• njikomes

Razib Khan: Population Genetics, Personal Genomics, History & Human Evolution | #91

Updated: Oct 10


Full auto-generated transcript below. Beware of typos & mistranslations!

Nick Jikomes

Why don't you just kind of give everyone a quick intro of who you are and what your academic background and pedigree is?


Razib Khan 5:26

Sure. All right, so my name is received Khan. In terms of education, I have degrees in biochemistry and biology from University of Oregon. I am everything but dissertation that UC Davis, my straight evolutionary genomics work on cat and aqua domestication. So I was really interested in those topics. I've also done a fair amount of consulting and work in the personal genomic space, I did not Family Tree DNA, original algorithm and National Geographic, the late, great National Geographic test, I did the autosomal algorithm, the latest version for that. So and I've done other consulting, in the private sector space, I can't disclose everything. But a lot of my work has been filed genetics and file genomics. Right now, I'm actually co founder of a startup called generate. And this is more about the future, basically helping with big data and precision medicine, that sort of thing. Another thing I'm super obsessed with and interested in personal genomics, aside from evolutionary genomics, and archaea, genetics, and my older son is the first human who was sequenced alive before he was born. So that's in the history books. And, you know, I see a future where we're all gonna be doing that sort of thing, and also the animals and the plants and the whole world. So that's what I do during the day, focus on that, you know, basically solutions for companies in the genomic space to make this new discipline, really efficient and scale up. But the other thing that I do, which I think you probably know me, for more, is my writing my substack, my blogs, you know, I've been, quote unquote, communicating population genetics and evolutionary genetics and human genetics for a couple of decades. Now, in the last couple of years, I have a sub stack, it was the.substack.com, where I mostly talked about human evolutionary history, human genetic history, very specific. So for example, you know, the genetic history of the Ashkenazi Jews, something like that. And my, my pieces, my posts, they tend to be pretty long. Sometimes they're multi part, but you know, they're always at least 5000 words. And you know, each part will be about 5000 words. And it's usually a mix of history and genetics, some of the genetics is from the scientific literature. But since I have 10s, of 1000s, of samples, myself, and I can do a lot of analysis, I often try to do original analyses for the posts as well.


Nick Jikomes 7:51

Ya know, a lot of your content on substack is super interesting. So it sort of appeals to people who are history nerds and genetics nerds. And it's sort of a, I guess, a historical, historical stories that you tell about different populations of humans, different different time points in history, recent history, or fairly deep history, often through the lens of genetics, and it is fairly unique in its approach to telling us historical stories. How did you even was just born out of personal interest? You were just always interested in that type of subject.


Razib Khan 8:23

Yeah, I've always been interested in that type of subject. You know, a lot of people that I meet in science who thought about being history majors as well. And you know, we took a different route. I'm one of those people probably well, I mean, I don't know if I ever thought about being a history major. I didn't minor in history, but was always an interest of mine. But you know, I was a quote unquote, stem person from when I was a little kid. And I knew I was gonna do STEM related things when I grew up. But I also have this other interest that kind of like, developed when I was a teenager. And I read a lot of volley forces, book history, geography of human genes, the unabridged form, when I was an undergrad at University of Oregon, I read it front to back that was published, I think, in 96. And it was a summation of his life's work from like, the 1950s Onward. And it used about it was it used about like, maybe like 40 markers in the human genome, and he had assembled a data set of a couple of 1000 individuals. And, you know, he just analyzed it with principal component maps and principal component analysis and phylogenetic trees, and kind of explained how the populations relate to each other and how they relate to history. And I was pretty captivated, captivated by it. But, you know, I didn't think it related to anything that I would ever do, because, you know, I was majoring in Biochemistry at the time and, you know, in the 1990s, when I was a kid, it was really the age of molecular biology still, that was the sexy area of life science. And I just assumed if I wasn't doing biological related, it would have to do with that scale. You know, I have a give you a quote, a friend of mine told me in the 1990s, a colleague of his was, was thinking about going to population genetics. And, and, you know, people were saying, you know, what do you call a population geneticist, and the punchline was Emeritus, because it was seen to be, you know, like, look, that was something from like, decades ago. And now we have all these DNA, you know, Analysis Tools, and then genomics happened. And all of a sudden, population genetics became relevant again, because instead of a single sequence, you know, we're talking millions of sequences, you know, so I have in my computer 10s, of 1000s of human genotypes with like, you know, four or 500,000 markers. And I just analyze or whatever I want to,


Nick Jikomes 10:49

can you, can you, for people that don't know, and don't have a background in genetics, what's the difference between genetics as a field generally, and population genetics specifically?


Razib Khan 11:00

Yeah. I'll give you like a really, really precise way to understand population genetics, really. So genetics is the field has origins in Mendel's theories. You know, Mendel's, like laws of inheritance, Independent Assortment law of segregation, but you know, easiest way to think about is to think about a pedigree. How are characteristics transmitted through the pedigree, and that's how you have like, recessive, dominant, all of these other things, right? And, you know, individual contributes half of their genes. And you know, the individual gets half from each parent. And you know, there's one half related across. So there's this whole formal framework using this concept of genes that are ended up in discrete particulate inheritance model that are inherited, right. And so that was rediscovered around 1900 by a bunch of different people, and blossomed into this whole field. And they discover things like genetic recombination, which is actually a way to explore assortment and stuff like that linkage. Excuse me. So that's genetics. Simultaneous to that there's another field called biometry. And that comes out of, you know, original Darwinism. Charles Darwin's theory is kind of more formalism. You know, Karl Pearson, you know, it came out of a lot of Dalton's work. And it's a way to kind of like understand evolutionary change over time quantitatively right. And genetics mentalism and its biometry originally seemed to kind of be rivals. And then ra Fisher, basically ra Fisher, that one of the major founders of modern statistics, you know, you know, p values, all that sort of stuff. Actually, uh, no, but I think p value was moment anyway, aside from that, he's a frequentist. Right. And, but he was also an evolutionary geneticist. And he, he was trained in mathematics, but he was super interested in evolution his whole life. And so what he did is he showed how using the medalion particulate inheritance model, when you scale up with the number of genes that actually result in the same type of patterns that you would expect, from biometry. So the two are actually just different ends of the distribution. And what population genetics did was fuse the two and then kind of start out with this formal model. So Hardy Weinberg p squared plus two PQ plus q squared equals one, everyone knows that that's kind of like a topology. So it's like two variations of a gene at given frequencies. How will they combine in terms of the different genotype frequencies? Right? So this is like, Okay, you start out with these premises, a population genetics really is kind of the working out of these premises and understanding formal models of how gene frequencies change as a function of time, due to different evolutionary forces. So one of the forces is random genetic drift, we can people understand that just random and stochastic, boom, boom, boom, here there. Another the forces of natural selection, this is Darwinism, right? So natural selection changes gene frequencies by the gene is good for you, it'll increase the frequency up to 100%. Like kind of like sweeping up, right? If it's bad for you, it'll keep it out of the gene pool. And then you have other things like migration, and then of course, mutation. So mutation is a genetic process, you know, the term emergent genetics in the early 20th century. Today, we know what mutation is biophysical Lee because we understand DNA, which is the substrate of inheritance. So originally mental didn't know about DNA, right? He just saw the inherited patterns. So he understood how DNA manifested itself in genetic analysis, genetic, you know, and then later, in the first couple of decades, I think they understood there was little suspicions that DNA was the substrate of hairs considerably before Watson and Crick nail down its exact structure, right? But once you nail down the structure, okay, you know what it is you can target it, you can figure it out. And so things like meiosis, for example, which is a genetic understanding of how you take diploid genomes, and you you know, do recombination and then you turn them into haploid copies for your sex cells that's originally understood in genetics and classical genetics as a model. And now we have DNA and you understand actually bio physically how the strands of the DNA are snapping together and why they snap together where they snap together. And so everything kind of comes together. Today, in genetics, where you have molecular biology that understands the biophysical process, you have classical genetics that's focused on the inheritance. And then you have like, say, genomics and population genetics, that's more interested in aggregate numbers and frequencies and say, like, summary statistics, maybe.


Nick Jikomes 15:38

Okay, so So population genetics is really the branch of genetics, which is concerned with quantifying how genetic variation changes throughout populations of


Razib Khan 15:50

organisms. Yeah, that's it. Yeah, that's a perfect definition.


Nick Jikomes 15:53

And so you know, when you start to think through the lens of population genetics, and you think about how patterns are changing, and entire populations of organisms, you can start to understand how a given group got where it is today. And you can also make predictions about where it might go tomorrow, depending on the evolutionary forces that are acting upon it. And of course, one of the most interesting stories is the story of human evolution, which is the one that you focus primarily on. Yeah, I want to ask you about a general topic, but you have this one article that is super interesting. And at the top of the article, you've got two cartoon images, they're basically two different versions of the human evolutionary tree. One of them is our 2000, the year 2000 state of understanding of human evolution. And the other one is are your 2021 are present day understanding? Yeah, they look very different from each other. So can you describe why that is? And how much the story of human evolution has changed in the last 20 years? And really, what was driving that?


Razib Khan 16:50

Yeah, so a lot of why it is, is data, we have a lot of it. Look, I mean, you know, just for the listener, or the viewers, the current generation of biologists are, you know, not not the ones that are like 25. But let's say like, mature career biologists, they emerged in a world where they did data analysis with spreadsheets, you know, and like maybe like a four by 30 matrix, you know, and out today, we have terabytes of data in any given lab, you know, so the data is there to test hypotheses that required a certain statistical power. Now, part of the issue is, and I think I did get into the, in the pieces a little bit is a lot of population and statistical geneticists, were aware that that model could very well be some too simple, because we didn't have the power to test more complicated models with more subtle effects, if that makes sense. Because we didn't have the data. And if you can't test it, it's just a conjecture. But you know, science has a, excuse me, it has a bias towards parsimony. And the model in the year 2000, where you have, you know, humans, expanding out of Africa from one small tribe and replacing, so expanding in Africa, and then out of Africa, like one small tribe. So in the late 1980s, is a documentary of Nova, you can find it on YouTube. And there, you know, the whole thing is like mitochondrial Eve, and she was a member of one small tribe that conquered the world. And that's kind of the story that was told in the public. Now, some geneticists and some paleo politicians say, well, it was more complicated, privately. I'm like, whatever, like, you guys, we're not saying this to the regular people. So you obviously didn't have the courage of your convictions? You know, I don't know, you know what I'm saying? It's like, you can say like, oh, well, we always had concerns, but it's like, that's not what you're telling everybody else. So you know, that that's my attitude about it. Because sometimes we will say, well, there's, and there's papers that showed, you know, like, we can't draw these conclusions with a high degree of statistical certainty, because we only have like a little bit of data, right? We didn't have any ancient data. That's also like everything that we had. So if you have a phylogenetic tree, before you had ancient DNA, you're sampling the tips of the tree, and you're using the distributions on the tips of the tree to figure out the topology of the tree to figure out the structure of the tree. Well, that's okay. That's not That's not totally off. Like, for example, all of the new genomics has confirmed the very early work that humans are more closely with chimpanzees than either our two gorillas, gorillas or the group and then you have the Ranga tangs and then you have the Griffins, right. So that was all discovered in the 70. Well, it was pretty much assumed and believed, I think through morphological analysis of bones and stuff like that and like looking at our characteristics, but The 1970s, the earliest molecular genetic, molecular evolutionists figured it out through I think proteins and stuff like that, right. And now we have much better understanding of that relationship and all the details. But the broad outlines are still there. What we do though, is like you add tips of the branches, you add connections. So you know, there is work that there might have been a period of chimp human hybridization, four to 6 million years ago, that doesn't change the broader story. But it adds a twist, you know, like a twist to the tree in the phylogeny, quite literal sense. Like it splits goes back up. And splits is what the model is. And that model comes from looking at Whole genome sequences of all 3 billion base pairs, which obviously, we didn't have to do 30 years ago,


Nick Jikomes 20:43

I see so So basically, if you go back to approximately the year 2000, the basic model that anthropologists and biologists had for how humans became modern humans spread throughout the world was, we're all descended from a single population that came somewhere out of East Africa. It then spread throughout the world and replaced all other populations of humans. And there were other what you're saying is there were other models that were more sophisticated and more complex, but around the year 2000, we simply didn't have all of the data, including ancient DNA from extinct humans and extinct other species that allowed us to test whether or not anything beyond that simple out of Africa model was true. But today we do. And that's why the story has changed.


Razib Khan 21:28

Yeah, we got way more data. Yeah.


Nick Jikomes 21:31

And so what are some of the most salient changes to that story? You know, one of them that I'll direct you to is just, you know, that there was always this sort of controversy, I think, between the world called the strict out of Africa replacement model of how humans became modern humans, versus something called the multi regional story. So what were those two ideas? And how did they sort the balance between them shift as we got more of this data? Yeah.


Razib Khan 21:55

So that out of Africa is really straightforward. So you start out with a single tribe, and everybody has descended 100% from that tribe. That means we were placed the underdogs. We placed all the other humans in the world it was, you know, it was known, but along with no, there must have been other hominins that I think determines hominins. I'll just say humans, okay. I think everything that's almost human, other types of humans in East Asia, because we found their artifacts, right? We know that there are people there. But it looks like everyone, Chinese Europeans, you know, points on your Aruba. Everyone's descended from these people that are probably African because Africa has most of the variation. And you know, when you look at distributions of trees, if everyone's nested within African populations, well, you know, it's not impossible that we're not African, but it's probably likely that everyone, if everyone looks genetically a subset of Africans, we're probably from Africans, we're probably African, right? I mean, it's, you know, it's just like, the statistical probabilities are so high. So that's what they saw. So, I mean, it's a part of the parsimonious model. And, you know, we saw it like 50 60,000 years ago, you know, new types of archaeological just like toolkits and stuff like that. And it seems like you know, something swept across the world well, that's something was like, you know, superhuman, Homo sapiens, sapiens, blah, blah, blah, all that stuff, right. There was a good model. And also we have, we have ideas, you know, from the monotheistic religions of the Garden of Eden and people expanding out it pretty well. You'd mitochondrial Eve and y chromosomal Adam like these, these words. So everything was kind of like rigged to make that plausible, right? The other idea is multi regionalism. Multi regionalism is really centered on University of Michigan. Milford, Wolfe, Mumford Wolff Wolkoff is a big motor that and multiregional isms like that, actually, like humans, over the last 2 million years have kind of evolved together across the world through gene flow and sweeps. And it's a much more complicated dynamic. And some of the features you see in modern Europeans and, and Australia's might be due to admixture with local populations in Eurasia, that lived there before the expansion of say, the dominant wave. And so you know, there are some forms of multiregional ism that are very balanced, where it's like, oh, well, like, you know, modern Europeans are like 70%, Paleo European and 30%, African and Africans are the reverse. There's always gene flow involved, right. But some of them are balanced. And then some of them admit that Oh, actually, there was a dominant African population, but there's a lot of ancestry from local populations in the model, so you know, everybody in Europe is 15%, Paleo European, everyone, Asia is 15%, paleo, so that creates the distinctiveness of the different human races, you know, in terms in some sense, they think that that some of the distinctive aspects of different human populations come from these deeply rooted differences that go back from Millions of years in the region. And so that's one of the arguments from Altai regionalism. And that's really based on fossils. It looked like the GNP, basically, the genetic analysis of mitochondrial DNA, which is passed maternally show that that couldn't be right. But there's actually statistical reasons why people oversold that, and a lot of geneticists actually did privately tell them multiregional was like, Look, we haven't totally really refuted your model. You know, we just made it maybe less likely, but we need more data. But you know, they, they will tell people that and people will just be like, okay, whatever, you're just racist. So that was, that was a serious problem. multiracialism had a era of racism, because they're saying that different human populations have deep differences, in some sense from the past, right? Whereas the out of Africa seemed like, oh, well, we're all is like, you know, Richard Dawkins used to wear like, we are all African, and we're all like, really young species, and we're all from the same population. The way that monta regionalist would respond is like, you're saying that we're basically genocidal like we exterminated everybody else like your your idea of human nature is really bad. So I think that there got to be some bad feeling. Some of the bad feeling even Chris Stringer, who was like the primary out of Africa guy in the Paleo space, and Wolpaw, who I'm friendly with both of them walk off in particular, still has bad feelings. He's told me he's told me publicly so I can repeat it on my podcast before years ago, with stringer about like, you know what happened. And I think a lot of it had to do with the sliming and the negativity directed towards the multiregional people as racist. You know, I think all of that stuff is kind of ridiculous. But whatever. What happened?


Nick Jikomes 26:38

Yes, I mean, setting setting sort of the the professional squabbles and debates aside, that got us to where we are today in our understanding of human evolution. When you talk about something like multi regionalism, or the idea that there was Gene flow between distinct populations of humans or even different species of humans that once lived, and modern humans. What exactly can you give some examples? So example number one that most people talk about is humans and Neanderthals. But what are some of the examples where we know there was definitely gene flow between different populations in different parts of the world?


Razib Khan 27:15

You mean, really ancient? Yeah. And the aerosols are ones with the NFL has diverged from our own lineage, depending on what molecular clock and estimator used probably 500 to 750,000 years ago. The deepest divergence between modern human lineages is the Khoisan, Bushman, and other humans at most 300,000 years ago, but really probably 150. I'm being a little cagey about it, because it depends on your parameter estimates and some models, right. But whatever. It's considerably more than between any of the modern human lineages, right. Denisovans are another population in Asia, there are multiple Denisovan populations. And with Neanderthals, it looks like they're genetically homogenous from the all tied to Spain. And it looks like they went almost extinct multiple times and came back from small founder populations. I think we understand why this is like look at the ecology of Northern Eurasia during the Ice Age. It's very, it's very precarious, you know, in contrast, Denisovans, they're named after green. So we cave in the Altai, in Mongolia. We're also Neanderthals lived and before I forget, geneticists, geneticists, have discovered a girl who's half Denisovan, half and Athol who's literally a first generation cross. So that happened. So it's not just modern humans, you know. So these events live in East Asia. They lived in Mongolia, they probably lived in Southeast Asia 5% of the ancestry. So Neanderthal ancestry is like 1.5 to 2.5%. Outside of Africa, it's much lower within Africa, it's probably mostly back to Africa migration, right. Europeans are like closer to 1.5. Middle Easterners in particular, and then East Asians are 2.5. In Papua New Guinea, you have populations that are about 5% Denisovan. Australian Aborigines, if they were pure, quote, unquote, are probably also 5%. But there's very few full blooded Australian aborigines left, right. There is a population in the Philippines called Philippine Negritos. And they're about like, 20. They're probably like, they're about like 60%. Like other Filipinos like Austronesian farmers are 40% of their ancestry is you know, the says decreto like little negro. It's like they're dark skinned, curly haired. They're clearly an indigenous population that lived there earlier. And these Philippine decretals have like a couple of percent, or like two to 3% Denisovan. So if you do the math, it looks like before they were mixed with the Austronesian. They were the most of these certain population of the world, more than 5%, right? The last few 1000 years, they've kind of been mixed in and so that's produced. These Denisovan populations are not like the Neanderthals. They're not homogenous, some of them are deeply separated. Some of them are separated. Did like 100 200 300,000 years ago. Right? So like I said, they're more distinct than modern human races, modern human populations.


Nick Jikomes 30:08

So in the past several 100,000 years, there were points in time when you had modern humans and other, I guess we would call them subspecies of humans that live in different parts of the globe. Neanderthals are one example. Denisovans are another example. And they're distinct. You can tell them apart genetically, and yet they were still capable of reproducing with one another.


Razib Khan 30:31

Yeah, no question. Actually, there's, But wait, there's more. So actually, there's Neanderthal genomes from the Altai. And also like, there's some evidence from y and mtDNA. It looks like so you know, the word modern humans a little fraught, like, I mean, I like to say like Neo, you know, Neo African. Another thing that we have in common is stem human, because it's the stem lineage to the modern populations, right. But in any case, I'll just say new African because it makes it easier. So there's no African populations, it looks like our the primary wave that led to everybody outside of Africa, me and you, just judging by your look, you know, left about like 60 65,000 years ago, 60,000 years ago, right. But it looks like there are Neanderthal samples from like 100 120,000 years ago in the Altai, that have something that's much closer to modern humans to Neo Africans than anything else. So that means to me that, that there were earlier migrations that were absorbed into the 100,000, probably Denisovans. Okay, they're not like really high fractions like, so we're like, you know, 2% or whatever Neanderthal? Well, those Altai Neanderthals were 2%, Neo African. So they themselves mixed in with the second wave of kneel, Africans had a little bit of modern human ancestry. And there's natural selection for modern some, some of this Neo African genomes, like genes had some useful things, and there's evidence of selection within the animals. So some of the multiregional arguments of these ancient populations exhibiting gene flow and influencing each other does seem to be true to me, where the multiregional argument is weak is the fraction of this Eurasian hominin, Eurasian humans are pretty low, everywhere. So I didn't mention to these one ancestry is actually present everywhere, pretty much detectable everywhere east of the Indus River. And so but it's a really low fraction. So you're talking about statistical power, you know, we have whole genomes now, with 1000s 10s of 1000s. Millions of genomes, you know, we can compare. So now we can detect it like Chinese people, Han Chinese are like point one 5% to the needs of people in mainland Southeast Asia are like point two 5% people in India like point 2%, point one 5%. And we even know these are low percentages, like less than 1%. But we even know that they're different Denisovan populations. So the reason we're population that's discovered and the only full genome that we have only full genomes, we have our for Mongolians population in the Altai. And researchers can tell that the Chinese have contribution definitely from that group. Papo people in Papa New Guinea have contribution from a different Denisovan group that's pretty distinct, which will make sense look, look at where Papua New Guinea is, right? So there's a Southeast Asian Denisovan group. And there's even research I mean, I'm like going on because it's complicated. But there's even research that people in pop will have like to two or three pulses of Denisovan admixture which means that they might have moved through territory in Southeast Asia before getting to Papa, where they mixed with a nice one group one that they miss Lisa and group three that are Denise one, group two and then three. Right. So this is it might be interesting might not be interesting. It depends on who you are. We have the statistical power, we have the technology we can now know and understand. Right?


Nick Jikomes 33:53

So just to make it really clear for people. So who were the Denisovans? And when did they live? In particular? You know, where on the globe? What did you find Denisovan populations? And what what did they look like compared to our direct ancestors?


Razib Khan 34:07

Yeah, so Denisovans are if if we are the cousins of Neanderthals, Denisovans are the brothers or sisters. So it looks like they come from the same. So there was an earlier out of Africa expansion is what it looks like. And it looks like Neanderthals and denisovans are part of that earlier out of Africa. I think homo Heidelberg that we're against this is the one but in any case, and so they separate, you know, pretty soon afterwards after that, so I told you 500 to 750,000. I think there used to be younger dates of three to 400,000. But I think that's out of date. So 500 to 750,000 is when our lineage the stem African Neo Africans, modern humans separates from the Neander Solons. And then, you know, 50 or 100,000 years after that, that Neander Sylvans differentiate. So if Neanderthals are in western Eurasia. So basically Neanderthals, they looks like they probably got as far south as Northern Israel sometimes when the climate was cold, you know, but in any case, they're always Western Mongolia, west of China, and looks like like, like they're in Iran, probably Afghanistan, but probably not India, it looks like Denisovans occupied Asia, east of the invaders like East Asia and into Southeast Asia. Right. So Denisovans are basically East Eurasian hominins, East Eurasian humans. And unlike the animals are very, very distinct multiple populations, I think, probably because Neanderthals didn't have ecological depth. So when the ice expanded, and there was a cold snap, even during the middle of the Pleistocene, they had, they had nowhere to go, like where they had to go to the Mediterranean, nowhere to go, right. So it was against the sea, whereas in the end, the Denisovan populations, you know, they had Southeast Asia, they had India that close, like, give Taiwan Japan was like, much bigger and connected to the mainland at that period. So they have large populations are quite diverse, before their absorption into modern humans.


Nick Jikomes 36:08

And, you know, another thing that's really interesting about how the story of human evolution has changed over the past couple of decades is the story of the Neanderthals themselves. You know, my understanding is, you know, 20 plus years ago, most people including Most anthropologists, basically thought of Neanderthals as you know, stupid cavemen that weren't, we're not nearly as sophisticated as early modern humans in terms of their cognitive abilities, their culture, their capacity for art and language and things like this. How has that aspect of what we understand about Neanderthals changed in the past couple of decades?


Razib Khan 36:42

Yeah, I mean, I think you've correctly correctly described the general shift. Basically, back when we didn't think the animals were our ancestors, their animals. Then, in 2010, discovered that, like everybody outside of Africa has Neanderthal ancestry, for sure. And now we know everyone inside of Africa pretty much does. It's just much lower levels. All of a sudden, we have a strange new respect, you know, and things that are automatically identified as modern human artifacts. They're revisiting, like, oh, actually like this probably Neanderthal artifacts. So I think Neanderthals have been, quote, humanized, I think they always work human. They haven't humanized and rehabilitated. And, yeah, you know, there's a paleoanthropology apologist at Stanford, Richard Klein, who wrote a book Dawn of, like, dawn of human culture. And it was like basically saying that, like, he basically was wondering whether Neanderthals can even have language. You know, I mean, so pretty intense. You know, there were like theories of Neanderthals, like they were like, as far as gorillas and all this stuff. And so yeah, we've gone a long way, partly just like it's human bias, where we don't want to think our ancestors were animals and be shields. So now we are looking really closely. Whereas before we just were like, you know, they're not us. So, you know, they can do jack, and that's why they're not around and we're around. So that was the theory. You know, there probably are still strong differences. And genetically Neanderthal, the Neanderthal admixture we have did seem to get selected against in the genome, it cost some incompatibilities, you know, so it wasn't a perfect fit, we diverge some. And I think we were quite different, probably, but, you know, not as different as we, as we would think. And like, look, I mean, we have sex?


Nick Jikomes 38:36

Well, I mean, what do we know that Neanderthals were capable of, in terms of, you know, anything that we would describe as more sophisticated behavior in terms of things like culture? Did they produce cave art? Do we, you


Razib Khan 38:47

know, alluding to as the result in Spain, where I think they assumed that it had to be moderns, and they had better dating. It looks like kind of like abstract representational art in Spain, is what it looks like. And I you know, they did so it does look like they did cave art, I think they found some, not ivory, but maybe some stone figures that might have been carved by Neanderthals. Like I'm not a paleoanthropologist. So I don't know this off top my head, but they are discovering a lot of things a lot of it has to do with rechecking the dates on things because it had just been assumed that this could only be modern human. And, you know, Neanderthals themselves seem to have. So they're I think they're, they're called archaeological toolkits called the most Luciferian. But by the end, they created it looks like they created a different a different archaeological toolkit with the influence and synthesis with the new humans coming from the south and the east, the chattel Curonian now there's arguments whether Neanderthals did or not, I think they probably did. So it looks like Neanderthals are dynamic and adaptive. We're in the face of the newcomers even though they were originally absorbed and there's good I mean, they were they were out competed, right? A lot of modern human groups are also out competed by other groups. It's not a big shock. So I think if you're if you, I think maybe, let me just ask like, are you asking whether we were superior to them or something? I don't know.


Nick Jikomes 40:13

No, I'm just asking sort of what's what's the state of knowledge in terms of what like Neanderthals were capable of cognitively? Like, did they have the major elements of culture that we normally associate with? higher cognitive function? You know, things like language, things like art, things like commerce and trade?


Razib Khan 40:29

Okay, so the art thing is a little weird, because like that depend on like, what archaeologists can find and how they datum it looks like they might have had some art trade, I think that they haven't, I think it's been found that Neanderthals didn't seem to go very far for their Flint and other things and learning to check them out. But I think like Matt Ridley just talked about how it's quite clear that the new modern humans have much bigger trade networks, okay. Um, in terms of language, I'm pretty sure that something like language, just because it seems like from what I've looked at, like the genetics of language and stuff like that, but this is not like a super new characteristic, it could be like basil to our group, like, it could be like a common characteristic pretty deep in human. Now their language might not have been, maybe it wasn't as complicated or it wasn't like eat as, as easy to use it for abstraction, I don't know. And that could have been genetic, their brains do look somewhat different, their shape somewhat different. And there's some genes related to brains or some papers out about it, we're, we're different than the Andritz dolls, I'm just very cautious about the idea that we can just look at differences and figure it out that way. You know, because you modern humans are also different. My brain is different than yours, you know. So we need to be careful about that. But I think I think if they were alive today, they would be different. But I also think no one would think about doing an experiment on them. I think they would be given human rights. But they are different. They would, I suspect.


Nick Jikomes 41:51

So we would, in your view, we would likely notice differences between them and the people walking around today. But we would still recognize them as humans basically.


Razib Khan 42:03

Yeah. And like so for example, it could be there could be something weird, like people have talked about, well, they lived in this, like, you know, kind of like open scrub land, like south of the ice sheets. As big game hunters. Maybe they had really good visual spatial fields. So maybe it was like, maybe you can like, maybe you can, you know, try to have a Neanderthal write a poem, and like, you ever felt gotten like no skills at that. But then you give them a Rubik's cube. And they're like, you know, I just wouldn't be surprised, because they had big brains like it's well known fact, their brains are bigger than most modern. Well, actually, like, I think Neanderthals and Ice Age humans, like 20, or 30,000 years ago in northern Eurasia had the biggest brains. And it seems like cold climates tend to produce bigger brains might be something metabolic. I don't know. But everyone knows Neanderthals had big brains. So what do they do with that brain? You know, maybe they just did different things with that brain. In terms of maybe it was like more, you know, visual spatial, like memory for the landscape and stuff like that. Right. So


Nick Jikomes 43:03

one of the key pieces that has informed how our knowledge has grown and change with respect to human evolution, the last 20 years, you mentioned this earlier has to do with technology, and in particular, our ability to obtain genome sequences from long dead individuals to go and actually take bits of Neanderthal bone or Denisovan bone, extract DNA from it, and look at the genome and sequence the genome and compare it to modern genomes and other ancient genomes. Can you give people like a basic sense for like, how that technology developed? And what the specific some of the specific major innovations were like? What exactly is going on there that allows us to do this ancient genomic work?


Razib Khan 43:41

Yeah. So high level because it's not my thing. But I basically, the initial time, the initial wave of doing this work was the 1990s. And it was all false positives. It was all it was all contamination. Right. So SWAT de Pavo Neanderthal man, you guys probably heard the book. You know, he did a lot of early publications that were probably all contamination. So that's one thing, right? So, you know, they have ancient DNA rooms are, you know, like, from what I've seen, there's, they're like, you know, Nuclear Laboratories, or, you know, people who work with HIV or maybe anthrax. They have to keep it really clean. Okay, that's one thing contamination. Now, the other thing is, when you extract DNA, when you extract sequences from ancient DNA, most of its crap. Most of us either, like a lot of times a lot of expect mostly bacteria, right? So you got to like pull the bacteria out. Well, we now have computational algorithms that can snap out all the bacterial reads. So when you do a sequencing, you generate reads like these strings, and, you know, modern computational algorithms, computers can just like slurp out all of the non human. Okay, so you could do something like, Okay, anything that's not ape pull it out. Right, because because it turns out there are no apes in Europe at that time, probably besides Neanderthals, right. So with htma is probably Neanderthal. Okay? Once you look at those sequences that are human that's less left, you can amplify a PCR amplification, all these things that those tools are available, but there is a pattern of DNA damage. Like in his DM amination, there's a pattern of DNA damage that they that we now know empirically occurs, as DNA gets older and older. And you can filter those, or you could try to infer, like, you know, what the undamaged ones like, what they changed from what they mutated over from, right. So we have a lot of learning in that area. I think the final thing I want to say, and like this is just really quickly, the final thing I want to say, because it's much deeper than this, obviously, like there's a whole area of forensic genomics actually, like my friend, David middleman is company author of, it's been like they've been in New York time, like dozens of times now doing cold cases. And it's they use the same methods, you know, and this is for cases of like humans, like 50 years ago and stuff. So it's much easier in some ways. But although like, you know, the police labs don't always have the best preservation methods. But any case, they found that there's a bone in your inner ear. So usually, you know, they were looking for bone, but then they were looking inside teeth, like, where's DNA going to preserve, there's a bone in your inner ear that's really sealed, and is extremely rich, and DNA. And so, you know, if they find that bone, it's much easier, they don't find that bone, they can get it from elsewhere, but the DNA yield is going to be as much lower probably, I think t does another thing they do. Right? So yeah, that's what they've learned a lot of things, but they didn't know anything. But that was just a random find out like, Oh, my God, this, this inner ear bone has all this DNA. So now they always look for that, right? So. So I think it's a combination of the molecular biology is getting better and better over time, computational techniques allow you to extract signal from the noise. And then finally, you have the issue with discovering this bone, like discovering which areas of ancient fossils have DNA, you know, particularly for humans, but not just for humans, because it's mammalian, ancient genome.


Nick Jikomes 47:15

And so as the technology improved, and we learned some of these tricks, and we got more of this data in a usable format, so we can compare genomes from all sorts of people, you know, how modern humans alive today Neandertals, our ancestors, you know, going back 1000s and 1000s of years. What, you know, to what extent is genetic evolution, still still driving human change today? Have we have we gone beyond the? No, no? Okay, okay.


Razib Khan 47:50

Yeah, so a lot of it is. Alright, so a lot of that is a lot of this has to do with the fact that we didn't have statistical power to detect recent evolution. So there were a lot of techniques developed with modern DNA that basically stopped like, it couldn't detect anything, like less than 3000 years ago, something like that. 2000 Right. And so, you know, of course, we're not going to just like selection sweeps that are recent, because we can now we can partly through like big genomic analyses, ancient DNA, etc, etc. So basically, like, you look at the whole genome, you look at lots of whole genomes, okay? Like you increase the sample size that way. And then you get temporal transect, you get ancient genomes. So then you can like see directly. So for example, you know, lactose tolerance, that gene that's common in Northern Europe in particular. You know, we now know that selection for that has been ongoing the last couple of 1000 years. Nobody knows why, but it has, it started really late. It started two to 3000 years ago, it was way later than you think, you know,


Nick Jikomes 49:00

that's so so that's the the gene that allows us to digest milk past childhood years,


Razib Khan 49:07

lactose, lactose, milk, sugar, right? So yes, as an adult, so that's a mutation. And it's prevalent low frequencies throughout like central Eurasia for like 1000s of years. And then it just started shooting up in frequency over the last couple of 1000 years. So that's a recent thing. Also, Europeans seem to view Northern Europeans seem to be getting lighter skinned over the last couple 1000 years. There's been multiple ups and downs in the height across rural populations, like, you know, populations adopt agriculture, they get small, they discover like, you know, agro pastoralism, they get big so weird things like that happen. Their selection, their selection, like in Bangladesh, where my family's from, there's actually selection, I guess, cholera. There's like genes that give you resistance to cholera. You understand why that would be right. But cholera is like hasn't been around forever. Color doesn't exist, unless you have like, pretty large Vil. judges with bad water like hunter gatherers would never really get cholera. I see that's definitely recent. So there's a lot of things related to the immunological function and stuff like that, that have been detected. I think, you know, to get like spicy, but uh, you know, I think they will probably discover, you know, personality and bio behavioral things, too. I think that I think a lot of hunter gatherers probably went insane in the first villages being around that many people over and over. So


Nick Jikomes 50:31

what why do you think that just because it was so different of lifestyle compared to why me?


Razib Khan 50:36

I don't know, are you introverted or extroverted?


Nick Jikomes 50:38

I'm definitely more introverted.


Razib Khan 50:40

Yeah. So introverted. I'm not, I'm not introverted. introverted. People tell me that, like, you know, they're in a room with a lot of people, and they just need time to recharge afterwards. You know, I like that, right? If you're a hunter gatherer, you might be like, really a lot like that. You know, because also, like you're living in a same small group your whole life, most of the people you encounter who are strangers, you're going to be kind of skeptical of, because you probably don't need to interact with other people, except for maybe like some religious ceremonies made exchange, and a little bit of trade. But aside from that, it's like, you know, it's what you'd call autarky. Right? Like, what does that mean? That just means like, you produce all of your economic goods and goods and services and necessities yourself within your community. You know, you're a hunter gatherer, right? You don't really that need that much from outside. And so why would you want to interact with other people, they're just dangerous, then you live in a village, and that's all different. All of a sudden, there starts to be specialization, you know, you need to have a big man, some people need to be artisans. And then also, you know, in a ancestral, like, clan, it's like, relatively small, everyone can keep track of everybody else really started getting big. You just started having rules that everyone agrees on. Yeah, have rules. Yes, talk about it. Yeah, talk it out. You gotta have priests maybe. And then also, you're a dude. And you're going out to the fields, and your wife is in the village. Now, there's all these other dudes around now you start to get sketch, you know, like, all sorts of things start crop cropping up. So a lot of these you mean cultural adaptations, but I would also wouldn't be surprised if social, like, for example, you know, this is like explicitly been hypothesized the psychopathic personality, which is very amoral. And they lie a lot. You know, a lot of con men are psychopaths, right? There's an argument that that personality could not have existed before large villages. Because you just


Nick Jikomes 52:33

because everyone knew everyone, and you'd be able to everyone would know that person and be able to


Razib Khan 52:37

do a psychopath. Honestly, they'll probably kill you, when you're 10 or something like that. They just they will start recognizing this person who is off. And because you can't run away to another tribe, probably, they would probably kill you like your stranger. Right? So psychopathic, with these sorts of like social, like pathological personalities probably would not been able to exist in in like, you know, a lot of these hunter gatherer societies. But once you have cities especially, that's perfect environment, like you can screw someone over, and then go to another neighborhood.


Nick Jikomes 53:10

So basically, what you're what you're saying is simply that as we transitioned from being small bands of hunter gatherers engaged in foraging behavior, where we lived with probably dozens, at most, probably 100, or two individuals in the tribe, and we became sedentary and villages and cities and civilizations grew, there was way more people around. And so that meant two things. A certain new personality types may have become adaptive, that weren't adaptive in the hunter gatherer foraging bands. And be there was just sort of a relaxation on the constraints that were present in the hunter gatherer foraging bands. Yeah,


Razib Khan 53:47

yeah. Yeah. Yeah. So I mean, it's just it's a new environment, right?


Nick Jikomes 53:52

Yeah. Yeah. I mean, one of the things that people always talk about when we talk about human evolution, and how humans sort of spread and took over the globe, and developed civilization, everything is of course, you know, our brains, we've got big brains. We're a really smart animal, we're capable of doing things like language and symbolic cognition. And we have, you know, we have a very rich set of cultural practices that just allow us to do things that most animals mostly can't do. And so but when we talk about the subject of intelligence itself, in the contribution of culture, and genetics, and things like that there, it's a fascinating subject. It's also a very controversial subject. So what do we know in the context of the story of human evolution and us, you know, expanding throughout the globe? What, what was the role of intelligence in that? How much did genetic factors influence that? What does Intelligence even mean? And how do we think about, you know, the types of cognition that humans have that really made the difference that enabled us to populate the entire globe? Yeah, so


Razib Khan 54:54

there's a couple of things like we can go over. Okay. So I So first thing is the human we from skulls, we know for a fact that our brains have been getting bigger, over like over a million years, million a half years up until about 200,000 years ago. And then it kind of petered off. In terms of like, we're not really that much. You know, like I said Neanderthals, and some, some modern humans, like 30 40,000 years ago, had the biggest brains of all right? So in terms of brain size itself, you know, we kept getting bigger and bigger to the point where, today, what is it like 25% of your calories are absorbed by your brain. Okay, at resting. That's a lot. Like we have massive brains for mammals. I'm sure some of your listeners and readers and viewers have seen the encephalization chart, which shows like for our body size, our brains are huge, right? Like, we're like, maybe we're not as extreme as a shrew. But we're like the mega mammal version of a shrew, you know? So, so that's the thing. And that's expensive, and you don't do anything expensive unless it's fit. And there's all sorts of hypotheses of why our brains got big, like, one of the hypotheses you probably heard about is it's gossip and socialization. You know, we have to keep all of these human relationships in our brain, etc, etc. But obviously, we're not stupid. It's we were doing, you know, with our tools, I mean, the gossip theory is fine. But like, we have some pretty intense tools, right? Like, we had some like, good things going on. And so I think we had general abilities, right. So there's a differentiation between specific competencies, and general abilities. So core competency would be like language is like a fluency that everyone has everyone can learn language in a way that a chimpanzee cannot. So let's say that's like, that's like, a specialized competency, right? It's a it's a faculty. It's a specialization. It's like we have an Oregon, a language, Oregon, in the brain, right. But we also have domain general intelligence. So domain general intelligence is basically our brain can adapt to new, new tasks that, you know, like, for example, I have a friend mathematical area of study is topology, I don't get it. None of his ancestors, new typology, and totally novel cave people. Right. But his general intelligence, and he figured it out. And so that's, I think, when you talk about intelligence, IQ, that's, that's that's that, right? Which is different from our other competencies, like language cetera, you know, so, you know, someone can have a PhD from Princeton, and someone could be, you know, get their GED. And there's big IQ difference, but they're both have core competency of language, in contrast, and someone maybe had more extreme form of Down syndrome or something like that, right. So we call this different ways, let's move on to like IQ, unlike general intelligence, because I think maybe what you're ultimately getting at, and that trait has a normal distribution of the population, as we know, normal distributions are generally associated with traits, whose selection whose benefit is not clear. For the population over a long period of time, otherwise, the variation would disappear as selection kind of just like, shifted everything to one direction, right? So in the pre, in the ancient past, you know, millions of years ago, there might have been some humans who are semi verbal, some that were not verbal, some that were verbal, and selection made it so that we're all verbal, like all the variation, and verbosity is basically gone, unless you have a disease when it comes to IQ. That's not like that. And that tells me that it's not always helpful. It's not always necessary. And, you know, we see in like, modern modern populations were extremely bright people, you know, they often have lower fertility for various reasons, you know, some of it could be structural, you know, like, if you seem kind of weird. You don't have a lot of people that you can like, potentially date.


You know, maybe that's not good to be that smart. Like, what are you going to, you know, like in a pre modern environment, what are you going to do with a really high IQ? You know, what I'm saying? Like, are you going to go be working at a quantitative hedge fund? No, you're not. You're probably another hunter, or gatherer. Like maybe you're a little better. But maybe you're also like, super weird now, right? So I think we're smart enough, or smart enough to create civilization. But we haven't been selected to be super super, at least like you know, by the lights that we measure it super smart or super dumb. You know, people with average intelligence do quite well actually, when it comes to reproduction from what I've seen, you know,


Nick Jikomes 59:57

so So you said a couple of things you you're talking about? intelligence and you're talking about IQ as a measure, or in this case, the measure of intelligence. So I want to talk about what how that measure actually works and what it's actually measuring. And B, you mentioned that this is a trait that's normally distributed, which means, which is associated with being something that's not clearly selected for one direction or the other. So I don't know if you can sort of compare that to something like height, like I just normally distributed. Yeah, is that maybe a simpler way we can get people thinking about that type of thing.


Razib Khan 1:00:27

So height is about like 10 times simpler than IQ in terms of the number of genes, genetic variation that affects it, but very similar. So is basically like a lot of genes are affecting it. And there's, you know, normal distribution occurs to just like, the random effects of all these genes, a central limit theorem, and you just get this like humps distribution. Now, you know, some of the nerdy listeners out there know that, like, you know, it can be normalized and standardized, because unlike height, IQ, something like you're measuring, and it's a test, that's kind of an artificial construct, right? So it's not. So when you do a normal distribution for height, aside from like pathologies, I think it's, I think it's a little fat tailed. But in any case, it's pretty much a Gaussian, right? But when you do like a paper and pencil test, obviously, you know, that test has been designed to come out as normal, just to be clear, just to give a good range, because you can give a test like basic arithmetic, and you're not going to get any distribution, aside from people that have like serious, you know, deficiencies, right. So these tests are designed to be hard enough to get the normal distribution, but not so hard that they're like Poisson distributed, where like, very few people get the answer, right, you know, so just to be clear, and in terms of selection, when you have variation in a population, it could mean a couple of things. It could be like mutation, and random genetic drift, as mutation random, are working together to maintain an equitable amount of variations of mutation as variation, random genetic drift removes variation, and you have mutation drift valance, and you have this normal distribution, that's always kind of metastable. Another thing is you can have diversifying selection. So you can have selection that's maintaining the diversity in some way. So that it could be that, you know, being smart is good, until you get too smart. And then it kind of bumps you down. And you know, and also it could be a situation where over time, things can change. So for example, this is not intelligence, but like, go back to the psycho psycho psychopathy example, it could be being a psychopath, initially, is actually a very good strategy for your reproduction. So imagine you just screw people over. And you just like leave children in your wake, you know, so your reproductive fitness as high psychopathy spreads. What happens in a society with too many psychopaths is like one, they start to develop mechanisms to detect them, and do away with them right jails. The other thing is also if society doesn't develop those mechanisms, the society will collapse, because psychopaths cannot exist without non psychopaths. They're a parasitic strategy, right? So your frequency dependence or psychopath, fitness is really high when it's low, and then it decreases as the frequency goes high. So that's negative frequency dependent selection. And so that's a common way of thinking how, you know, maybe selection is maintaining genetic variations, which are good. But once they get to high frequency, evolution pulls them back. So it's like, kind of like a bounce back and forth creates this, like normal distribution is another way to think of it. I can make it more concrete with height.


Nick Jikomes 1:03:36

Yeah, I can do that in the comment. I'm like, is this a common type of evolutionary phenomenon?


Razib Khan 1:03:42

Yeah, it's so high. It we know from most evolutionary psychology, and like modern sociological analysis, like being a tall dude, in particular is reproductively very, very beneficial. You know, like, you know, I don't know, like, if you're a five A, got that short, but like, you know, if you have a brother who is six, one, you're controlling for a lot of factors, you know, that they date much more easily. Just concretely I like, I'm sure, like, people have seen the dating studies, like women like they, you know, like, there was a big controversy where they let women keep the, it's mostly women who do the filtering, but men do sometimes too, but they let them keep the height filter, and that automatically means like, no guy under 510 is, you know, being seen by most women, right? Okay, so what what, why aren't we all taller? Why don't we all join us? Well, it turns out, there's a couple of things. One, once you get above six, two, or six, three, as a man, as a human being, I think in general, your body starts breaking down, so your mortality starts increasing. There's all sorts of weird things that only super tall people know about. So I have a super tall friend, and he was taking a shower when he got a tear in his lung. And his doctor had always warned them that could happen. He's six, seven, right? It's just something super tall. People know


Nick Jikomes 1:04:55

what is what is torn lung have to do with being tall.


Razib Khan 1:04:59

So apart Only the lung does not scale, volume wise, as you get bigger and bigger. Like, it's just like the anatomy and the physiology just start breaking.


Nick Jikomes 1:05:10

So as your body becomes literally taller, some of your internal organs like the lung just can't overcome it, hold it together,


Razib Khan 1:05:18

hold it together, the heart heart attacks are common. So people over seven feet, unfortunately, like, you know, you see this and like famous people like Andre the Giant and stuff like that, they just their body just collapses, it's over. Because the issue here is like, your, your mass is increasing by cube, but the cross section is square. And so it's like getting harder and harder to support everything and all this stuff, you know, and your heart has to get bigger and bigger, but it's not really getting bigger and bigger, proportionately to get to the whole all part of your body. So that's one thing that happens. Another thing in a pre modern environment is you need a lot of calories. So imagine you live in a society, there's a famine every three years. You just die.


Nick Jikomes 1:06:02

You know, I see. So even if even if those individuals are those males particular are preferred, they are just subject to these sorts of brute physical realities, the


Razib Khan 1:06:13

selection, natural selection can operate, you know, in something there pygmy populations in the world. And, you know, it could be that women in those societies are just like, That guy sexy, but he's not going to be around next year. I'd rather have a guy that's around to help out with the kids. You know, I mean, they could start like, thinking that way, you know, I mean, I think they can observe, you can observe people, right? So I think similar things with like intelligence, like I have a post on my substack, where I use John von Neumann as an example, John von Neumann, for those of you who do not know, it was one of the probably, you know, they say, he's a smartest man in the 20th century. I think, you know, I read a book about him recently, I did a podcast with the author, but von Neumann is like, he contributed, like, several dozen fields. And I think the way to explain him is, he's like, you know, unlike Einstein, who like thought really hard and a few big problems and made some like massive breakthroughs. Orthogonal, von Neumann, like took, like, what a mathematician could do and supercharged everything, right. And so he made like, many, many advances in many, many different fields. But maybe he did not, like transform our understanding of existence like Einstein. So that's why we know Einstein, even though he he made achievements in like, you know, a handful of fields. I mean, that's actually way more than 99.9999% of people not hating, but von Neumann, literally, if you go to the Wikipedia, you'll see that he made achievements in computing, mathematics, game theory, economics, and a couple of other things and other things I can't even think about, like, you know, like, you know, he did a, he was co founded Rand, you know, he did like, you know, stuff about foreign policy, whatever. Right. And so he had one daughter. Okay, and I think she has two kids, by normal reproductive fitness is quite low. And yet, he's like many, many standard deviations above the norm. Right. So what's the point of it? Well, I mean, it's cool for human race and stuff like that. But the game of illusion only cares about how many children you have. Right? And like this is, this happens with this one, like, not a Hilbert, the guy who got ill, he starved himself to death. Because after his wife died, because he didn't trigger people, who did you know, girdles incomplete Incompleteness Theorem, mathematician. After his wife died, he didn't trust anybody to give him food. Because he thought they would poison him. So he's starve to death. Right? So really brilliant guy. Obviously, my only point here is like, I'm not really sure how, in an evolutionary sense, this is really beneficial. Unless like, unless you're awarding harems to guys that the guys that are like winning the Fields Medal or women that are, you know, giving them like, lifetime, like, you know, 10 nannies or something, you know, they're not going to, it's not going to benefit their genes in any way. And so if they're not gonna benefit their genes, and who cares, if it's not benefiting the genes, it's going to be random genetic drift. And you're probably going to have like, drift mutation equilibrium around some average average amount that's necessary for functioning as a human in a society, right? Because as your intelligence goes lower, your fertility does drop, like really at the low end, it does drop like people are like, don't want to hook up with you. You might also have some disease issues, just because that's might be why it's lower. But, you know, people that are in the mid range, they do fine. You know, and as you know, like there's a correlation with like, basically, there's a correlation where like, education tends to drop your fertility, especially for women, you know, so There's just a lot of reasons where it's like, I don't see any evidence by being super smart as it is super beneficial.


Nick Jikomes 1:10:07

What? So when you think about intelligence, in comparison to something like height to say, what is the heritability of these traits? And how much how much of each trait is due to genetics versus environmental factors?


Razib Khan 1:10:19

Yeah, so I gotta like the heritability of height, like is, you know, people usually say 80 to 90%. So that means 80 to 90% of the variation is due to variation in genes. So the correlation between siblings is, is a point five, I think, like, Pearson's correlation, you know. But I have a friend who says probably a little bit lower for various reasons, although he's hedging on that. So I will say, probably 80, or 90%, but it could be as low as 65%. Okay. Just to be just to be clear, if there's any, like quantitative geneticists out there who are like, received doesn't know the latest work, I do know it, it just hasn't been totally accept it. So anyway,


Nick Jikomes 1:10:55

so So what you're saying is that the majority of the variation we see in something like height, is due to variation in genes.


Razib Khan 1:11:01

Yes. That's why That's why when you do a scatterplot, you do a scatterplot, you do the, the offspring of the parents. And you see there's like a pretty tight correlation, right? So like, point five would be the art. But anyway, and then the regret of slope of the regression line is the heritability. And the slope is not one, but it's like, it's like that it's high, right? With IQ, like I usually give 50%. But that kind of depends, because high IQs probably is little bit more sensitive to like, in the developed world, very few people do not eat enough. On the contrary, you know, so. But you know, as we know, eating too much doesn't make you tall, it makes you wide. So all of the variation you see, and how tall you are pretty much is due to your genes, like there's a little other stuff. And a lot of it is also like it could be like stochastic, like randomness within developmental biology. It might not be environment environment, like what you eat, or what anything like that probably is not actually. Right. So with IQ, it's a little bit more like environment can probably matter, in my opinion. And in terms of what inputs you get, and stuff like that. There are some evidence from the early 2000s, that lower socioeconomic status, the United States has lower heritability, probably because of more environmental stochasticity. That's a very popular study. But I'm going to be honest with you and say that, like people privately been like, that needs to be replicated, because it hasn't been replicated. And people just really liked the study, because they liked the conclusion. All right. So I'm going to say that, but other people say that the heritability in upper middle class families could be could be like in height range, maybe not quite as high, maybe like 70%, point seven, but it could be quite high. And the way you're really going to analyze and understand this is looking at siblings, because siblings are on average, 50% related, but they're not exactly 50. I have like two sibling pairs, I have siblings, I have siblings. And I have, well, I'm not gonna say the sex, but the 40% related, right? There's a lot of siblings that are more than related than 50 and unrelated and 50. And looking at their phenotype looking at their traits, and how related they are genomically, that's going to tell you how genetic quote unquote, something is, like how heritable it is, right? So if siblings that share a lot, share more than average, genomic Lee, almost always, you know, like, the correlation with sharing, the trait characteristic is quite high. That's probably a really heritable trait. If they don't, then it's probably not a heritable trait, you know, because siblings raised in the same environment, same socioeconomic status, and cetera, et cetera. And so I think that's going to answer the final question. But I think if probably, I think like, you know, between 30 to 70%, looks safe, I usually get 50 to split the difference, because it makes math easy to do like back of the envelope chemical calculations.


Nick Jikomes 1:13:53

So, you know, when it comes to the subject of genetics, generally and human genetics, in particular, why do you think this area is one that attracts so much spirited interest, but also controversy and conflict among people? You know, you mentioned you just mentioned a result where, you know, you said something like, people like that, that type of result. And I think human genetics is an area where people have a lot of motivated reasoning, and they're looking for answers that they want to find. Why do you think that is?


Razib Khan 1:14:20

Nazis? I can say more, but, I mean, there's not very much genetic research done in Germany, you know, it's not because they don't have the ability to, you know, they're scared of it, you know, Nazis, eugenics, like, that scared a lot of people. And, you know, this idea of like, group differences bell curve, genetic discrimination, Gattaca. I mean, I could probably just keep going.


Nick Jikomes 1:14:47

People are just scared of the dark places it could potentially go,


Razib Khan 1:14:51

or it has gone. I mean, that's, you know, that's legit, you know. So, yeah, that's why they're motivated. I mean, like, you know, I mean, you've done research I mean, like, I I'm in trouble. I'll just say what I think that gets me in trouble. I'm not going to like, you know, okay, like this, I'm going to be honest here. It's like, I do get frustrated when a lot of geneticists just like, Guess what, like, every liberal thing in the world is confirmed by genetics. I'm like, Oh, that's really coincidental. You know, I'm just, I just say, like, when people are telling you that everything in their field confirms everything you want to believe. I think maybe you should take them with a grain of salt. Right? So I think that is what's happening a lot of times now, because geneticists are very scared about their, you know, what is it like harm reduction, that's a new thing in academia. And they think that, you know, their field is going to be used in negative ways. You my opinion on that is like, fundamentally, it's pretty, my reading of history of science is actually far less influential in shaping ethics and thought, than it is executing things. Whether it's a green revolution, or war, like war is bad. Green Revolution is good. Right? But science did not tell us that war was good. Or that, you know, the Green Revolution was good, or bad, right. So that's my attitudes, but people think so for example, there was a huge controversy about genetic variants associated with homosexual behavior in the British biobank data set a couple of years ago, and people are like, they're gonna screen for gay babies and all this stuff. And, you know, my question to scientists, and they routinely do not answer this is if if you're worried about that, why are you not saying anything about the mass extermination of Down syndrome fetuses in the United States? Today? It's happening, you know, why are you not saying anything?


Nick Jikomes 1:16:46

Do you just mean? Like, what do people are people doing like in vitro fertilization, and they can check ahead of time.


Razib Khan 1:16:52

So insurance makes it free to do non invasive prenatal testing for Down syndrome now. And so if the numbers are hard to get, because people don't want to report it, but it's quite if you're, if you're 35, and above, this is last I checked, because like my, my wife, and kids, whatever, they actually recommend it and it's free. If you're under 35, they don't necessarily recommend it. But I think it's usually free too. And it's basically a blood draw. And it tells you if you probably have a Down Syndrome fetus. And if you, you know, have a Down Syndrome field, you could choose to terminate an early second trimester. You know, it's nothing invasive, like no amniocentesis. So there are countries like Denmark that almost have no Down Syndrome children. Now, Iceland is at zero, you know, so is that good or bad? I mean, what do you think?


Nick Jikomes 1:17:37

Well, I mean, it's, I didn't even know this was true. Yeah.


Razib Khan 1:17:40

We don't know. Because I don't want to talk about because it's happening. And I'm gonna be honest, I think a lot of it is that the socio economic status people who do this sort of thing are exactly the people who are scientists, they're academics, they're doctors, they're lawyers. They wait until their late 30s to have children. And so if I ask them that question, I'm getting annoyed because it really annoys me that they don't answer the question when they get all righteous about other things. Because they don't answer the question, because it's them that's doing it like that. They're doing eugenics. But they call everything else eugenics is really frustrating to me, you know, I'm just like, if you call that eugenics, what you're doing is eugenics, you know, like,


Nick Jikomes 1:18:18

so let's just formally define eugenics just just to be careful. So eugenics is basically just when you can either be selecting certain individuals with, you know, good traits that you want to have, or it can be getting rid of bad, you know, quote, unquote, bad thing such as Down syndrome or so yeah.


Razib Khan 1:18:38

And technically, and I've gotten to this technical argument, because I think that it's important, technically, eugenics should be changed in frequencies. That was like people that are transmitting it. Most people with Down syndrome are not fertile. The females have a last I checked, a very subfertile of the males just sterile. So you can say that that's technically not eugenics, because all it's doing is just preventing those people from existing, but they wouldn't have transmitted it in any way. Right. But I think like the way most people use the term eugenics is it clearly is eugenics colloquially, and the way scientists even use it, even though they should know better in my opinion, anyway, I'm sorry that I kind of got annoyed, but it's like I have asked multiple times people and they just refuse to answer me on that question. Because, because when they bring up eugenics, I'm like, Well, what about this? They don't answer the question. The only people who care about that are pro life activists, because they're pro life. You know,


Nick Jikomes 1:19:30

I see. Yeah, I mean, it's definitely an interesting question, Should you be you know, allowed to terminate pregnancies simply because a given child will have a predisposition to some some disease, whatever, you know, cancer or some metabolic disease or anything. And it's it's definitely not super straightforward to think about.


Razib Khan 1:19:47

Well, I'm Emma is going to be here, like within the next decade. I mean, you can already do it now if you really wanted to, but like I did in 2014.


Nick Jikomes 1:19:56

So how does how did you actually do that? Just


Razib Khan 1:19:59

I got like, sampling from chorionic villi and then basically bullied the lab and just send me the amplified DNA.


Nick Jikomes 1:20:06

Oh, and then you just check yourself because you know how to look for it. Yeah, I


Razib Khan 1:20:11

got this girl. So I see.


Nick Jikomes 1:20:12

Okay, so yeah, it's possible to obtain the right DNA sequences to check. And if you're capable of checking yourself, you can just look for these things.


Razib Khan 1:20:19

Yeah, so I did. But it's not easy for regular people, because they haven't operationalized it, there's not a high demand for it yet. But in the near future, it's gonna start to happen where you can sequence in the first trimester, and then it's then then it can I swear, yeah. Oh, yeah, go for it, she's gonna get real. So you start to figure out like, what people want what they don't want. You know, I think most people are not going to get too worried if their child has like, a two times greater chance of type two diabetes. But you know, there's going to be stuff besides Down syndrome that people are going to pick up? And, you know, I don't know, I don't know why, or


Nick Jikomes 1:20:54

even something as innocuous as height. Right? You can imagine that people might have a very strong preference only to have a pregnancy if the child's gonna be above some height. Yeah,


Razib Khan 1:21:03

well, I mean, look at sperm banks, sperm banks are very general. Basically, you know, you know, sperm banks are subject to winner take all like, like that parental principle. It's only like a few of the donors that ended up like getting requested, because they're tall, athletic, good looking, you know, blue eyes, like there's certain things that people want, you know, and so I'm assuming you'll see that with, with a, you know, probably, instead of doing like, you know, first trimester treatment as a termination, termination is less traumatic. Probably you do embryos reading, embryo selection, these sorts of things. Some of its already being done by certain companies, certain high income net worth individuals, some of its publicized, but most of it is not. So it's not mass consumer. But the Down Syndrome thing is, you know, like, they're probably like, 10s of 1000s Down syndrome, fetuses being aborted every year now, you know, looking at the numbers,


Nick Jikomes 1:21:58

what about like, the more popular and accessible personal genomics products out there? Things like 23andme, and ancestry.com? How good are the services? Are they telling people information that is legitimate and usable? What are people using it for? How sophisticated are those things? Actually?


Razib Khan 1:22:16

Yeah, they're really sophisticated. First of all, and I know a lot of people that work there. I think I can say this, because I had it on a podcast podcast, I'm friends with the co founder, 23andme. You know, I've done work for Family Tree DNA, which was a pioneer in the field. So yeah, I think it's great. It works. There are limitations because genetics can only tell you so much. The killer app has turned out to be actually ancestry and genealogy, and relative matching. And that was kind of like done as a lark. But it turns out, people are interested in that. So that stuff is pretty accurate in a lot of ways. There's some details which like, I don't want to get into, like, the weeds on because it's confusing, but they're pretty accurate. Okay? If they tell you if 23andme tells you your point, 5% sub Saharan African, you are okay. I'll tell you that. Right? If it tells you this point, 1% Okay, maybe there's a 50% chance they're wrong. But if it's point 5%, that's, that's a true positive, like they know, right? They, because remember, they're looking at hundreds of 1000s of markers in your genome, they're not looking at just a couple of markers, you'll get hundreds of 1000s of martial art aggregating all this information. Most people don't have actual medical instrument and results. And that's good, but some people do. So you know, I know people who found it out pretty disturbing things, just for their life course. Not many, but enough. So that's legit, that stuff is pretty vetted by their doctors internally, from what I know, once you have whole genome sequencing. With larger populations, you're gonna have more and more insight coming out of it. I think there's always gonna be limitations, because not everything is genetic. And as you get older and older, you have a better sense of what your risk susceptibilities are, because that stuff would have happened, right? So I think that the real real benefit is going to be when you start when everyone starts sequencing infants, and then you start to have the sense of like the life trajectories and other things, but you can modify certain things in certain ways. So let's say that, let's say that we have a pilot, we have a polygenic index that predicts schizophrenia, right? Well, I mean, that's pretty important to target people with those susceptibilities early, you know? Yeah, it's not perfect. But you know, schizophrenia is 80% heritable, by the way, as is autism. So they're highly heritable. So you're gonna have that risk. I mean, it's not guaranteed, but you have much higher risk of the average person.


Nick Jikomes 1:24:33

So if you if you knew that from birth, you could just be much more cautious about trying to minimize your exposure to risk factors.


Razib Khan 1:24:41

Yeah, well, I mean, do you have any kids? No, I don't. Not that you know. But no, what I'm saying is, like, you know, with autism and lesser extent, schizophrenia, definitely autism. I know a lot of parents who are really paranoid when they're when their kids are babies, because they're always looking for signs of autism because it's just it's scary, you know? You don't know, that guy, he's not making enough eye contact? Or is he just like a dick, you know, whatever. If you if you had, like, if you had a polygenic index score, that would tell you something, that would be great. A lot of it happens to be people who have who have family history of autism, so they're cute and for it. And, you know, the reality is, you might have family history, but it might not actually matter. And you might be able to tell, or it might matter. What I'm trying to say is like, let's say your uncle has autism, you have a son, and you're paranoid that your son's gonna get autism, because you know that the numbers are like, way higher for your son, potentially, even if you didn't manifest, right. So you take a test, and it's like, oh, risk is like, 10% greater than average. Okay, that's not really that big of a deal. Right. But what if it comes back into risk is like, 250% greater than average? 2.5? Okay, that's a big deal. Right? I don't know, I think those are the sorts of things that are gonna happen in the future. And I think that's, that's how genetics can make. Genetics can enable for greater human flourishing in that way, by giving us more information to make decisions, right. That is the dream, in my opinion, you know,


Nick Jikomes 1:26:09

interesting was,


Razib Khan 1:26:10

you know, just individual choice like that is a dream.


Nick Jikomes 1:26:14

And so do you. Do you have any of these personal genomics resources that you recommend that like, Are there any that you think are the best in terms of do they do partial genome versus whole genome sequencing? Are they giving people more information about health? Or are they all pretty similar in terms of their quality?


Razib Khan 1:26:31

Yeah, this one, I'll say 23andme. They do genealogy and health. So if you want health go to 23andme. They're on a snip array, I think their snip array of 650,000 markers. Now, ancestry uses a more generic snip array 100,000 markers, that difference doesn't make any matter. But they don't do any health right now, I think, I think they have some health options now. But they're just starting it. If you're really interested in genealogy, you might want to do ancestry, they have a bigger database. Family Tree DNA does really detailed stuff on Y and Y chromosomes and maternal lineages. mtDNA. So if you want that do that, if you want a whole genome sequencing go with nebula genomics, they're based in the Bay Area, they're pretty good. And if nebula doesn't work for you check out Dante genomics, they're based in Italy. I mean, I've done Dante and Nebula, they're both similar. Dante is a little bit. Well, I mean, not Dante is a little slower. That's all I'm gonna say, Okay.


Nick Jikomes 1:27:22

Is there any risk? What are the risks for using these products in terms of how secure your data is? And if it's going to be you? I mean, do these companies then own your genomic data?


Razib Khan 1:27:33

No, they don't, they can't legally own it, although like you can opt in to allow them to do research on it. In terms of security, I think they're pretty secure. But I think they really, what I'm honest with people about is I think the weak point in security is going to be in hospitals, hospitals have notoriously bad security. And I think most of us are gonna get genome sequenced in a hospital in the next 10 to 20 years, and a lot of us are going to get our genome sequence stolen. So now I'm here to be honest about that. I mean, hopefully, like me saying stuff like this to people, Well, someone in power will listen and actually do something. But um, you know, hospitals, like, have you heard about hospitals, like losing records on like, you know, patients, patients with HIV and stuff they get like, so like, now people know, these people have HIV, like they get like, slapped on the wrist? You know what I'm saying? Like, they're just they take security really lackadaisically from what I've seen in furnished security. So I


Nick Jikomes 1:28:24

see just because they don't have to, because there's nothing they have to


Razib Khan 1:28:28

do. Yeah, no, they do what they have to do. So I'm really like, assuming that that's going to be the that is the weak link, it's going to be hospitals, because these private companies they're under, they're under scrutiny. When you go to the hospital, one, you're probably like, not super excited, you're in the hospital. And two Americans have a certain amount of trust for doctors and hospitals. You know, but I have heard of way too many things. I know way too many things related to hospital information, tech security, that I think that that's gonna be the weak link, because sequencing is a commodity product. You can do sequencing. So the first human genome was $3 billion. Current human genome at high quality is now $200. A Nebula, I think.


Nick Jikomes 1:29:16

So that's what it costs to get a full genome sequence.


Razib Khan 1:29:18

Now it does, yeah. Wow. Yeah. So I mean, hospitals can do it easy, right? They've already started doing it with newborns that are not flourishing. So their newborns that like aren't like, us, they are latching, we're not crying, and they're know what's going on. So they sequence start sequencing them, 40% of them, they immediately detect what the disease was. So they're gonna start to see its utility. And there's gonna be things like cancer genomics, where, you know, they'll, they're gonna want to do repeated sequencing on your tissue, possibly. I mean, Steve Jobs was one of the first that did that. I mean, he spent like 100,000 or whatever dollars because he had money, but now it's, it's like 1000s Mac's, like, it shouldn't be that expensive, right? So hospitals probably do that. And where are they going to store the data? Where are they going to do with the data? You know, I mean, I'm not like a not shilling for my company here. But that's one of the reasons that we may generate, that I started generate with my co founders. There's really a lot of data out there. And we got to figure out what to do with the data and what's human data? There's all sorts of ethical governance, security issues related to it, right. So, you know, I think the estimate is by 2024, there will be as much genomic data in the world as there was like, data on the internet in 2015.


Nick Jikomes 1:30:36

Oh, wow. So okay, so your company's doing stuff related to storage and security of genomic data?


Razib Khan 1:30:41

Yeah, analysis, like, we have a pretty ambitious goal, even though we're small. But yeah, but I mean, basically, the company, I started, I helped start the company, with a co founders, partly just because I think the need is going to be here. And you know, you're gonna have nations like Britain that do socialized medicine, that are going to come online with full genome sequencing of everybody. And where are you going to put that, you know, right now, they put on clusters, but I think I think we need like more rational, scalable solutions soon. It's gonna, you know, the technology that we have the technology, what do we even do with it?


Nick Jikomes 1:31:16

What's it been like balancing your startup duties and responsibilities? And your day job, so to speak with the subject that you run?


Razib Khan 1:31:25

Yeah. You know, honestly, the substack is, like, you know, people are surprised that it's very part time. And, you know, I do have editor who helps, and stuff like that. Podcasts I have an editor for. And I have a guy who does transcriptions and the stuff that I write, I'm not saying it's easy, but everything that I write, I have to do very little research. I know that stuff. Right. So it hasn't been the easiest, but it's been easier than you might think, based on the output. Just because if you look at what I write, if you've been following me for a couple of decades, you're like, Okay, he knows this stuff, like in terms of like, you know, exactly why I know this stuff. If you just come to it fresh, you might be like, Wow, he's doing a lot of research for this. I'm not doing any research for anything, you know, like, I can do the analysis easily. Like I've done it for decades. I love history, you know, until I had, I have not read many books, frankly, over the last year and a half, because of the startup time. But I you know, spent, you know, you go to my good reads go to receive Khan good reads like, I put like 12 1300 books that I've read there, you know, so I know all that stuff. And so it's just putting it all together and presentable way. So anyway, I just, I've told people this because they're like, you know, how can you juggle it? I'm like, Well, I think what you don't understand is, I don't write about things. I don't know. And if you don't write about things you don't know, it's much easier.


Nick Jikomes 1:32:56

Yeah. Right. And so how much? How long you've been doing the subject down? How much has it grown in that time?


Razib Khan 1:33:03

It is since November of 2020. So I started on 911 of 2021 when I imported my MailChimp list, but I didn't post anything until like, November. And then so it's been like almost two years. And it is the number two science substack on subject so


Nick Jikomes 1:33:23

and so how many look roughly how many subscribers? Do you have? What order of magnitude?


Razib Khan 1:33:27

1002? I can't tell you that I have 21. Second, let's just do it right now. I can tell you how many free subscribers I have at least I think that will give you less resistance. I have 21,333 emails right now.


Nick Jikomes 1:33:47

And so is that. Is that surprising to you? That's that number of people are interested in this subject? Or is it not surprising this because, you know, there's billions of people on the globe and the internet just allows almost everyone to, you know, with an interest in any subject to now find it so easily.


Razib Khan 1:34:03

Yeah, that's a good point. Um, yeah, I'm surprised. I'm surprised at the attraction. I think a lot of people are surprised at the attraction. Some people are happy, some people are angry. But, you know, there's, there's a hunger for whatever I'm giving people, you know, there's a hunger for it. So I'm giving it to him. In terms of like, you know, your point about the whole world out there. Yes. So, you know, I have a core audience of like, 1000s that I've cultivated, I guess, over the years that have followed me, you know, my, like, 60,000 Twitter followers now. 50,000. But a lot of people don't know me, but you know, there's poor people have been reading me for 20 years since I was like, It's a weird thing, like, you know, someone who's like, been on the net. I mean, I've been on the net since 94 on Usenet, but like as like, like writing on blogs and stuff since 2002. You know, I was, I was a boy, you know, and now I'm old. You know, All those there have been people who like aged with me and that they follow Me. But then there's a whole new groups of people that have discovered me through substack and do my writing and my writing is my I've tried to make it much more writerly, much more accessible. You know, I read dry things, and I'm okay with it. But I understand that other people or so it's not written like, and you probably read some of it, it's not written like a scientific paper, you know, like, it's written in a way that can engage people. So, you know, there are people a lot, for example, a lot of retirees read me, a lot of people that are like retired engineers, you know, retired doctors, and they have time now, they have the income to like spend on sub stacks. And they want to know about stuff. And they never had time to know about stuff, besides what their professional duties were, for many, many decades. And so a lot of them have, like, some technical skills, so they can follow some of it, you know, so it's like one of my posts, I, you know, I posted about, like, you know, the revolution in biology and genomics, and I decided, like, I like it, I want to do it, like I started talking about hmm. And some people were some of the engineers were like, really appreciate you talking about hmm, because what's what's hmm, hidden Markov models. So it's just like a, in genomics, it's useful, because sometimes you see part of the genome but not other parts of the genome. And you can infer other parts of the genome from the parts you see. And it's just, I think it's something out of it. I think it's out of information theory. And so I think it's been used in like, signal transmission and stuff like that. So engineers know all about it. Yeah. Biologists do not know all about it. But you know, misses have started to use it. So I just put that in there. Mostly because I'm like, I know, I have some engineer readers. And I know, they're kind of lost in the biology a lot of the time, but here, I'm gonna, like, throw them a bone. And they were like, multiple were like, Yeah, I know what a hidden Markov model is. Now I understand what you guys are doing with this with the sequence. It's like, okay, yeah, that's, that's just gives you a sense of my readership.


Nick Jikomes 1:36:52

Interesting. And how do you think like, substack has grown so much in the past couple of years, I subscribed to several substack authors myself, I love the fact that I know that I'm reading an individual person that sort of unfiltered. Yeah, in a way that you're not going to get at, like a mainstream corporate media outlet where you've got, you know, at least one layer of filtering happening through an editor, and just sort of the larger corporate culture that's gonna, you know, tend to shift things in a certain direction. So so I like that it's sort of unfiltered in that sense. Do you think that substack and, and things like it will continue to see more and more momentum?


Razib Khan 1:37:30

Yeah, I mean, it is right now. So you know, and it's its growth phase. You know, money is finite. On the other hand, you know, people love following individuals. So there's some sub sects that are kind of like more conglomerations, so the dispatches on sub stack, and they're leaving sub stacked mostly because I do want to create an institution. Whereas like, I don't want to create an institution, like maybe I'll do guest authors, guests, guest people, but I think people want to hear from me, honestly, and you know, I do podcasts with people, but the ones that I've done as monologues are actually quite successful as well. And so this is, I think, I think I will be successful as long as I keep my voice because that's what they want to hear. And I have a certain reputation, whether it's good or bad, it is what it is, and so they know what they're gonna get. Whereas, like, if it's an institution, well, who knows, new editor comes in, you know, they decide to change the narrative, these sorts of things. I just try to think of why I subscribe yourself stacks. And a lot of it has to do like, you know, Glenn Greenwald is gonna take Glenn Greenwald, just whether you hate them or love them, right? Because he knows who he is. Whereas like, you know, I have subscribed to institutions that have changed over time. And subscribe to The New York Times a lot, like since I don't know, 2005, like, forever, right? It's changed a lot since 2005. You know, and it's, I still I'm still subscribed to it, but you know, you got some, you know, got some things to say about that. Right? So, you know, so I think subjects are gonna be pretty successful because of that. And also, people are sick of the algorithmic social media blast furnace. There are people that are like removing Twitter and just reading sub stacks. I'm not but I respect that choice. That is a choice you make. And I think that's fine. And you know, part of it is also substack defends a view of speech and I don't like nude I didn't like do like research on you. I don't know what your view of speech is. Right? But substack the people the people who run substack the founders and you know, I've said this in public before so I can say it again like I met them and they would not say free speech with quotes ever. Like they actually believe in it to a pretty extreme extent. I my opinion, that's good. You know, to quote Wildwater extremism in defense of liberty is no vice. You know what I mean? They're kind of like of that when it comes to speech there of that. That's not very common on the internet today. A lot of people are very worried about disinformation, which is basically information a lot of times they just don't like, you know, let's be honest, in my opinion, I don't know what your opinion is, but I And so it's a different older throwback philosophy to the internet that I remember, which was, you know, the internet that I remember, when I got on the internet, the 1990s, I remember seeing all these people, a lot of them kids getting on the internet, and like finally saying things that they couldn't say in their small towns, you know, that was the, that's the internet spirit that I loved. And that spirit is gone now. Because, you know, people are always worried about what they're going to say, and what people are going to think and all of this stuff, you know, someone was, I saw someone on Twitter, saying, like, you know, not surprised receivers, posting most of us most of his stuff on substack. And they were just like, it just sucks. You know, basically, the person was pissed that I had a platform. And that that is that is today. That's 2022 A lot of people, you know, where they're just their goal is to like, get people not to have platforms, like you're not platform, this person, do not platform, that person do not be seen with that person. It's like middle school, except it's the whole world now, you know?


Nick Jikomes 1:41:01

Yeah, I mean, to me are escaping. I mean, to me, though, the only logical conclusion of that attitude is that the person saying that should be the decider for what gets said and doesn't get said everywhere. Oh, yeah, no, yeah,


Razib Khan 1:41:16

it's quite clear. It's quite clear the way they're talking. Like, they believe that they get to decide what other people get to hear, you know, what is licit. And not listen, and all of that stuff. And substack does not have that philosophy. And the founders do not have a philosophy. So as long as the founders are involved, I'm pretty confident, I have a mailing list. I'm not gonna try to give Gmail, any ideas. But you know, as long as they don't start to like, as long as Google doesn't start to, like, you know, censoring emails, newsletters will still continue, you know, I mean, ultimately, you can't really censor anything in some deep way, people will always find it, you can just make it harder. And that does change the discourse, in terms of, there was much more freewheeling engagement in the internet in the 2000s. Before the rise of social media, social media has like resulted in you know, it's resulted in like a certain tone. And this tone is not just social media. I was. I was watching YouTube's on the new Rings of Power show, and there's some who are pro and some who are against, and they have opposite views. But I was listening to the youtubes. I was like, they use the same smug, sneering tone, though. Both of them do. I mean, can I can I just like, hear your opinion about this without you being so fucking smug? You know, that sometimes I agree with you. But like, Damn, you're smug. You know, sometimes I disagree with you. And I'm just like, You know what, maybe you would actually convince people if you weren't so smug? I don't know. But it was like it's the same tone. It's just like smug tone where everyone listening to you agrees with you. And they know you're right. And that's the tone you see on Twitter. You know, and I'd probably do it, but it's probably like a structural issue with the platform. But I substack is trying to go beyond that. I'm trying to go beyond that, at least in my long form pieces. And, you know, there's truth out there, we just need to find it, we need to grasp it. You know, I think the earlier stuff we talked about, I mean, I think the listeners and viewers, I think they get a sense that I've seen some of the people you talk to, like they know like we live in an age of wonders. We just need to open our eyes and look at look at that and focus on that. But instead, we're keeping up with the Kardashians. Like I really sound like a curmudgeon here, but you know what I'm saying?


Nick Jikomes 1:43:32

Yeah, no, no, I think I think that's that's a great place to end it actually receive. On that note, I definitely agree with you about a lot of what you just said, Do you want to just remind everyone who you are, where your stuff SEC substack is and what you tend to cover for those that are interested because I think you know, at least a portion of my audience will be interested in the types of stories that you tell. I've read several of your works. And I really liked the stories that you put together.


Razib Khan 1:43:55

Yeah, so my name is Eve Khan. You can find all of my stuff@reserved.com. Right. My substack is reserved.substack.com. And so I mostly talk about like, you know, genetics and history and the intersection between the two. So for example, genetic history of the Ashkenazi Jews genetics history of the Romney's, you know, genetic history of Greece. I did recently, I also like talk about human evolution, obviously about Denisovans, Neanderthals, these sorts of things. Sometimes I talked about characteristics. So I did do a post on height. I did do a post on skin color and other characteristics like that. I also contribute to unheard and a few other places. So if you go to reserve.com, you'll see that and if you want to follow me on Twitter, it's reserved con Arese IB Kjn. That's pretty straightforward. And I have some blogs and I have some other stuff and yeah, so it's been great talking to you, man, and I hope your listeners and viewers enjoy it.


Nick Jikomes 1:44:50

All right, reserved, calm. Thank you for your time.


On today's show is brought to you in part by dosed an all natural cannabis company specializing in dose controlled cannabis products made with plant based ingredients to learn more about dosed their products and where they are available please visit their website through the link in the episode description

6 views0 comments