opens with yet more strawmen fallacies, confusing evolution with neodarwinism, skeptics of neodarwinism with skeptics of evolution, species with subspecies, and more.
Last week we saw Dennis didn't know that the original lactose tolerance and peppered moth stories fell apart years ago. He didn't know most of his examples weren't even about new species. And of course he misrepresented Vox's argument. It was a complete and highly embarrassing disaster.
Neodarwinists and creationists are equally naive and unscientific. Neodarwinism has become a cult defending a naked emperor, and Dennis cannot escape this cult because a large part of his own identity, life and work are tied up with this cult. It's sad, but we cannot change it.
Better spend some more time on Shakespeare.
Since you exerted so much argumentative energy in the canid direction, I’m surprised that you didn’t include an extensive discussion of the thylacine — morphologically so similar to a placental canid that it’s difficult to tell its skull from that of a wolf, but genetically more closely related to a numbat, a possum, or even a kangaroo than to any placental mammal. Just sayin’.
Good point!
Dennis is telling a story and Vox is crunching the numbers that the story is founded upon. The math doesn’t give you the story that Dennis is telling.
Excellent follow up, Dennis. I don't really have a dog in this fight but it seems like you've poked a bear. Ursa Creatio?
You’re saying 50 novel genes spread across the human species every generation? Are you fucking insane?
I don't understand why McCarthy even bothers to debate you lot.
He said right above: "50 more mutations from the 2nd generation should fix around the 40,001st generation. And so on. "
Then you deliberately misrepresent this claim as it happening immediately
Vox Day and his audience are filled with bad-faith debaters constantly making straw-man arguments and outright lies because their actual arguments are just that bad.
Liars for Christ.
Smiling. Yes, while many are nice and sincere, a few do seem to be frustrated and trolling—and doing so dishonestly.
There are 3 generations of humans now alive. Go demonstrate the 100 to 150 fixations you claim. Should be easy.
Vox used data. You don't even understand what he's saying, and you just tell stories.
Before fixation, just looking at frequencies, the "fixed" mutation looked like a nearly-universally shared version. You would have called it the "baseline", and the few holdout individual "mutants". From a time-myopic perspective, if you just look at previous and next population gene count, you saw a 0.0000..001% variant going extinct (which is textbook Kimura BTW). Not surprising.
The surprising part is that if look back in time and plot generation vs frequency, then 10 generations ago its frequency was (random numbers, only for illustration purposes, perhaps our host could give us more accurate ones) 99.8%. 20.000 generations ago, ~50%. 50.000 generations ago ? 1/20000, it appeared as a de novo mutation in an individual in some ancestral population.
It’s a bit less surprising if you consider that the frequency of any such mutation will follow a random walk (because : we’re considering neutral ones) and there are million of them. Some of them will get lucky.
It also explains why it’s very hard to answer : "point me to anything that got fixed in the human genome recently. If there’s 50 per generation, it should be easy, right ?"
What you’re looking at from "today" point of view : a very rare "mutation" (but which is not a de novo mutation : it was inherited, and inherited from a very distant ancestor), happening in 0.0000...0001% of individuals (so good luck finding those) that also happen to be the "main" variant in the ancestral population 50.000 generations ago (do we even have the data to identify that ? Perhaps our host can shine some light here too). It does not seems something easy to show.
50 mutations fixed every single generation? Read the fucking book, you retard. That’s not reality. That is not happening. That has never happened in any species. That is several orders of magnitude away from anything ever measured.
You have no clue what you are talking about.
Why do you have to be so rude, Zaklog? It is not beneficial.
Because I’m tired of arrogant morons running their mouth when they don’t know anything. He acts like he’s so superior, and then spouts utter madness.
Fifty fixations per human generation? There are at least 3 generations of humans alive today. Let him show the 100 to 150 fixations he claims.
"Read the fucking book, you retard."
I have discarded your advice, oh great sage of wisdom.
McCarthy is just making shit up. If he had actual data on this, he’d do so. If it happened every generation, we have at least three living generations to draw on. He could point to this in real data. He didn’t because IT’S NOT HAPPENING.
Not novel genes, no. Out of many hundreds of billions of tiny, harmless mutational differences that have entered the population of 10,000 individuals over many thousands or millions of years, roughly 50 of those hundreds of billions will reach fixation in a generation.
So you’re just making things up. That’s all this is. You’re playing pretend because facing the problem makes you feel bad.
By your fairy-tale math, 100 new genes have fixed all across the human species since I was born. How could that possibly happen? It makes no sense. We don’t reproduce in a way that would cause that.
“New mutations” is not the same as “new genes”.
But we're not just talking about mutations. We're talking about *fixations*. Please pay attention.
Yes, fixation of new mutations, which is what the word “fixation” refers to in evolutionary biology, and which you misunderstood as referring to new genes.
He claims 50 fixations complete every generation. Where is his evidence of this? It sounds insane.
You say the average is 50 and “there have been essentially no dramatic jumps”, which means each generation can’t deviate from that average by whatever constitutes a dramatic jump (a vague term that can mean whatever you want it to mean). There can be some 0-5 fixation generations but not too many otherwise a dramatic jump will be necessary in order to maintain the average.
By dramatic jumps, I mean a bird turning into a horse or a dog giving birth to a cat or any other such obvious impossibilities that some evolution-skeptics believe is part of the theory.
Well, to be fair, the theory doesn’t say that can’t happen because the theory is built on models rather than empirical data.
All Dennis’ male subscribers: “We want bear dogs!”
50 fixed mutations per generation? A fixed mutation is that which the entire species or kind has. So you are saying, for four hundred thousand generations every single generation, every single one, 50 mutations fixed in every member of the population, and these exact mutations passed on to every other member of the species.
Come on bro. That's impossible, as we know that in no generation is everyone descended from the same lineage, except in tiny, tiny, one family based populations.
So either every single human today is related to one family stretching back 400,000 generations, or that mutation spread through every member of the species, while humans were living short, brutal lives, where few women even survived childhood, and where there were many fathers and many mothers.
Now, we know from modern genetics that it takes well over a thousand generations to fix a gene. That is based on data that comes from non-sexually reproducing species. And that includes parallel fixation. (So it is a massive under-estimate).
Hence, You are arguing that evolution was not just a thousand times faster in the past (that is for one gene fixing in the entire population per generation), but more like a factor of 50,000 times faster (though someone should check my math). This absolutely destroys your argument that it is the product of consistent slow change. Your own argument undermines your premise.
You've literally decided to argue for the hardest possible naturalistic position (consistent, slow change). And the math, by your own arguments, massively disagrees with your slow gradual premise.
Brother. Come on.
That’s only fifty fixing out of the hundreds of billions of harmless mutations that have entered the population of 10,000 individuals over millions of years. Most of those tiny mutational differences disappeared. A small fraction drifted around at low or moderate frequency for many thousands of years, and only rarely—roughly 50 of the hundreds of billions—reach 100% in a generation.
Rev. Littlefield writes: Come on bro. That's impossible, as we know that in no generation is everyone descended from the same lineage, except in tiny, tiny, one family based populations.
Dennis: Actually, while this seems counter-intuitive, everyone shares ancestry with each other over remarkably small time frames. A famous example of this is that every European descends from Charlemagne. Why? Well, to quote a Guardian article on the subject: “If you’re vaguely of European extraction, you are also the fruits of Charlemagne’s prodigious loins. A fecund ruler, he sired at least 18 children by motley wives and concubines, including Charles the Younger, Pippin the Hunchback, Drogo of Metz, Hruodrud, Ruodhaid, and not forgetting Hugh.
“This is merely a numbers game. You have two parents, four grandparents, eight great-grandparents, and so on. But this ancestral expansion is not borne back ceaselessly into the past. If it were, your family tree when Charlemagne was Le Grand Fromage would harbour more than a billion ancestors – more people than were alive then. What this means is that pedigrees begin to fold in on themselves a few generations back, and become less arboreal, and more web-like. In 2013, geneticists Peter Ralph and Graham Coop showed that http://www.nature.com/news/most-europeans-share-recent-ancestors-1.12950. Basically, everyone alive in the ninth century who left descendants is the ancestor of every living European today, including Charlemagne, Drogo, Pippin and Hugh. Quel dommage.” https://www.theguardian.com/science/commentisfree/2015/may/24/business-genetic-ancestry-charlemagne-adam-rutherford
How could a hypothetical genetic mutation who's origin is in Asia fix to my European grandchild in 50 years if none of my grandchild's parents or grandparents have any sexual selection exposure to anyone with that genetic mutation.
Exactly! That's why the expectation for mutations to fix among the entire human race in the way humans have been distributed all across the Earth for the last 50,000 years is absurd. And evolution doesn't predict it. This falsifies the rest of the examples in "Probability Zero" regarding Genghis Kahn, Black Death, and Lactose intolerance. Now you're getting it.
But for everyone to have everyone else's fixed genes, everyone has to be descended from the same line. Hence, if all Europeans are descended from Charlemagne, which is not correct, as there were other powerful men in Europe at the time with large families, but even if it were, then they are not descended from the Slavic equivalent, or African equivalent, etc, etc. My friends kids don't have my genes, and my kids don't have his genes. And this reality has been true for countless generations in history. So, you go back 50,000 generations to find a small population where these genes could fix in a smaller population, but still not everyone was descended from the same people even in those smaller populations. Everyone had their own lineages. So 50 generations fixing in every generation could still not happen. Either way, your math does not work.
So your slow gradual change model fails on the data, and how asexually reproducing species create offspring.
To make up the short fall you'd need massive jumps, massive jumps.
Matthew Littlefield: "but still not everyone was descended from the same people even in those smaller populations"
Dennis: You don't believe every human on Earth is descended from Adam?
Let me reword my point. Every generation has its own mutations. Hence Seth did not have the same mutations as Cain. The lineages branch, exponentially, as eazh new generationis born, and older lineages keep reproducing in many cases, as well. So we may all have Adam's genes, but evolution does not stop there, if it is, as you say, a gradual ongoing process, each lineages branches. Of course through intermarriage some lineages will interconnect. But many wont, either way, 50 mutations per person, per generation may be easy, but not 50 fixations. Because it is not just based on the probabilities of how how many mutations are happening, but how much the branches are diverging. Observed fixation rates in asexually reproducing species are too slow, even in smaller populations they are not fast enough to make up 50 genes fixed per generation.
Yes, Seth did share those mutations with Cain that were fixed at that time--and that Cain shared with his wife (where did Cain's wife come from again?) And once the mutation is fixed among the small population, it will remain in the population no matter how large it becomes. You get half your genes from dad, half from mom, so if both have the same mutation, you get it. If everyone has it, all their descendants will have it, forever.
But as the branches increase, the lineages further increase. And the fixation across population falls of a cliff. Hence, its not just a case of arguing we are all descended from Adam. Its a case of saying we are all descended from Adams, and his tight knit family group thay stayed interacting for tens of thousands of generations, until the recently era. But we know that's not how the story goes, Cain was exiled far away, lineages diverge very quickly. Fixation then takes many generations.
Well that new mutation would have to fix, also there’s gene loss as I noted. But it’s rare
The entire article is yet another case of the motte-and-bailey fallacy: it pretends to defend neodarwinism (evolution by random variation and natural selection) but then quickly retreats to defending evolution in general, without providing any evidence for neodarwinism.
And again, almost all of these examples aren't even species but merely subspecies. They were incorrectly classified as different species only because scientists at the time didn't know they could still interbreed, and now scientists don't want to change their textbooks.
That's the real irony: Darwin based his theory of evolution on the dog breeding example, but dog breeding doesn't involve natural selection and doesn't produce new species.
He then thought the Galapagos finches confirmed his theory, but it turned out they weren't different species, may not have evolved through random variation and natural selection, and weren't even finches...
The emperor is naked. Some people have noticed this, others probably never will.
It's fascinating that all evolutionists seem to argue with this same motte-and-bailey. Everyone knows their claim: that humans formed themselves through random processes out of primordial goop. When pressed, they start arguing things that no one ever disputed even pre-Darwin, like that dogs can be bred or that species change over time.
I have to laugh at the OP's AI-generated images, which seem seem to tell where his mind is stuck. Last time we had an AI picture of a mantis with flowers sprouting from his back, to illustrate the wonders of evolution. The one this week is especially revealing about the word-games that evolutionists like to play. The propaganda is that "birds evolved from dinosaurs". Shocking, right? The massive, scaly land reptiles that we've all seen pictures of evolved over quombatillions of years into cute little tiny feathered creatures? Wow! Evolution! And then when pressed: oh, you thought I meant that? Oh no, haha, by "dinosaurs" I really meant birds, please see this fake picture for reference.
Yet nother identically worded “motte-and-bailey” accusation. The problem is only Creationists ever imagined the bailey—sudden, shocking mutational leaps in which a fish births a monkey and a monkey births Scarlett Johansson. Evolution has always ever been about microevolution, in which changes over generations are barely noticeable, and nothing else.
That is not the bailey. Everyone understands that a colossally long time would be required for your argument.
Very interesting, although it seems a lot of the discussion in general, and I mean beyond Dennis McCarthy and Vox, looks like its about evolution happening at all, not the mechanism
If I understand correctly, Vox has simply proven that this cannot possibly happen through natural selection. Not that evolution cannot happen through ANY means. Perhaps MITTENS could be considered the continuation of evolution, through other means...
For me, seeing the astronomical chances of a single protein forming randomly was enough, seeing how transitional fossils don't exist was another, seeing paleontologists making up BS like Punctuated Equilibriums to justify the lack of transitional fossils and daring to call that philosophically and intellectually dishonest idea science, was yet another. Too many operate with the assumption that TENS cannot possibly be wrong, so if anyone finds a "problem" then the problem is apparent, and a series of explanations are quickly cooked up as if its infallible. Its looking like a Frankenstein monster with the doctor quickly rushing to replace the parts that are the most rotten in a desperate attempt to keep the corpse alive, but its too late. The way journals keep rejecting Vox's most important papers show interest beyond science are at play. And what we think of science and what it actually is are nothing alike
Anyway, I am enjoying this exchange. It has been one of the most interesting reads in the longest time for me, can't wait for Vox's reply, and the subsequent (probably) by Dennis McCarthy 👍
Wade Wilson: If I understand correctly, Vox has simply proven that this cannot possibly happen randomly and by chance.
Dennis: No, as Vox Day has explicitly admitted, his attack on neutral theory has nothing whatever to do with evolution by natural selection. And by the way, when I start proving cases of evolution by natural selection, many creationists will say those examples are so mind-numbingly obvious that they are trivially true--and no one denies them. I have to point out: Nope, people still deny evolution by natural selection.
Wade Wilson: "seeing how transitional fossils don't exist was another"
Dennis: I provided pictures of transitional fossils with the horses since the Eocene --evolution of body, skull, hooves, everthing. It's all there. Just click the link and keep your eyes open: https://en.wikipedia.org/wiki/Evolution_of_the_horse#/media/File:Equine_evolution.jpg
A couple of animals that happen to look alike aren't evidence of accumulation of small changes through natural selection, the differences between them aren't small, saying you imagine one eventually became another isn't proof of anything, as some paleontologists noted, time and time again, their imagined assumptions tend to be proven wrong when they look at the evidence. There is no evidence of small incremental changes in the past, there is nothing but entire groups of extinct animals that remain unchanged for long periods of times and then suddenly they cease to exist and are replaced by others. Which is why some of them retreat to intellectually dishonest and borderline insulting "theories" like PE to explain what TENS cannot
This was described in greater detail by Richard Milton over 30 years ago, but he wasn't the first to notice this
Well, what do you expect to find for transitional fossils except a steady progression of fossils from smaller ancestral, three toed forms to the more modern forms today? If those aren't transitional fossils, what would be?
But to be clear, prior to 30 million years ago, we find no horse fossils at all. None. We find no modern mammals at all, in fact. No giraffes. No elephants. No wolves. No tigers. No human beings. All we find are primitive, obviously ancestral versions of these mammals. Around 10 mya, we still find no modern mammals, but the fossils are clearly much closer to the modern forms than ones from 30 mya. We find this across the world with all mammals that have left a fossil record. Saying this is evidence of nothing is just to put blinders on.
Oh yeah I wanted to clarify, it is certainly evidence of something. But not of very small changes randomly accumulating over time
Regardless, you claim there are no transitional fossils. Well, what could paleontologists have possibly found to meet your definition of a transitional fossil in, say, horse evolution?
What is it evidence of? God first filled the earth with primitive mammals? Then he wiped them out and filled the Earth with slightly more modern looking mammals? Then he wiped them all out and filled the Earth with even more modern looking mammals? etc?
Vox attacks both neutral theory and natural selection, in different chapters, in two separate books: PZ and TFG.
Arguing he does not just shows a lack of engagement with the material you are supposedly criticizing.
Without trying to be rude may I ask have you actually read the books?
If not, what are we even discussing here?
Okay, but even considering the source of the following: who still needs to catch up? Most horse lovers, myself included, long ago read up and followed this closely. And let's REALLY not get me starts on my fav, the beautiful brontosaurus!! NOT (and never WAS) an Apatosaurus!!
==================
"The acceptance of Eohippus (now more accurately referred to as Hyracotherium) as a transitional fossil in horse evolution is highly contested and no longer widely accepted in mainstream paleontology as the direct ancestor of modern horses.
Eohippus/Hyracotherium was once presented in textbooks as the earliest and most primitive ancestor in a linear evolutionary series leading to modern Equus. However, modern research shows this view is outdated.
Fossils of Hyracotherium are now classified as part of a broader group of extinct odd-toed ungulates called paleotheres, not true ancestors of horses.
The supposed "transitional sequence" from Eohippus to Equus is now recognized as a misrepresentation. Many of the species once thought to form a clear lineage (e.g., Mesohippus, Merychippus, Parahippus) lived at the same time, not sequentially, and often overlap in the fossil record.
Major scientists, including Colin Patterson (British Museum) and David Raup (Field Museum), have criticized the museum displays of the horse series as misleading, even removing Eohippus from displays due to its inaccuracy.
Today, the horse lineage is understood as a complex radiation of species rather than a simple, linear progression. The morphological changes seen—such as increasing size, reduction in toe number, and dental adaptations—are better explained by adaptive radiation and microevolution within a single lineage, not a direct ancestral-descendant sequence from Eohippus.
In short, while Eohippus remains a fossil of an early horse-like animal, it is not considered a transitional fossil in the direct evolutionary lineage of modern horses by current scientific consensus. The traditional "horse evolution" story is now seen as a simplification that no longer reflects the complexity of the fossil record.
Smiling. Btw, what do you think all those early-primitive horse types were? God's first drafts?
Anyways, Eohippus is considered a transitional fossil--and obviously the evolutionary branch of horses (as with essentially all animals) is messy and not a thin straight line. Here's a full picture and explanation of the multi-branching tree: https://www.britannica.com/animal/horse/Evolution-of-the-horse
And yet, they are ALL horse-like.
(And I have no 'grip' on an actual source; first we need to untangle the "they look alike and so must be related." Solve that and then we can look for Who-dun-it!)
We REALLY NEED a good time machine -- go back and get some genetic material from all these lineages.. Might make things worse; might clear some things up. Just don't let anyone step on any butterflies!
And in the meantime, going back-and-forth on this or that animal and "I want to classify it as one of THESE" and "no no! It's one of these OTHERS!" {shrug} I may have different view once I've read Vox's two books.
(But thanks for the jousting, it's been fun!}
"But we should not expect these 50 mutations to fix immediately, but after 40,000 generations. 50 more mutations from the 2nd generation should fix around the 40,001st generation. And so on."
This is just a retreat to parallel fixation. By this logic after 20.000 generations you will have 50*20.000 = 1.000.000 mutations somehow "being fixed" in parallel. This is already addressed in PZ and TFG via the Bernoulli Barrier, and contradicts the Law of Large Numbers.
No, once a mutation reaches 100% then everyone who mates shares it. So all descendants inherit it by default— unless something else removes or alters it (eg. through gene loss). So after 20,000 generations, one million will have become fixed indefinitely. But many wold have been fixed for over 10,000 generations, etc. Once it’s fixed, it’s fixed.
No, after 20k generations one million mutations will not "have fixed", as per initial claim it takes 40k generations to fix even the 50 from the first generation. In fact 50 of them will only be one generation old and won't fix for another 40k -1 generations, as per provided math.
But they are on their way to fixing, eventually, so as I initially said, you have one million mutations on their way to fixing, in parallel. That has been mathematically proved impossible both for natural selection and for drift, if only one would read the books in question.
"once a mutation reaches 100% then everyone who mates shares it"
So a mutation that gets 'fixed' amongst the 10,000 humans around the fellow what birthed that miraculous gene-bit... somehow managed to reached 100% of humanity across the entire planet? They say Genghis Khan spread his personal mutation(s) very VERY far-and-wide... but how did he get it to cross the Atlantic into America and South America, and how into Australia and Africa? If his beneficial mutation reached every human who had sex and made a baby across the ENTIRE planet, what then of the OTHER 49 fixed genes from the original 10,000 humans in the set?
Again, I find it odd that you are not discussing the proofs of speciation and evolution of birds from dinosaurs--and are now focusing on Kimura's neutral theory which has nothing whatever to do with speciation or natural selection. Also I explained this in the post. Quoting:
Vox Day defines fixed as occurring in 100% of the human race everywhere around the globe and now refers to beneficial mutations appearing in humans within the last thousand years. But evolutionary theory predicts that no mutation, whether neutral or beneficial, that has arisen in the last 50,000 years or so can reach and spread throughout all populations on the planet. The reason is that over that time—and especially over the last 10,000 years—human populations have become fragmented and geographically isolated in places such as New Guinea, Australia, Tasmania, the Andaman Islands, the Pacific islands, and the Americas.
Most people of these regions have had effectively zero genetic contact with the rest of the world until very recently, if at all. Under such conditions, it has been impossible for any single mutation—whether neutral or beneficial—to reach fixation across the entire human species. The genes that helped some Europeans survive the Black Death, for example, could never have also raced across the Americas, let alone manage to have spread to the Hewa people of New Guinea, who had not seen a white person until 1975.
Instead, the roughly 20 million fixed genetic differences between humans and chimpanzees accumulated during the millions of years when ancestral hominid populations were relatively small, geographically concentrated, and tightly interconnected by gene flow.
Given this, the rapid rise of lactase persistence in Northern Europe between roughly 8,000 and 3,000 years ago is especially revealing. Although Northern Europeans already numbered in the millions, they still constituted a far more cohesive, interbreeding population than humanity as a whole. This makes them a much better proxy for the small, interconnected ancestral populations in which essentially all human fixations occurred. The near-fixation of lactase persistence across this region within a few thousand years demonstrates just how quickly a strongly beneficial mutation can spread through such a population.
Your arguement for 50 fixations 40 occuring 40k generations later is rediculous if you are arguing for natural selection.
Are you trying to argue constant selection pressure across all humans across 40k generations despite geographical movements and dietary changes? Why would a mutation to adapt to stimuli even 20k generations prior be beneficial enough to fix that much later? If that is your arguement again it is you who abandoned natural selection. Which mathmatically you have to. As is the entire point of the discussion.
You are right this has nothing to do with natural selection. It’s based on the science of neutral mutations (which have no physical consequence on the individual). These appear in and spread throughout populations at rates that are predictable. And I am not the one who focuses on neutral theory; it’s Probability Zero that does. This explains the science in more detail: https://en.wikipedia.org/wiki/Neutral_theory_of_molecular_evolution
Dear Dennis, thank you so much for engaging with Vox in this way as it's clearly going to lead to some interesting results and a part of me is somewhat loathe to intervene. It's taken me this long to see where the impasse lies as it's such a simple and fundamental one. Your quoted figure for "fixation" is for an individual to carry the gene, not the for it to have propagated across an entire species. Fixation means being fixed across the entire species, not the pick up of the gene in an individual. I hope this is helpful to both sides as there are clearly some inflamed emotions among some. I'm quite sure that Dennis is not at all being disingenuous here. May we all come to a greater and more accurate understanding by our discourse.
Hey Nemesis, thank you for the question. No, that number (50) is the quantity of mutations (out of hundreds of billions), on average, that after hundreds of thousands of years have finally been passed on to 100% of the 10,000 babies born in a particular generation. Most of the hundreds of billions of neutral mutations actually disappear, but a handful (roughly 1 out of 20,000) beat the odds and spread to the entire population.
No, Dennis is NOT saying that fixation is just for "an individual to carry the gene, not the for it to have propagated across an entire species". He is saying fixation is what it is, that it has propagated across the entire population.
"50 more mutations from the 2nd generation should fix around the 40,001st generation. And so on."
This is a mistake in your reasoning. There are many mistakes in your reasoning. But this one is something you need to understand or continued dialogue is pointless.
Here is the first e. coli study: https://www.cs.unm.edu/~eschulte/data/Barick%20et%20al%20E%20coli%20evolution%202009.pdf
Here is a follow up of the same ongoing study 20,000 generations later: https://europepmc.org/article/med/29045390
This image depicts the 12 petri dishes of e. coli and the allele frequency from DNA samples every 500 generations. Fixation occurs when a line hits the top of the graph. As was noted in the follow up paper, there is a constant rate of fixation. However, many alleles that fixate for a time later go extinct while no single allele replaces it.
https://europepmc.org/articles/PMC5788700/bin/nihms908078f1.jpg
For some reason you assume that it takes 40,000 generations for the "engine" of fixation to get started and then 50 fixations occur every generation. But that's not what the LTEE experiment shows. Fixation is rare. Stable fixation is even rarer. However, it continues at a very stable pace of one fixation per 1,600 generations on average in e. coli.
Edit: Another fundamental error that the LTEE shows isn't true.
"When there are fewer mutations available to fix, then you end up with fewer fixed mutations"
No, from the second paper "The difference between the appearance and fixation times of each successful mutation (the “transit time”) is a proxy for the strength of selection acting on a lineage. Despite the declining rate of fitness gain (Fig. 2a), we observe a broad distribution of transit times throughout the experiment (Fig. 4b). Even after 50,000 generations, some mutations appear to fix nearly as rapidly as those that occurred in the first 5,000 generations of evolution. This observation suggests that fitness differences between cohorts of mutations can remain high, with selection coefficients at least ~2log|1-Δf|/Δt~1%, even after many beneficial mutations have fixed."
Thinking skeptics don't claim birds popped out of dinosaur eggs ex nihilo. That's a strawman designed to inflame and distract from the central issue. Skeptics understand the claim is gradual change.
The problem is demonstrating how gradual changes, even neutral ones, can accumulate into functional wings, complex eyes, or new body plans within the available generations. You can't just assert "small steps accumulate" without addressing the combinatorial mathematics.
The actual debate is this: Skeptics say "Even granting gradualism, the math doesn't work. There aren't enough generations, reproductive events, or selection opportunities to coordinate the necessary changes." Evolutionists say "We've observed small changes accumulate, so given enough time they can accumulate into any structure," without actually addressing the combinatorial constraints.
Yes, this equation is correct: 400,000 gen × 50 fixed/gen = 20 million substitutions. But this assumes 50 mutations can fix (not just appear, but fix) per generation, that they fix simultaneously without interference, and that a new batch of 50 appears and fixes every subsequent generation.
Under neutral theory (where the 1/20,000 comes from Kimura's neutral fixation probability, 1/(2Ne)), mutations are selectively neutral (neither beneficial nor harmful), fixation happens by random genetic drift, and the 1/20,000 represents the probability that a neutral mutation randomly drifts to 100% frequency.
Yet for evolution to explain complex adaptation, you need beneficial mutations to fix at higher rates than neutral (which requires selection), deleterious mutations to be eliminated (which also requires selection), and coordinated changes that build functional systems (which requires selection). So using Kimura's neutral fixation is a bait and switch.
You can't have it both ways. If you invoke neutral fixation rates, you're describing random drift of inconsequential changes. If you want functionally significant evolution, you need selection, which means selection coefficients, fitness landscapes, and far more restrictive probability calculations that account for combinatorial constraints. The math in this response shows that neutral mutations can accumulate randomly, but it doesn't show they can build the coordinated, functional complexity evolution requires.
Icky writes: "If you invoke neutral fixation rates, you're describing random drift of inconsequential changes. If you want functionally significant evolution, you need selection, which means selection coefficients, fitness landscapes, and far more restrictive probability calculations that account for combinatorial constraints."
Dennis: Most mutations are neutral, a few beneficial. Day's mathematical argument was all about neutral fixations, and I showed why it was wrong, even though it has nothing to do with natural selection. Regarding beneficial mutations the math is much simpler--and *obviously* a beneficial mutation can and will proliferate in a population in severe circumstances, often reaching fixation. This is the point that everyone--including Creationists--agree is trivially true. Do you want me to provide the math for that?
I'm happy to look at the math, but while Day's post the other day may have been addressing the neutral fixations point, the overall argument(s) from PZ and TFG need to be taken in totality. For instance, unless the math addresses the population constraint issue, it is nothing more than (admittedly compelling) sci-fi story telling.
There is no bait and switch. Neutral alleles drift to fixation and beneficial alleles are selected while deleterious ones are eliminated. All of this happens simultaneously which is what the modern synthesis describes.
Just to be clear, the Modern Synthesis does not include Neutral Theory. Huxley and Mayr predated Kimura by 26 years. You're presumably referring to the ad hoc Post-Modern Synthesis, which isn't really a coherent theory per se, but rather two very distinct theories running in parallel.
To the best of my knowledge, no one has tried to synthesize the Modern Synthesis and the Neutral Theory in any mathematically serious way.
If 50 neutral alleles drift to fixation, how many of them are beneficial vs deleterious? For the math in the article to work, you have to assume all 50 are beneficial each time, every generation, simultaneously across the entire population. If there is assumption that "pops up out of an egg" analogy in play, it is the assumption that 50 beneficial alleles pop up miraculously in the ENTIRE POPULATION every generation with none of them being deleterious. If you want to claim that there is a net 50 positive out of some greater number of total alleles drifting each time (two steps forward, one step back type of situation) every generation, you need more than 50 shifts per generation. Which makes the math explode.
If they are beneficial, then they are not neutral by definition and they will fix faster than neutral alleles. Fixation does NOT require or imply that the fixes happen instantaneously and simultaneously across the population, I don't know where you're getting that from. And you DO have more than 50 mutations every generation. You have 50*N where N is the amount of new individuals.
E.coli have one single new mutation in one single individual every few hundred generations, it can literally be over 500 generations before a single new mutation happens in one offspring. Animals like humans have several dozen mutations, even up to a hundred, in every single individual that's born. There are literally orders and orders of magnitude more mutations per generation in humans than there in e.coli. The statement that "e.coli have the fastest fixation rate per generation ever observed" is a blatant lie and false.
Precisely, if they are --beneficial--, you cannot call upon Neutral Drift mathematics to make your case. Fixation requires that the beneficial mutation, spreads throughout (enough of) the ENTIRE population. How do you propose that happens across the ENTIRE population every single generation? The math on the 1/20,000 is assuming the SAME 50 FIXES every generation. That's how it becomes fixed, everyone now has it.
A neutral allele that fixes in a deme in some generation g did not start the previous generation, it started thousands of generations ago and has been drifting towards fixation for thousands and thousands of generations. And no one claims that all the differences between species are only due to neutral alleles, so that is a strawman.
Every single generation experiences neutral fixations that started thousands of years ago. The overwhelming majority of mutations are eliminated as the generations go on (and during this time there are tens of millions of new mutations happening every generation as well), and a small handful fix after ~4Ne generations. Every single generation has neutral fixations (at the rate of average neutral mutation) where those mutations that fix started thousands and thousands of generations ago.
Thus, the relevant number is "mutations per generation".
E.coli literally do not even have a single mutation per generation on average. Humans have 60*N mutations per generation on average, where N is the number of new births. The claim that e.coli has the "fastest fixation rate ever observed" is a straight-up blatant lie.
You are correct. Dennis is saying the same 50 fixations, not different ones, are occurring in every individual at the same time, every generation. No study has shown this happening.
People are comically expressing surprise and horror at the fact that
20 million/400,000 = 50.
Yes, 20 million fixed mutations in the human race acquired over 400,000 generations equals an average of 50 fixed mutations per generation. (Again, this is out of 400 billion mutations.) And it's not me who says it, but geneticists, evolutionary theorists, etc.
Your math isn’t a surprise,but your ability to link it to reality is. There’s no study, no model, that validates your average.
I’ve never understood why naturalists say that dog breeding isn’t an example of natural evolution. Humans are natural biological beings that operate according to natural biological and physical laws. The genomes are altered in micro evolutionary fashion according to those natural laws. It’s all natural. We’re told that you can’t include intelligence or design in any theory that explains the process in an attempt to differentiate it from natural evolution, so that’s what it must be.
Because "natural" is being contrasted with "artificial". An undirected random process, versus an intelligent directed process.
Dog breeding thus falls under "artificial evolution".
If everything is "natural evolution", then the adjective "natural" no longer carries any meaning and should be dropped. It's all just "evolution".
Dennis, you're still using opinions as if they are evidence.
No. All the fossils, the salamander ring, the results of the moth study, the DNA studies, the animal and plant distributions, taxonomy, etc., etc., etc.-- those are all facts and data. I just don't shut my eyes to them.