The Y Chromosome Bottleneck

Genetics of a gender gap in consciousness

The male family tree, most notable for who’s missing

You may have noticed that men and women think differently. Don’t worry, it’s not a sin. In fact, female preference for people vs things is a well-studied phenomenon in psychology. If the self was discovered, then the explorer of the interior would likely have been a woman. This is central to the Eve Theory of Consciousness.

If men had to “catch up” on recursion (ie. self-awareness, language), then this would be accomplished on the Y chromosome¹. Therefore, the claim that men only recently became conscious is a claim that:

1. there was recently massive selection on the Y chromosome and

2. the Y chromosome is involved in the construction of self.

Like Jaynes, I interpret the Fall of Adam to be the moment he discovered duality and identified as a moral agent. Unlike Jaynes, I entertain the veracity of other details of the myth—that women made the discovery and that agriculture was a result. These claims are, of course, correlated with other streams of evidence (see a list of 6 supporting the idea of a female vanguard). Duality leading to agriculture deserves its own post, but for now let’s consider that the Agricultural and Sedentary Revolutions, the start of the Holocene, and the end of the Sapient Paradox are all intertwined changes that are poorly explained. If something like The Fall (or Bicameral Breakdown, if you prefer) actually happened and was recorded in Genesis then it’s worth taking the timing seriously as well; it could explain our species’ incredible, synchronized turn to art and agriculture. So an auxiliary third claim would be: the selection starts in earnest just before 12 kya².

EToC holds that creation myths contain kernels of truth and can be used to understand when and how we became human. Molecular biology allows us to test that. This post will address the claim # 1 about selection, the next will cover the construction of self.

Previously, I reviewed how recursion is related to consciousness and various attempts to date when that evolved. EToC holds that there were glimmers of recursion 100-50k years ago. In the beginning, this would have been sporadically experienced, and more commonly (or exclusively) by women³. Self-awareness became the base state for men around the Holocene. This and the next post explore the genetic predictions of that theory.

A note about the scientific method: hypotheses can be falsified, not proven. Hence if you want to show a model is accurate, you use it to make predictions that can be shown to be false. You can’t prove the model, but if it makes strong predictions that turn out to be true, it increases your confidence. The more surprising the prediction, the more your confidence increases. Hence the theory of evolution isn’t a fact, per se, it’s just that Nothing in Biology Makes Sense Except in the Light of Evolution. It was surprising that humans share a common ancestor with apes. Likewise, it would be surprising if the Y chromosome recently underwent massive selection all across the world.

Outline

1. For decades and across many research labs, it has been observed that there is way less diversity on the Y chromosome than is expected. The most natural explanation is selection.

2. But a 2015 paper found that the decrease in diversity happened fairly recently (20-5kya). This makes selection an awkward explanation because it would have to be very strong and spread over multiple continents.

3. Several papers have tried to explain this without recourse to selection. The proposals are fantastic in their own right.

4. EToC predicts Y chromosome selection, and it may in fact be there. This won’t be a mystery for long, it’s an active area of genetics research.

Papers in this post are discussed chronologically, in order to better make sense of the uncertainty. This also serves as a tour of the exponential progress in the field of genetics. By 2018 our genetic history is interrogated with hundreds of samples from across the world. The first paper from 2014 uses just 16.

One downside of this strategy is that it somewhat buries the lede; if the article is too long or technical feel free skip to last paper which discusses direct evidence of selection.

Natural Selection Reduced Diversity on Human Y Chromosomes (2014)

First some terminology. The genome consists of sex chromosomes (X and Y), the autosome (22 non-sex chromosomes), and mitochondrial DNA. mDNA and the Y chromosome are passed down without recombination. That is, if you are a male, your Y chromosome is identical to your fathers, other than some de novo mutations⁴. Selection refers to an organism having less surviving offspring (genetic fitness) because of a certain allele (version of a gene). The effect can be very small, but even if an allele reduces fitness by 1%, this will be enough to remove it from the gene pool in short order (thousands of years).

This paper cites several papers going back to 1990 that report reduced diversity on the Y chromosome and offer various explanations⁵. Many labs have observed this phenomenon. You really can’t miss it (even in the 90s, I take it); there is a full order of magnitude less diversity on the Y chromosome than geneticists expected. Compare this to most effects we see in science which are advertised as significant but are visually indistinguishable from noise.

The lack of diversity could either be due to a few guys hogging all the ladies (for a hundred thousand years) or selection on the Y chromosome. The first scenario reduces diversity by decreasing the population size of males that successfully reproduce. The lopsided reproductive sex ratio in turn would cause changes in the relative diversity of the autosomal , mitochondrial, X, and Y DNA. For example, fewer males reproducing, reduces diversity on the autosome as well (men have chromosomes beyond the Y). The authors compare the observed ratios to several hypothetical hogging scenarios: 4 reproducing females for 1, 2, 3, and 4 men. As you can see in the Y/A columns (Y vs autosomal), the observed Y chromosome diversity is much less than even the 1:4 hogging scenario predicts. Therefore, it must have been selection; some Y chromosome variants were deleterious and nature got rid of them.

Figure 1. Observed and expected ratios of normalized X/Autosome, Y/Autosome, and mtDNA/Autosome nucleotide diversities. Notice that Autosomal, X, and Mitochondrial diversity ratios are all close to what is predicted by a neutral model. Y chromosome diversity is an order of magnitude less than expected.

This seemed straightforward enough, until the story became more complicated.

A recent bottleneck of Y chromosome diversity coincides with a global change in culture (2015)

The next year dozens of authors put together a worldwide dataset of 456 Y chromosomes and found when the diversity went missing⁶. Shockingly, it was mostly in the last 10,000 years. Below is a plot of the global phylogenetic tree displayed over the effective male and female population sizes by year.

Effective Population Size on the Y chromosome vs Mitochondrial DNA (like the Y, it is non-recombining, although passed down on the mother’s side). Notice that only a few male lineages make it through the bottleneck at ~10 kya. Most of the diversification happened in the last 6 kya. MtDNA, on the other hand, expands dramatically after both 50 kya and 20 kya.

In their words:

The surprisingly low estimates of the male N_e [effective population size] might be explained either by natural selection affecting the Y chromosome or by culturally driven sex-specific changes in variance in offspring number [what I described as hogging]. As the drop of male to female N_e does not seem to be limited to a single or a few haplotypes, selection is not a likely explanation. However, the drop of the male N_e during the mid-Holocene corresponds to a change in the archaeological record characterized by the spread of Neolithic cultures, demographic changes, as well as shifts in social behavior (Barker 2006). The temporal sequence of the male N_e decline patterns among continental regions is consistent with the archaeological evidence for the earlier spread of farming in the Near East, East Asia, and South Asia than in Europe (Fuller 2003; Bellwood 2005). A change in social structures that increased male variance in offspring number may explain the results, especially if male reproductive success was at least partially culturally inherited (Heyer et al. 2005).

Selection is deemed as unlikely because it would have to affect haplotypes (genetic groups) across multiple continents. But note that it’s not discounted based on the data. It would just be surprising if there was such a steep and widely shared fitness slope in our recent past. They offer one alternative scenario: the invention of farming allowed some men to hog all the women. As farming spread, so did the hogs. It seems to me the spread of more complex society could also introduce selection. Even without a full-bore change in consciousness, those with better theory of mind and ability to abstract would do better.

Here’s another view of the data, the ratio of the male vs female effective population size.

Figure S5 – The temporal dynamics of the ratio of female (Nf) and male (Nm) effective population size in the last 140KY. The ratios of the global accumulative Ne estimates of mtDNA (Nf) and Y chromosome (Nm), as presented in Figure 2, are plotted against the time (in thousands of years) back from the present (0).

It’s hard to overemphasize how bizarre this is. Imagine you find a tribe and there is one dude with 17 wives while no other males reproduce. This wouldn’t be an outlier, it would be the norm. Brigham Young’s 56 wives would have made him a local legend, rather than a national scandal. And this social structure coalesced all around the world for a few thousand years before disintegrating just as quickly. It’s wild!

Now, it probably was not one dude with 17 wives; other scenarios can deplete Y chromosome diversity. But first, let’s get a handle on the timing. Notice first that the ratio starts rising 20-25 kya. And also that there is a precipitous drop after 6 kya, just before the Bronze Age. If it was technology that allowed some alpha men to kill off their competitors, then one would think the invention of bronze weapons and the domestication of the horse would be a boon to this trend. Instead, it happens at the reversal.

One final view of the data. Here are the ratios broken down by region. The major dip is quite synchronized across multiple continents. This is why the authors consider selection to be unlikely. What could cause such a stark gendered effect all over the world?

Effective population size of males vs females, by region. Each region is normed against it’s value at 12 kya; if you look at the figure above it’s clear that some regions were already starting their bottleneck (specifically Near East / Caucasus based on the first fig). All the usual caveats about inferring a hundred thousand years of genetic history on multiple continents from a total of n = 456

Fit with EToC

This is, admittedly, not a home run for EToC. In each region, the bottleneck is a bit too close for comfort, and a dealbreaker in the Andes which only declines in the last 2,000 years. Incan men were not in the process of becoming conscious when Pizarro encountered them (whatever Jaynes says contrary).

Some of this must be noise, there are sampling problems inherent to reconstructing 100,000 years of genetic history from n = 456, especially when one starts splicing this by region. Indeed a 2018 paper with 3x the samples highlights Y chromosome expansion 15 kya there, coinciding with populating the hemisphere⁷. Central Asia, too, shows a recent bottleneck. The next paper discusses how that could be due to Ghengis Khan and crew, the ultimate reproductive outliers; selection isn’t the only game in town that can decimate Y chromosome diversity.

The plot is normalized to each region’s effective sex ratio at 12 kya. But based on Figure S5 (one above) the bottleneck starts 25 kya, hence there is going to be some distortion. The start of the trend is truncated. Looking at the first plot, it seems the decrease is first evident in Africa and the Near East / Caucasus at about the time of the Broad Spectrum Revolution in the Levant. This included a more flexible and efficient extraction of resources; more plants gathered, and more small game hunted.

I keep going back to the quote from linguist Jacques Coulardeau:

“If we want to understand the phylogeny of language or any human production, we have to keep in mind the following timeline. What is most important is that an essential divide occurred around 15,000 BCE, but it took several thousand years before being effective, and in many areas in the world the transformation may have started later and may have taken longer to become effective.”

Figure S5 is a good fit for this. The bottleneck starts just before there is an obvious change in global culture. It then spreads across the world. EToC holds that this change had to do with men catching up on recursion. In that case, by 4 kya the fitness landscape is no longer gendered; men have caught up. This still requires that 6 kya in many places there were marked sex differences in language, introspection, and hearing voices, which would be absolutely wild. (And would definitely show up in myths.)

Using the dates from the bottleneck, EToC holds that ordinary male consciousness was a 20,000-year worldwide genetic process completed just before the invention of writing⁸. One way to interpret Jaynes then, is that he picked up on the tail end of this transition. From the Slate Star Codex review of Bicameral Breakdown:

Jaynes interprets basically everything that happened between about 1000 BC and 700 BC as increasingly frantic attempts to bring the gods back or deal with a godless world. Now, to be fair, he cites approximately one zillion pieces of literature from this age with the theme “the gods have forsaken us” and “what the hell just happened, why aren’t there gods anymore?” As usual, everyone else wimps out and interprets these metaphorically – claiming that this was just a poetic way for the Mesopotamians to express how unlucky they felt during this chaotic time. Jaynes does not think this was a metaphor – for one thing, people have been unlucky forever, but the 1000 – 750 BC period was a kind of macabre golden age for “the gods have forsaken us” literature. And sometimes it seems oddly, well, on point:

My god has forsaken me and disappeared
My goddess has failed me and keeps at a distance
The good angel who walked beside me has departed.

Or:

One who has no god, as he walks along the street
Headache envelops him like a garment⁹

Actually, one of my criticisms of Jaynes is that he lacked the courage of his convictions. For Jaynes, Genesis is an account of Bicameral Breakdown written soon after the fact. (Indeed some date Genesis to before Jayne’s proposed 1,200 BC for Breakdown.) But Adam only eats from the fruit because of Eve’s encouragement; why not treat that seriously? Especially considering his theory hinges on hemispheric cognition of which there are large sex differences¹⁰. Further, if “I” was invented recently then use comparative linguistics to date it. There is a case to be made “I” (well, “na”) spread to many places 10-15 kya. Finally, muting the gods must have ultimately been genetic. Schizophrenia is highly heritable (0.8!), why not the Bicameral Mind? Yet Jaynes is essentially a blank slatist on this point; for him the only difference between then and now is a realization which is now culturally dominant. It is a theory of consciousness that makes consciousness very small, it’s diffusion across the world going almost unnoticed.

A 20,000-year process is still quite fast, but there is precedence in other important traits. Since the domestication of dairy animals ~10 kya, lactase persistence has been driven to 90% frequency in Europe. A similar and independent process is also observed in Africa after animal domestication. We are a species defined by higher-order thinking; there could have been even stronger selection for recursion than for digesting milk.

In terms of corroboration the Y chromosome bottleneck is a bit more recent than the work on pronouns would suggest (which seem to go back 10-15k in many areas). And it does seem that the process would need to have started before the New and Old World split ~15 kya¹¹. This could be the downfall of the theory, and that’s a good thing. In the search for truth, we want theories vulnerable to falsification. Especially in the realm of consciousness, where falsification is notoriously difficult. For now though, it makes sense to wait for more data on the bottleneck. You have to make a lot of assumptions to model 100,000—or even 10,000—years of history from 456 samples. And there are some regions that are left out of this study. Australia, for example, could prove to have no Holocene bottleneck (though that is about the time the Rainbow Serpent takes the continent by storm). EToC predicts there must be. Again, vulnerable theories are good.

Cultural hitchhiking and competition between patrilineal kin groups explain the post-Neolithic Y-chromosome bottleneck (2018)

If you ask a geneticist the status of the bottleneck, they will point you to this paper published in Nature. It proposes a model where males compete for a shared resource (women). In it, large cultural groups of men tend to get larger, and small groups tend to die out. If cultural groups are patrilineal kinship groups (all or most males descending from a recent male ancestor), then this dynamic could drastically reduce the effective population size of males. A tribe of brothers doesn’t look much different from one man, as far as the Y chromosome is concerned. So what they suggest is that the meme “only related males are allowed in the tribe” diffused across the entire world and was ruthlessly enforced. A kind of ur-patriarchy that spread like wild-fire only to flame out just as fast.

Remember, the global male effective population size 6 kya is 1/17 that of women. In the paper, I can’t find how strict enforcement of the rule must be to obtain such a ratio. But it seems that each group would have to do a very good job maintaining their patrilineal purity.

There are many counter-examples where even brutally conquered males reproduce. The Ottomans captured Christian boys and trained them as Janissaries. Comanche, Maya, Ancient Egyptians, Romans, Mamluks and Barbary Pirates had similar practices. Slaves often had some autonomy. In fact, the most ancient Y chromosome lineage, A00, was sequenced in an African American whose great-grandfather was born into slavery. So there is the question of maintaining purity, but also the spread of the meme.

“A second caveat for our model is that patrilineality is treated exogenously. It is treated as a preexisting distinction between two sets of social ecologies whose origins are left unexplained.”

This is the “cultural hitchhiking” part; the meme to organize your society as a band of murderous brothers spreading across the world. Supporting this claim is considered beyond the scope. The implications of a meme traveling so far so fast would be huge. If it managed to re-organize people’s way of life, did it travel with other ideas like farming? Those who study the agricultural revolution believe farming was invented independently in Africa in the mid Holocene. This is during the bottleneck; was that too perhaps a transplant? It would be a great victory of science if genetics could shed light on this question.

Then there is the question of why the social structure would break down just as fast. 6,000 years ago the effective gender ratios are 17:1. Just 2,000 years later this had dropped four-fold. If patrilineal kinship groups had so little staying power, why was it such a dominant mode in very different places 6,000 years ago?

Call me a pedant, but it seems the proposed mechanism can explain rather than does explain the bottleneck. Simulations can tell us whether an idealized situation could produce the statistics in question. And I’m sure they could! But the cultural hitchhiking should leave marks too; that should be investigated before the mechanism is accepted. If this model does explain the bottleneck, we should be much less certain Agriculture was independently invented in the Mid Holocene; memes don’t travel alone. Further, selection would produce other effects beyond general diversity statistics. If there are sites on the Y chromosome that code for (or can disrupt) recursion, then once recursion became a phenotype even Y chromosomes in very distant haplogroups would tend to look the same in these locations. This can’t be tested in simulation, you have to compare actual DNA.

Selection Has Countered High Mutability to Preserve the Ancestral Copy Number of Y Chromosome Amplicons in Diverse Human Lineages (2018)

This paper does exactly that. The ampliconic region of the Y chromosome is a specific area characterized by the presence of highly similar DNA sequences called amplicons¹². It makes up about 15% of the total length and is of particular interest because it contains numerous protein-coding genes.

Copy number variation (CNV) is, as you would expect, variation in the number of copies of a segment of genome. The effects of genes aren’t always additive, but one way to think of CNV is a way to change the “dose” of a gene by repeating it.

The paper compares the CNV in the ampliconic regions of 1,216 males from all around the world. If that still sounds like gibberish think of it as a subset of the type of variation, as well as a subset of the loci. They find “each amplicon’s reference copy number is scrupulously maintained among divergent branches of the Y chromosome phylogeny, including the ancient branch A00, indicating that the reference copy number is ancestral to all modern human Y chromosomes.”

That is a lot of selection on the Y chromosome! And to be clear this could be consistent with the neutral bottleneck model above. There could be low-level amounts of selection on the ampliconic regions + a neutral bottleneck that does most of the work in the Holocene. But it is direct evidence there was some selection.

From this data, one can’t say when the selection occurred. Diverse branches now look the same, but it would take ancient DNA to say they also looked similar 40 kya. Because this region is expressed in the testis and variation can result in male infertility selection here is not really a mystery; there is a very obvious candidate that would apply for at least the last 200,000 years. But this region is also tied to the brain (which has arguably been under recent selection), which I’ll discuss in the next post.

Conclusion

The most common objection to the Eve Theory of Consciousness is that humans everywhere now develop self-awareness without any (snake venom) intervention. It’s certainly a conundrum. If we have myths of men obtaining consciousness, then it must have happened sometime near the Holocene. Are men genetically different now? Surprisingly, maybe.

Extraordinary claims require extraordinary evidence and I pony up that EToC is extraordinary. However, the narrow claim that there was selection on the Y chromosome is not so extraordinary. The Holocene ushered in incredible changes to the lifestyle of Homo Sapiens. Different behaviors were rewarded with greater survival, including the ability to integrate into a more complex society. There is ongoing debate on whether there was war before the Holocene, an invention that would obviously favor men better at the activity.

If I’m positing an ur-matriarchy spread across the world spreading consciousness, the cultural hitchhiking paper posits an ur-patriarchy spread, ending most male lineages. Their model is “neutral”, while mine implies changes to the fitness landscape (to be self-aware). It’s a very different account of who we are and where we came from. Either way, a lot of diffusion was required. That fact prompted me to try and falsify my theory using pronoun similarity; diffusion on these time scales should leave linguistic marks. In the case of the patriarchy, how did the meme get around? Were there other cultural riders? What made it so dominant? What caused it to fall apart just as fast? If the model is correct the answers to these questions would drastically update our beliefs about the past. Surely as much as “society became more complex which caused intense selection for better Theory of Mind, particularly in males”—a watered down EToC.

EToC predicts massive selection on the Y chromosome before and during the Holocene. Before because it caused the Holocene, and during because selection takes time¹³. Surprisingly, this may have actually happened. It is currently debated in the literature. The good news is that eventually, we will know. Geneticists will test for a recent bottleneck in Australia (would the cultural hitchhiking group predict their patriarchy made it there?). Or they will compare distant haplogroups to check if selection has made them look similar on the Y chromosome. This has already happened for the ampliconic regions, where similarity was found to be maintained “scrupulously” via selection. In the case of selection, affected chromosome regions will be investigated; EToC predicts they will be related to brain function.

A future post will make the case that the Y chromosome, and particularly the ampliconic regions, affect Theory of Mind.

Lineages left behind

1. Technically, males could express non-Y-chromosome genes differently and the “catching up” could happen exclusively there. This would actually serve to protect EToC in the case that there is no selection on the Y chromosome. But my instinct is to look at the most obvious places. The claim is absolutely massive amounts of selection on a male phenotype, it would be very surprising if that didn’t include the Y chromosome. Falsifiable theories are good!

2. I lay out more reasons for this timeline in Eve Theory of Consciousness.

3. Hope to flesh the transition out more. But as far as falsifaction bang for buck, makes sense to focus on the more recent proposed changes to male psychology.

4. De novo mutations are new and unique to you. There is also a small section of the Y chromosome that recombines with the X chromosome.

5. Malaspina P, Persichetti F, Novelletto A, Iodice C, Terrenato L, et al. (1990) The human Y chromosome shows a low level of DNA polymorphism.

Pritchard JK, Seielstad MT, Perez-Lezaun A, Feldman MW (1999) Population growth of human Y chromosomes: a study of Y chromosome microsatellites.

Wilder JA, Mobasher Z, Hammer MF (2004) Genetic evidence for unequal effective population sizes of human females and males.

Rozen S, Marszalek JD, Alagappan RK, Skaletsky H, Page DC (2009) Remarkably little variation in proteins encoded by the Y chromosome's single-copy genes, implying effective purifying selection.

6. This is possible because the Y chromosome does not recombine. Therefore, with a relatively small number of samples, one can create a phylogeny that estimates the global family tree. Making assumptions about the mutation rate and generation time provides an estimate of number of 100x great-grandpas, 1000x great-grandpas, and so on with surviving male lines.

7. Punctuated bursts in human male demography inferred from 1,244 worldwide Y-chromosome sequences

“As the haplogroup expansions we report are among the most extreme yet observed in humans, we think it more likely than not that such events correspond to historical processes that have also left archaeological footprints. Therefore, in what follows, we propose links between genetic and historical or archaeological data. We caution that, especially in light of as yet imperfect calibration, these connections remain unproven. But they are testable, for example, using aDNA. First, in the Americas, we observed expansion of Q1a-M3 (Supplementary Figs. 14e and 17) at ~15 kya, the time of the initial colonization of the hemisphere. This correspondence, based on one of the most thoroughly examined dates in human prehistory, attests to the suitability of the calibration we have chosen.”

8. When it started and ended would be different by region. Also, there was surely selection on the autosome (non-sex chromosomes) as well.

In a previous post I criticized Robert Proctor for his failure to directly answer “when did we become fully human”. He basically punts: “That is exactly the question we need to problematize. It’s what I call the Ghandi question. When asked “what do you think of western civilization?” he said “it would be a good idea”. So, when did humans evolve? Well not yet…”. And yet my answer is not so far from that. If recursion was gendered 4 kya, then surely we are not yet a finished product. We actually were born yesterday, in evolutionary terms.

9. One of the drums I have been beating is that if there was a fast brain reorganization, it would have hurt. Why were so many people 5,000 years ago willing to undergo stone age brain surgery?

10. “Male brains are optimized for intrahemispheric and female brains for interhemispheric communication…The observations suggest that male brains are structured to facilitate connectivity between perception and coordinated action, whereas female brains are designed to facilitate communication between analytical and intuitive processing modes.”

11. It’s an interesting time for our understanding of how humans got to the Americas. For a long time Clovis-first held, and it was thought that humans came over the land bridge and covered the whole continent ~15 kya. More recently, there is evidence that humans made very primitive tools 30 kya in Mexico. This is strange because there is no genetic contribution to current Native Americans. Where did they go?

As far as EToC is concerned, the group at 15,000 years is the first that is really obviously using technology. For example, there was a copper complex 9,000 years ago. Other homo sapiens being there isn’t really a problem for EToC. And indeed may help explain why the first group left so little impact genetically.

12. A more technical definition from the linked paper: “The third class of euchromatic sequences, the ampliconic segments, are composed largely of sequences that exhibit marked similarity—as much as 99.9% identity over tens or hundreds of kilobases—to other sequences in the MSY. We refer to these long, MSY-specific repeat units, of which there are many families, as amplicons. The amplicons are located in seven segments that are scattered across the euchromatic long arm and proximal short arm (Figs 1 and 2), and whose combined length is 10.2 Mb.”

13. I know this sounds fantastic, but my project is to develop the idea that self-awareness was a realization. You have to be willing to go to strange places to build out such an idea.