Why do black pupils in the US underachieve academically when one factors out poverty?

Great Debates
521

With all due respect, I think you would benefit from some additional reading about how the world was populated.

It is not the case that there was sort of a central population which then migrated and diverged such that modern populations are all the terminus of spokes emanating from a central hub. In that kind of model, all modern populations are more or less equally diverse, more or less similarly related or unrelated, and more or less similarly clinal from one another. Another incorrect analogy might be the way leaves on a tree are all sort of equidistant from the trunk, and likely to be about equally related or unrelated to one another, since all the branches are sort of coming off the trunk the same way.

But that’s not what happened. Consider our mitrochondrial DNA lines. mtDNA is one way to trace ancestry. This has nothing to do with how much DNA is affected. It’s a way to trace antecedent groups.

If you start with mtDNA haplogroup L0 in africa, and get to the splitting off of L3 into M-N, you are roughly at the point of out of africa into the Levant, perhaps 50-70kya-ish. From there (broad picture; some exceptions for earlier exit groups yada yada) humans spread out to populate the world. Perhaps 30-40 kya the non-african ancestral group gets genes from a Neanderthal daddy. Other new variants appear (MCPH1 haplogroup D, e.g…that haplogroup happens to be named for Y chromosome markers, but doesn’t change the story). Out of africa, the gene variants that show up early in the Levant expansion (either ordinary evolution or archaic introgression) become an ancestral gene pool for the almost the entire out of africa group, and certainly most european and asian groups, who have M-N mtDNA lines as their ancestral group.

This is not the sub-saharan story. At L3/M-N, L3, L2 and L0 lines remain in africa, with all of the incredible diversity associated with those lines. But the gene variants which exist, and the new gene variants that arise, do so largely without exposure to out of africa gene variants post-Levant expansion.

That’s why I use that “M-N descendant lines” as shorthand, under the assumption that you are interested enough in the topic to get what I’m talking about.

To summarize: Most modern eurasians descended from the Levant population(s). Most modern sub-saharan africans descended from L0 without the Levant population being represented in their gene pool, because back flow did not occur on a large enough scale.

It’s easy to let wordsmithing around clinal populations, amount of SNP variations and so on mask the central concept here. When we self-identify into black, white and asian, we associate ourselves with very different average gene pools, on opposite sides of the out of africa splitting point.

Look at the diagram on page 614 here. The fat green lines are the main migrations and gene groups; the red crossbar lines show that there has been some gene flow between these large branches. Notice the splitting points, and the splitting point for europe, asia and the americas. Starting about 40kya, this is largely its own branch, with some backflow to NE africa in relatively modern times. Gene variants that arise among all of these branches flow mostly to descendant lines. A gene variant that arose in the Levant population 40 kya would flow to europe, asia and possibly the americas depending on the exact point where it arose. It might flow to NE africa. It would not be very represented for the rest of the african lines.

Notice also the language in the text: “sub-saharan populations…non-african populations…” You will see this language over and over again from population geneticists even when they take great pains to talk about quantity of diversity, no biology of race…etc etc.

Why do they use that language? Because “sub-saharan populations” and “non-african” populations are two different and diverged gene pools. And the modern populations they contain are nowhere near clinal with one another even though they may be clinal with their ancestral populations back to some original population (archaic introgression excepted for the M-M descendant lines).

This is why when we self-identify with “black,” “white” and “asian,” we self-identify with different gene pools even though there is no such thing as race and even though SNP variations run amok within all populations. And our migration history is the reason those self-identified genetic pools vary in average frequency for all sorts of gene variants.

522

Point of clarification: When I mention the Levant group, I am talking about the second group to reach there; not the original group >100kya, which is though to have died out.

See this popular Journey of Mankind graphic illustration.

523

You say they’re not clinal, but you haven’t provided evidence across the entire genome. Mitochondrial or Y DNA tells us about mother’s-mother’s-mother’s line, and father’s-father’s-father’s line, but that’s not “antecedent groups” – it’s just ancestors along the edges of the “family tree”. That’s only a very small portion of a group’s ancestry. Further, even this small portion of ancestry demonstrates that certain African groups (TODAY, not in the past) are more closely related to non-African groups (just in terms of this small portion of ancestry) then various other far-flung African groups.

Show me two neighboring populations, that aren’t separated by a huge geographic barrier (ocean, impassable mountains, etc.), that have lived there for a long, long time, that are not at all closely related across the entire genome (not just mitochondrial or Y DNA) compared to their relatives on opposite sides. Seriously – name these groups, if they exist. My understanding is they don’t exist – or at least that they don’t exist in the specific geographical area we’re talking about (Red Sea coast of Africa and Arabia and the Levant). That we could go from village to village from Khartoum to Baghdad, and except perhaps for places like parts of Israel with lots of recent immigrants, we would find that each village is pretty closely related across the entire gene pool to the next village along the way. If this isn’t true, where is the cutoff? Which village isn’t closely related to which neighbor?

524

First of all, I think you misunderstand mtDNA and Y DNA lineages. These are not references to the amount of DNA; they are markers on how to track ancestry of genetic pools or lineages. It is not the case that “Well; my mtDNA and YDNA might be very removed, but most of my DNA is from the same recent ancestry as yours.”

Second, I don’t think you understand what being clinal is all about. There is no way modern populations in sub-saharan africa are as clinal to eurasian populations as adjacent african or eurasian populations are to one another. If you read a little more about how the world was peopled, you will understand better why every population genetecist, after reassuring us about quanitity of diversity, happy single race and so on, lapses right back into talking about “sub-saharans” and “non-africans.” Why? Because those are two broad divisions genetically. Why? Because the out of africa migration was largely one way and largely isolated two broad groups for tens of thousands of years.

Maybe the graph here will help you, although I am beginning to lose hope.

Notice the sub-saharan lines in green. About 70kya we get the out of africa migration and a split at L3 into M and N lines. The M lines and their descendants are in blue and pink; at the bottom are the N lines and their descendants in blue, pink and red.

Now look at the groups we have today by focusing on the terminus of each line. Notice that the green circle at the far left representing haplogroup L0k shows a separation from any of the eurasion groups by more than 150,000 years. So populations where that haplogroup is highly represented (the Khoisan, for example) have a gene pool that is on average separated from any given eurasian population for tens of thousands of years.

If you look at the general grouping of sub-saharan pools from the post L3-M-N descendant lines, you have to go back 70,000 years to get to the confluence of pools (roughly the out of africa splitting point).

What this means is that those sub-saharan pools we see today are nowhere nearly as related (clinal) to a eurasian population as eurasians are to each other. Of course, for an individual population we choose to define, we can figure out how much of their gene pool represents how much separation from any other gene pool.

When we take the broad grouping that self-identification of black, white and asian puts us in, you can see we have to go back about 70kya to get those two broad groups to coalesce–and even then, we only pick up the L3s. To pick up the L0, L1 and L2 haplogroups in africa, we have to go back tens of thousands of years more to get to a coalescence point. So african groups whose gene pool has a predominance of those markers have genes which are way, way removed from eurasian lines.

525

Link to the site containing the graph just above.

My apologies for omitting it.

526

They’re “markers” that often disagree with each other. They’re just a small part of someone, or some group’s, genetic ancestry. Tracking someone’s (or some group’s) mitochondrial DNA tells you about mother’s mother’s mother’s mother (and so on), but nothing about the other ~99% of their ancestry.

“Largely” and “largely”, not “entirely” – not even close. Further, you’re just making assertions, and linking to data about mitochondrial and Y DNA without actually answering my questions about groups’ genetic ancestry in its entirety.

More assertions and data about mitochondrial or Y DNA without actual data about groups’ entire ancestry. Further, your link actually supports my point (which you continue to ignore) – there exist multiple African groups which are more closely related in terms of genetic ancestry (measured by mitochondrial DNA, at least) to Eurasian populations than to certain other African groups. How can it be possible to separate them into two larger conglomerations (say, A for Africans and E for Eurasians) when some groups in A are more closely related to E than to certain other groups in A?

Again, none of this answers my questions. If the relationship is not clinal, where on the map is or was the split? Where is the line across which two neighboring villages are not more closely related to each other than they are to villages much farther away? What boundary must be crossed such that nearly everyone in the new area is mostly unrelated to the previous one?

You’re giving me broad assertions and data about mitochondrial and Y DNA without making much of an effort to answer my questions or provide information about ancestry over the entire genome.

527

You are very wrong about this, but I understand your need to maintain an incorrect paradigm. The truth flies in the face of your idea that our gene pools are all sort of mixed together.

When we use mtDNA and Y chromosomes as markers for ancestry, we get very good correlation about roughly which populations were ancestral, how they migrated, and to what extent they were isolated. You won’t find the Khoisan’s L0 in east asians,. This is not because they share 99% of their other gene variants. It’s because that population has been separated by over a hundred thousand years, ancestrally. You won’t find very many post-split gene variants in non-descendant lines because post confluence points, the ancestral groups have been separated. That’s why the marker haplogroups become name-able haplogroups: separated gene pools. You’ll find a teeny bit of the post-split gene variants in both groups because there is always somebody boinking someone somehow. But as an overall percentage, these are widely separated pools. For the reasons I keep reiterating, there is a broad split at out of africa into “sub-saharan” and “non-africa” and this language is used all the time in the literature.

But if you want to make your own graphs and use your own crayons for coloring rainbows in all the circles, have at it. Pollyanna might buy it. You won’t sell it to the modern academic world studying population genetics. In the end we kind of like that image better, philosophically, even though evolution keeps forgetting not to diverge separated populations.

528

You’ve completely failed to support this. If I’m wrong, it’s not proven by any data and graphs about mtDNA and Y chromosomes. It would be proven by data about ancestry across the entire genome.

I’m not just going to take your word for it. If you want to make assertions about the entire gene pool, show me data about the entire gene pool – not just markers.

If you want to skip this side discussion, go right ahead. Even if you’re right, it provides no information whatsoever about the genes for intelligence among black people.

529

At roughly the L3/M-N split out of africa, 70kya, there is a broad separation of the human gene pool into “sub-saharan” and “non-african.”

See the nice graph I just gave you and look for the blue "M"middle left, and the pink “N” bottom left. These are the two descendant lines for most eurasians at the out of africa split.

The boundary that was crossed for the second group to make it into the Levant and become the founder group for most eurasians might be somewhere around the southern part of the Red Sea at the Gates of Grief. See the Journey of Mankind simplified video, and play it at least until you get to 85kya. “All non-(subsaharan) african people are descended from this group.” (This is slightly wrong, but as an average it’s fine.) That migration gate closed, and so two broad pools were created. Out of africa descendants (mostly M and N lines and their descendants), and sub-saharan lines (mostly L0-L3 and their descendants). And the migration gate kept the two pools from mixing to any significant extent.

Sarah Tishkoff puts it this way in the paper I gave you (with another graph on P 614):
“These observations indicate that populations in northeast Africa might have diverged from the rest of sub-Saharan Africa early in the history of modern African populations and that a subset of this northeast African population migrated out of Africa and populated the rest of the globe. Analysis of mtDNA and Ychromosome diversity support a single East African source of migration out of Africa.”

In other words, most peoples outside of africa came from a founding population that started in NE africa and made it out across the Red Sea (or other) geographic gate in a one-way direction.

If you look at that graph on page 614, the splitting point for groups out of africa is pretty obvious (Tishkoff labels it “Phase III: Migration out of africa.”) It should also be obvious that gene variants acquired in any branch flow mostly to descendant populations. For this reason by the time you get down to the bottom of those branches in the graph–to current populations–the current groups on the right represent an entirely different branch from the sub-saharans. Any gene variants acquired between that “Phase III Migration out of africa” and before subsequent branches will accrue only to the populations on the right for the most part. This is why, for instance, we see Neanderthal genes in non-african groups only (for the most part). That even introgressed genes somewhere around 40kya, after the out of africa split and before most of the M-N lines went their separate ways.

Hope this helps. Stop obsessing about “clinal” and begin thinking how you can only get gene variants from your ancestral pools plus new evolution. You can’t get them if they are not in your gene pool.

530

So, we know that traits like skin color are genetic. We also know that there are strong correlations between skin color and other physical features which are genetic. We also know that intelligence has a genetic component. But when it comes to the proposition that intelligence might be another trait that aligns with skin color, you claim—that’s just about impossible! Well, you might not go quite that far, but you place the burden of proof on those that hold that this is a very real possibility. Seems to me that a long as someone is not flatly asserting that blacks are —in fact—less intelligent that whites or asians, that the burden falls to you as to why this correlation, based on the admittedly weak evidence we have to look at is almost impossible. Or, at the very least, why it should be the default position. Mind you, it is my guess and desire that if all were revealed that we’d see that nurture would be able to explain any and all differences. But that does not mean that the other side should be saddled with in inappropriate burden of proof.

531

A claim you have totally failed to back up by citing any studies on Africans.

And something I doubt - if distinctively Levantine genetic material can make it all the way to pre-colonial South Africa (as in the Cohanim markers in the Lemba), any claims of severely limited backflow start to look a little silly. Did only their Jewish DNA make it south, not their Neanderthal DNA?

532

Bullshit, as has been repeatedly shown to you. And repeating the bullshit claim doesn’t make it any less bullshit.

There are no strong correlations between skin colour and other physical features. At least, none that you have ever shown.

533

You keep saying this, and you keep asserting that mtDNA is enough to say this. Because… just because. No – mtDNA is a tiny part of a person or a group’s ancestry. My dad has different mtDNA than me, but I’m much more closely related to him than to my great great grandmother with identical mtDNA. This is an interesting way to track maternal lines of ancestry, but it doesn’t track the ancestry of entire groups without further data. Again – I’m not just going to take your word for it. Show me data about the entire genomes of groups.

Huh? What strong correlations do we know exist between skin color and other physical features? Are you talking about things like sickle-cell trait? If so, that’s the wrong example – it is prevalent in West Africa, but not prevalent in many areas of Africa in which people have just as dark (or even darker) skin than West Africans. Plus, it also occurs in Mediterranean regions and tropical regions of Central and South America among populations with much lighter skin than various regions of Africa without a high prevalence of the trait.

I’m not placing any inappropriate burden of proof – I’m objecting to claims that it’s been proven! Do you seriously think that CP’s assertion that we have enough data right now to conclude without doubt that the remaining test score gap is explained entirely by genetics is reasonable?

And when have I said such an explanation is impossible? It’s definitely not impossible. But it’s far from the best explanation with the data we have – for one thing, we have no data on the actual genes involved or their prevalence, and for another, we have plenty of data about other test-score gaps that were explained by matters of society and culture and not genetics. Further, we have actual experimental evidence that refutes the genetic explanation, with no actual experimental evidence in support of the genetic explanation. This is why CP’s assertions and his own confidence is so ridiculous. That’s what I object to.

534

Bad example. The “Kohanim markers” are on the Y-chromosome, which is not subject to chromosomal crossover like the other chromosomes are, and so it’s quite likely that the tiny amount of Neanderthal DNA brought there over a thousand years ago would be mostly gone and hard to detect in the present population.

535

I’d hesitate to say it was likely or not for the Neanderthal DNA to be bred out - I mean, it wasn’t in other, Eurasian populations, was it? This is the difference between 40 000 years and 3000.

The only way to tell, would be to check the group. Something which, as I said, hasn’t been done.

I don’t expect the Lemba to have as high levels of the DNA in question, nowhere near it, but I do expect them to have non-zero, detectable levels, something CP seems to think impossible. I mention the Kohanim markers only to show there’s proof of quite a bit of backflow (the Southern Europe markers in the Khoisan genome have also been mentioned) into Africa.

536

But most Europeans have been interbreeding with other Europeans, most all of whom have some Neanderthal DNA. The Lemba have been interbreeding with other Africans, most of whom have little to none. Neanderthal DNA.

If Vikings had bred with a few Native Americans 1,000 years ago, we might find some European Y-DNA in “pure-bred” Native Americans, even if they had no other measurable European genetic markers since those would have lost during the centuries of chromosomal crossover. You likely have few, if any, genes that came directly from any one of your 30th great grandparents (other than Y-Chromosome and mt-DNA genes). That is to say that the % of your genes coming from any given 30th great grand parent is vanishingly small.

537

Ordinarily, yes, but the Lemba (like some Jewish groups) have a large amount of inbreeding by being a cultural isolate.

My point being, we can speculate, but we don’t know, because no-one’s checked.

538

I think asking why they don’t show appreciable Neanderthal lineage is like asking why they don’t generally have a skin tone like Barack Obama. One is half European in lineage and the others are some tiny fraction Near Eastern. In that tiny fraction, the Neanderthal DNA would likely be lost, or at “background noise” level.

539

What is wrong with this ad?:

"Wanted, one entire race to fill 2 or 3 billion positions with up-and coming company. Applications will be received from the entire Caucasian, Negroid and Mongoloid races. Each race is required to submit its race-wide IQ scores. Successful applicant will be the 2 or 3 billion people in race with highest score.

I have seen the arguments in this thread over and over for the past 50 or 60 years, and I always wonder what the point is. What practical application would there be to determining that one race is more intelligent? Obviously, nobody is saying that every single member of race A is smarter than every single member of race B.

I have had occasion to hire people in the course of my career. If I have a black person and a white person submitting applications, race “averages” are completely useless in helping me decide who is the best candidate. The black guy could be an idiotic underachiever, and the white guy could be a super-smart worker, or it may be the other way around.

Quite simply, who do you know who has had occasion to hire all the members of one entire race instead of another?

540

How long have we been at this? Ten years? Five? Anyway, little has changed with the exception of Unz and his research assistants discovery that the strong genetic hypothesis based on standardized test scores just doesn’t work. (Unless the Irish and several other European groups underwent some sudden, Bruce Banner like genetic mutation. That’s right, something that could only happen in a comic book.)

Anyway,