Michael Blowhard writes:

Dear Blowhards --

It's day four of our week with Gregory Cochran, celebrating the startling and exciting new book that Greg has co-written with Henry Harpending, "The 10,000 Year Explosion: How Civilization Accelerated Human Evolution." (Buy the book here; explore the book's excellent website here.)

On Monday we talked about the fact that human evolution didn't -- as we were once told -- come to a screeching halt 40,000 years ago. Tuesday's topic was culture and its impact on evolution. On Wednesday, we discussed Cochran and Harpending's contention that not only has human evolution continued during the last 10,000 years, it has sped up considerably. Today: Did Neanderthals and modern humans interbreed?

***

A Q&A With Gregory Cochran, Part Four

2Blowhards: Your contention is that 1) modern humans and Neanderthals may well have interbred. 2) Any beneficial alleles we picked up from these interbreedings would have spread throughout the population even if the instances of interbreeding were few. 3) These developments may help explain the cultural explosion of 40,000ish years ago. Is that a fair summary?

Gregory Cochran: Yes.

2B: How should readers take this contention? As a hypothesis? A provocation?

GC: A hypothesis that we think probable. But there are unlikely circumstances that might have blocked such gene flow, so we could be wrong.

2B: What kinds of evidence is there so far for interbreeding with Neanderthals?

GC: There are a lot of paleontologists who think that the skeletal record suggests some interbreeding. In addition, there are a few genes with strange patterns that might have parachuted in from another hominid species such as Neanderthals.

2B: How and why did you come to this hunch?

GC: Before I learned some of the relevant theory in population genetics, playing with simulations left me aware of the fact that even a few copies of a better version of a gene had a good chance of becoming common. Turns out that there was an exact result that shows just how likely this is: for a single copy, the chance of ultimate success is twice the gene's advantage. For example, one copy of a gene with a 1% advantage has a 2% chance of becoming universal in the population.

2B: Why do you think that the possibility of interbreeding is a useful idea to float?

GC: Well, if it is correct, we'll understand a bit more about prehistory and the development of modern cultural capabilities.

2B: Where might evidence for interbreeding pop up?

GC: Out of the current work on sequencing the Neanderthal genome. With a little luck, we might also find some better skeletal evidence for interbreeding.

2B: What kind of evidence would nail it down one way or the other once and for all?

GC: Neanderthal DNA. For example, if we found that Neanderthals had a particular favorable version of a gene before we did, a version that was quite different from the version humans used to have (and that some people might still have), we'd have to think it originated in Neanderthals.

2B: Might it be useful to sketch in the current picture scientists have of how modern humans spread? My impression is that the old argument was: One out-of-Africa event vs. multiregional. The new picture is: numerous out-of-Africa migrations, with backflows and overlappings. Is that a semi-useful picture?

GC: Modern humans appeared in both Australia and Anatolia about 45,000 years ago, according to the best archeological info we have. Creatures that look a lot like us had been in East Africa and in and out of the Levant for tens of thousands of years, but some important change occurred that finally let them expand out of Africa and displace archaic humans like Neanderthals and the evolved erectus types of East Asia. This expansion had a northern arm into Europe and central Asia and a southern arm, around the Indian Ocean, to Australia and Oceania. This process ought to have led to lots of genetic differences since most of the population growth occurs from colonizers and founding populations at the leading edge of a wave of advance. There’s no reason to think that populations stayed small after colonization though: in fact, data from mitochondrial DNA work suggests growth in all but some Arctic populations, and among Amerindians after they managed to get south of the ice. There may also have been other local expansions and replacements of peoples: we know this sometimes happened in the last few thousand years and it may well have happened back in deeper prehistory as well.

2B: Before reading your book I hadn't been aware of how quickly beneficial alleles can sweep through entire (or nearly-entire) populations. Is there a simple, short, plain-English way of explaining this phenomenon?

GC: If a particular version of a gene in one way or another increases the number of children of carriers, it gradually increases in number. Imagine that a mutation creates an allele that confers a 5% edge. It might disappear by chance, but with some luck, it continues to exist and tends to increase. By the time there are 100 copies, it is unlikely to disappear by chance, and tends to grow in a fairly predictable manner. In a thousand years, it increases by a factor of 7. In 7 thousand years, it's ~800,000 times more common, at least in this simple model of a well-mixed population. In a more realistic model, one based on the fact that people marry near neighbors, it spreads out as a wave from the place it originated [Fisher-Kolomogorov equation].  If the average person is born 20 miles from the birthplaces of his parents, such a gene would spread about 1800 miles in 7000 years. In real life this spatial spread seems to occur more rapidly, perhaps driven by the occasional trader, colonist, or invading army.

I know that this is somehow unintuitive, but it's a lot like what happened when someone released 24 rabbits in Australia. Obviously 24 rabbits were tremendously outnumbered by native Australian animals like kangaroos and wombats, but rabbits reproduced rapidly. Assume that rabbits could increase their numbers by a factor of four in one year -- then, starting with 23, you have ~20,000 rabbits after five years and ~25 million after 10 years. That is roughly what happened. Or, there were originally about 7200 French settlers in Quebec. Today they have six million descendants -- compound interest can do this in a few centuries.

2B: How might interbreeding-with-Neanderthals help account for the cultural explosion of circa 40,000 years ago? What is it about that explosion that needs explaining? And is it true that in many parts of the world such an explosion didn't occur?

GC: If the Neanderthals happened to have some nervous-system genes that had an advantage (even if Neanderthals were uncompetitive on the whole), we could have gotten a significant cognitive boost out of such interbreeding. It's called an explosion because of the (apparently) fairly abrupt appearance of real art, more complex tools, etc. I say apparently because the archaeological record is far from complete. The complex stuff doesn't seem to have arrived everywhere at the same time: stone tool techniques in Australia were similar to those of Neanderthals until a few thousand years ago.

2B: Neanderthals didn't range all over the world. That means that not all modern-human population groups would have benefited from the consequences of interbreeding with them. Australian aborigines, for instance, would be unlikely to be carrying any Neanderthal alleles in them at all, where Europeans would be very likely to be carrying Neanderthal alleles. Fair and reasonable?

GC: It happened a long time ago -- really advantageous genes might have had time to spread everywhere. But some might not have made to Australia -- it's possible. In addition, some Neanderthal genes that were useful in northern climates might not have been useful in the tropics and thus would not have spread there.

At any rate we expect that any advantageous genes acquired from Neanderthals are likely to have reached Australia about 4,000 years ago when the number of archaeological sites suddenly increased by a factor of ten, stone tools became more complex, and the dingo arrived. These events may be associated with the great Austronesian expansion at that time, which may also have contributed to the contemporary Australian gene pool.

2B: If "when did your population group encounter agriculture" is one of the major ways in which population groups diverge, could "exposure to Neanderthal alleles" be another?

GC: If we had some populations that could be shown to have some Neanderthal gene flow (more than other populations) but never adopted agriculture, we could answer that question. But there are no hunter-gatherers left in Europe or the Middle East.

2B: There were other human-esque creatures out there around the world too, no? Is it a possibility that other population groups interbred with the human-esque creatures they shared territory with in the same way that western/European/etc humans may have interbred with Neanderthals?

GC: Yes, we mention them briefly in the book. There were archaic humans in East and Southeast Asia, generally called 'evolved erectus'. In general, the story would be similar: interbreeding looks probable. But we didn’t talk much about them, because we know a lot less about them than about Neanderthals.

2B: I imagine that the interbreeding-with-Neanderthals hypothesis will be the most controversial thing in the book. Do you too?

GC: Probably not controversial at all since Neanderthals have no lobby. So far I haven't seen anyone getting all het up about it. In fact, for a least a few people, it's the least controversial thing we have to say. Now when we clone a few Neanderthals and the Europeans let them run casinos (in compensation for exterminating them), things may change.

As we mention in the book, there's good evidence that our ancestors (about 30 million years ago) picked up a working gene from a virus. Compared to that, a Neanderthal in the family tree is nothing special.

***

Henry Harpending's faculty page is here. Here's the Unofficial Gregory Cochran Site. Harpending and Cochran often make appareances at GNXP.

Here's a site that the authors made for their book. Be sure to visit it; the website provides an excellent overview of the book's contents and includes some well-worth-reading outtakes. Buy the terrific book here.

Some of the buzz around the web:

• Randall Parker takes a brainy look at the authors' thesis
  
• Tyler Cowen didn't love the book
  
• Steve Hsu gives the book a technical inspection and likes what he sees. Steve also supplies some excellent linkage
  
• Congenial Times found it provocative (and supplies a good summary)
  
• John Derbyshire is brilliant and appreciative
  
• A couple of rowdy discussions are underway at GNXP: here and here
  
• ShrinkWrapped thinks that "The 10,000-Year Explosion" is "a dangerous book" -- but in a good way
  

Back here and here, I interviewed Greg Cochran about the Iraq War.

Greg promises to respond to a selection of your comments and questions tomorrow, which will be a treat for all of us. Since he'll need a little time to do that, if you want your question or comment to be considered as a candidate, please leave it by midday today. And, everyone, be sure to return tomorrow.

Best,

Michael

I'd like to know how Dr. Cochran sees the relationship between his theory and the theory of race submitted by Rushton.

Posted by: LemmusLemmus on January 29, 2009 7:51 AM

What is GC view on "Idiocracy"? I.e. the recent phenomenon where rich countries' economic surplus, created by the successful evolutionary adaptation of their population acts nowdays to yield a sizable demographic surplus of gene-carriers of the less successfull adapters.

Posted by: zeev on January 29, 2009 11:05 AM

Just a quick thought: The usual example for the "evolution stopped 50,000 years ago" argument is warm clothing -- you don't need to grow fur, because you've invented clothes. The thing is, the argument is in fact valid FOR THAT PARTICULAR EXAMPLE! Warm clothing does indeed give people a microclimate much like the one they first evolved with in Africa, so we really don't need fur.

But most cultural changes are not like that at all. Learning to farm makes it easier to survive, because it provides you with a large and fairly steady supply of food, but it in no way resets anyone's personal environment back to any sort of hunter-gatherer norm. So my point is that, in explaining your ideas to people, it might be useful to point out the anomalousness of the usual warm clothing example.

Posted by: JohnB on January 29, 2009 12:23 PM

Are there any archeological sites in Australia that predate the dingo (thus presumably predating the possible spread of Neanderthal genes), that have a chance of yielding sequencable human DNA? Such direct evidence--if it exists--would seem to be helpful in evaluating that portion of the book's theory.

Posted by: AMac on January 29, 2009 12:47 PM

I hope that this is not an ignorant question, but when Mr. Cochran says that a gene confers "a 5% advantage", is he referring to 5% more descendants, 5% better survival chance, or some combination perhaps? Or does the percentage not translate directly, and if not, what does it mean?

Posted by: Dennis Mangan on January 29, 2009 1:20 PM

Did Neanderthals mate proto erectus lineages? Specimen one: George W, Bush. Q.E.D.

Posted by: Lisa on January 29, 2009 2:37 PM

Dennis,
It means 5% more descendants. Surviving is part of fitness, of course, but only because it usually helps increase one's number of descendants.

Posted by: Eric J. Johnson on January 29, 2009 3:11 PM

Any guesses about the relationship between "evolved erectus" and modern Asian populations?

How many full sequences will we need to get a good genome->IQ mapping? Do you think it's more like 10^4, 10^5, 10^6 or above?

And are there any archaic hominids that would make good pets? If so, which?

Posted by: Mencius on January 29, 2009 11:05 PM

I think AMac hinted at this on earlier thread. In the book, C&H say that people have a high miscarriage rate from lots of driving genes causing chromosomal abnormalities. Dr. Harpending says that BRCAI and Melanisian Oocytosis homozygotes are nonviable.

Is it plausible that lots of alleles have heterozygote advantage, but we don't see homozygotes? It could even be an adaptation: losing a pregnancy has a much lower fitness cost than losing a child.

SSADH, an enzyme that destroys GABA, has 2 common alleles, in studies, the T version reduces cognitive ability(1) and lifespan. The alleles aren't in Hardy-Weinberg eq. in the families Plomin studied(TT frequency lower that predicted)

Is it possible that a big fraction of relatively recent "overclocking" from heterozygote advantage is masked by spontaneous abortion? In this scenario, a trait can be increasing in frequency even though lower-IQ people have more babies on average? Heavy selection of alleles like this would lead also lead to our anomolous high miscarriage rate. Possibly we could see evidence of the strength of relative selection in miscarriage rates between different populations.

Link

Link

Posted by: rob on January 30, 2009 8:43 AM

Michael, have you considered adding a 'your comment has been submitted and is awaiting approval' screen? I never know if I have, so I hit post over and over, if you don't see repeats, no biggie to me.

Dr. Cochran, is there any pigmentation change in albinism carriers? Southern Africa has relatively low UV, and the Bushmen are pretty light-skinned. Is it plausible that Southern African Bantus are depigmenting from whatever selective advantage led to light skin other places?

Posted by: rob on January 30, 2009 10:12 AM

Hey Rob -- It does sometimes take us a little while to approve and post comments. A pain for everyone but it seems to be our best defence against spam comments. But you do get an indication that comments won't post instantly. Hit "Post" and these words come up:

"Thank you for commenting.

Your comment has been received and held for approval by the blog owner."

Posted by: Michael Blowhard on January 30, 2009 10:38 AM

"But there are no hunter-gatherers left in Europe or the Middle East."

Not the Lapps?

Posted by: Chris B. on January 30, 2009 12:22 PM

To clarify, I know that the Lapps are/were herders, not hunter-gatherers, but AFAIK they were never farmers.

Posted by: Chris B. on January 30, 2009 12:29 PM

Pupu bought the book last night and read a half of it. It is riveting!

Pupu seconds Michael's assessment on the interbreeding-with-Neanderthals hypothesis, and finds it potentially controversial. This is not because the theory is unconvincing but because the evidence indicating the said event has both happened and left a trace on the human footprint is lacking. What may be even more controversial is the connection between this hypothesis and the event of "out of Africa."

The mathematics of the theory is so elegant and powerful, which potentially makes pointing at one specific event as its manifestation more controversial. The question is why it has to be the Neanderthals. We do not know.

Posted by: Pupu on January 30, 2009 1:47 PM

Michael, so it does. I guess I was too impatient.

Posted by: rob on January 30, 2009 9:07 PM






Post a comment

Name:


Email Address:


URL:


Comments:


Remember your info?