Local Optima, Diversity, and Patchwork

Local optima–or optimums, if you prefer–are an illusion created by distance. A man standing on the hilltop at (approximately) X=2 may see land sloping downward all around himself and think that he is at the highest point on the graph.

But hand him a telescope, and he discovers that the fellow standing on the hilltop at X=4 is even higher than he is. And hand the fellow at X=4 a telescope, and he’ll discover that X=6 is even higher.

A global optimum is the best possible way of doing something; a local optimum can look like a global optimum because all of the other, similar ways of doing the same thing are worse. To get from a local optimum to a global optimum, you might have to endure a significant trough of things going worse before you reach your destination. (Those troughs would be the points X=3.03 and X=5.02 on the graph.) If the troughs are short and shallow enough, people can accidentally power their way through. If long and deep enough, people get stuck.

The introduction of new technology, exposure to another culture’s solutions, or even random chance can expose a local optimum and propel a group to cross that trough.

For example, back in 1400, Europeans were perfectly happy to get their Chinese silks, spices, and porcelains via the overland Silk Road. But with the fall of Constantinople to the Turks in 1453, the Silk Road became more fragmented and difficult (ie dangerous, ie expensive) to travel. The increased cost of the normal road prompted Europeans to start exploring other, less immediately profitable trade routes–like the possibility of sailing clear around the world, via the ocean, to the other side of China.

Without the eastern trade routes first diminishing in profitability, it wouldn’t have been economically viable to explore and develop the western routes. (With the discovery of the Americas, in the process, a happy accident.)

West Hunter (Greg Cochran) writes frequently about local optima; here’s an excerpt on plant domestication:

The reason that a few crops account for the great preponderance of modern agriculture is that a bird in the hand – an already-domesticated, already- optimized crop – feeds your family/makes money right now, while a potentially useful yet undomesticated crop doesn’t. One successful domestication tends to inhibit others that could flourish in the same niche. Several crops were domesticated in the eastern United States, but with the advent of maize and beans ( from Mesoamerica) most were abandoned. Maybe if those Amerindians had continued to selectively breed sumpweed for a few thousand years, it could have been a contender: but nobody is quite that stubborn.

Teosinte was an unpromising weed: it’s hard to see why anyone bothered to try to domesticate it, and it took a long time to turn it into something like modern maize. If someone had brought wheat to Mexico six thousand years ago, likely the locals would have dropped maize like a hot potato. But maize ultimately had advantages: it’s a C4 plant, while wheat is C3: maize yields can be much higher.

Teosinte is the ancestor of modern corn. Cochran’s point is that in the domestication game, wheat is a local optimum; given the wild ancestors of wheat and corn, you’d develop a better, more nutritious variety of wheat first and probably just abandon the corn. But if you didn’t have wheat and you just had corn, you’d keep at the corn–and in the end, get an even better plant.

(Of course, corn is a success story; plenty of people domesticated plants that actually weren’t very good just because that’s what they happened to have.)

Japan in 1850 was a culturally rich, pre-industrial, feudal society with a strong isolationist stance. In 1853, the Japanese discovered that the rest of the world’s industrial, military technology was now sufficiently advanced to pose a serious threat to Japanese sovereignty. Things immediately degenerated, culminating in the Boshin War (civil war, 1868-9,) but with the Meiji Restoration Japan embarked on an industrialization crash-course. By 1895, Japan had kicked China’s butt in the First Sino-Japanese War and the Japanese population doubled–after holding steady for centuries–between 1873 and 1935. (From 35 to 70 million people.) By the 1930s, Japan was one of the world’s most formidable industrial powers, and today it remains an economic and technological powerhouse.

Clearly the Japanese people, in 1850, contained the untapped ability to build a much more complex and advanced society than the one they had, and it did not take much exposure to the outside world to precipitate a total economic and technological revolution.

Sequoyah’s syllabary, showing script and print forms

A similar case occurred in 1821 when Sequoyah, a Cherokee man, invented his own syllabary (syllable-based alphabet) after observing American soldiers reading letters. The Cherokee quickly adopted Sequoyah’s writing system–by 1825, the majority of Cherokee were literate and the Cherokee had their own printing industry. Interestingly, although some of the Cherokee letters look like Latin, Greek, or Cyrillic letters, there is no correspondence in sound, because Sequoyah could not read English. He developed his entire syllabary after simply being exposed to the idea of writing.

The idea of literacy has occurred independently only a few times in human history; the vast majority of people picked up alphabets from someone else. Our Alphabet comes from the Latins who got it from the Greeks who adopted it from the Phoenicians who got it from some proto-canaanite script writers, and even then literacy spread pretty slowly. The Cherokee, while not as technologically advanced as Europeans at the time, were already a nice agricultural society and clearly possessed the ability to become literate as soon as they were exposed to the idea.

When I walk around our cities, I often think about what their ruins will look like to explorers in a thousand years
We also pass a ruin of what once must have been a grand building. The walls are marked with logos from a Belgian University. This must have once been some scientific study centre of sorts.”

By contrast, there are many cases of people being exposed to or given a new technology but completely lacking the ability to functionally adopt, improve, or maintain it. The Democratic Republic of the Congo, for example, is full of ruined colonial-era buildings and roads built by outsiders that the locals haven’t maintained. Without the Belgians, the infrastructure has crumbled.

Likewise, contact between Europeans and groups like the Australian Aboriginees did not result in the Aboriginees adopting European technology nor a new and improved fusion of Aboriginee and European tech, but in total disaster for the Aboriginees. While the Japanese consistently top the charts in educational attainment, Aboriginee communities are still struggling with low literacy rates, high dropout rates, and low employment–the modern industrial economy, in short, has not been kind to them.

Along a completely different evolutionary pathway, cephalopods–squids, octopuses, and their tentacled ilk–are the world’s smartest invertebrates. This is pretty amazing, given that their nearest cousins are snails and clams. Yet cephalopod intelligence only goes so far. No one knows (yet) just how smart cephalopods are–squids in particular are difficult to work with in captivity because they are active hunter/swimmers and need a lot more space than the average aquarium can devote–but their brain power appears to be on the order of a dog’s.

After millions of years of evolution, cephalopods may represent the best nature can do–with an invertebrate. Throw in a backbone, and an animal can get a whole lot smarter.

And in chemistry, activation energy is the amount of energy you have to put into a chemical system before a reaction can begin. Stable chemical systems essentially exist at local optima, and it can require the input of quite a lot of energy before you get any action out of them. For atoms, iron is the global–should we say universal?–optimum, beyond which reactions are endothermic rather than exothermic. In other words, nuclear fusion at the core of the sun ends with iron; elements heavier than iron can only be produced when stars explode.

So what do local optima have to do with diversity?

The current vogue for diversity (“Diversity is our greatest strength”) suggests that we can reach global optima faster by simply smushing everyone together and letting them compare notes. Scroll back to the Japanese case. Edo Japan had a nice culture, but it was also beset by frequent famines. Meiji Japan doubled its population. Giving everyone, right now, the same technology and culture would bring everyone up to the same level.

But you can’t tell from within if you are at a local or global optimum. That’s how they work. The Indians likely would have never developed corn had they been exposed to wheat early on, and subsequently Europeans would have never gotten to adopt corn, either. Good ideas can take a long time to refine and develop. Cultures can improve rapidly–even dramatically–by adopting each other’s good ideas, but they also need their own space and time to pursue their own paths, so that good but slowly developing ideas aren’t lost.

Which gets us back to Patchwork.


Recent Discoveries in Human Evolution: H. Sapiens 300,000 years old?

Welcome back to our discussion of recent exciting advances in our knowledge of human evolution:

  • Ancient hominins in the US?
  • Homo naledi
  • Homo flores
  • Humans evolved in Europe?
  • In two days, first H Sap was pushed back to 260,000 years,
  • then to 300,000 years!
  • Bell beaker paper

As we’ve been discussing for the past couple of weeks, the exact dividing line between “human” and “non-human” isn’t always hard and fast. The very first Homo species, such as Homo habilis, undoubtedly had more in common with its immediate Australopithecine ancestors than with today’s modern humans, 3 million years later, but that doesn’t mean these dividing lines are meaningless. Homo sapiens and Homo neandethalensis, while considered different species, interbred and produced fertile offspring (most non-Africans have 3-5% Neanderthal DNA as a result of these pairings;) by contrast, humans and chimps cannot produce fertile offspring, because humans and chimps have a different number of chromosomes. The genetic distance between the two groups is just too far.

Oldowan tool

The grouping of ancient individuals into Homo or not-Homo, Erectus or Habilis, Sapiens or not, is partly based on physical morphology–what they looked like, how they moved–and partly based on culture, such as the ability to make tools or control fire. While australopithecines made some stone tools (and chimps can make tools out of twigs to retrieve tasty termites from nests,) Homo habilis (“handy man”) was the first to master the art and produce large numbers of more sophisticated tools for different purposes, such as this Oldowan chopper.

But we also group species based on moral or political beliefs–scientists generally believe it would be immoral to say that different modern human groups belong to different species, and so the date when Homo ergaster transforms into Homo sapiens is dependent on the date when the most divergent human groups alive today split apart–no one wants to come up with a finding that will get trumpeted in media as “Scientists Prove Pygmies aren’t Human!” (Pygmies already have enough problems, what with their immediate neighbors actually thinking they aren’t human and using their organs for magic rituals.)

(Of course they would still be Human even if they part of an ancient lineage.)

But if an ecologically-minded space alien arrived on earth back in 1490 and was charged with documenting terrestrial species, it might easily decide–based on morphology, culture, and physical distribution–that there were several different Homo “species” which all deserve to be preserved.

But we are not space aliens, and we have the concerns of our own day.

So when a paper was published last year on archaic admixture in Pygmies and the Pygmy/Bushmen/everyone else split, West Hunter noted the authors used a fast–but discredited–estimate of mutation rate to avoid the claim that Pygmies split off 300,000 years ago, 100,000 years before the emergence of Homo sapiens:

There are a couple of recent papers on introgression from some quite divergent archaic population into Pygmies ( this also looks to be the case with Bushmen). Among other things, one of those papers discussed the time of the split between African farmers (Bantu) and Pygmies, as determined from whole-genome analysis and the mutation rate. They preferred to use the once-fashionable rate of 2.5 x 10-8 per-site per-generation (based on nothing), instead of the new pedigree-based estimate of about 1.2 x 10-8 (based on sequencing parents and child: new stuff in the kid is mutation). The old fast rate indicates that the split between Neanderthals and modern humans is much more recent than the age of early Neanderthal-looking skeletons, while the new slow rate fits the fossil record – so what’s to like about the fast rate? Thing is, using the slow rate, the split time between Pygmies and Bantu is ~300k years ago – long before any archaeological sign of behavioral modernity (however you define it) and well before the first known fossils of AMH (although that shouldn’t bother anyone, considering the raggedness of the fossil record).

This was a good catch. (Here is the relevant Dienekes article, plus Model-based analyses of whole-genome data reveal a complex evolutionary history involving archaic introgression in Central African Pygmies, and Whole-genome sequence analyses of Western Central African Pygmy hunter-gatherers reveal a complex demographic history and identify candidate genes under positive natural selection.) If the slow mutation rate matches the fossil record better than the fast, why use the fast–except if the fast gives you inconvenient results?

But now we have another finding, based on the Bushmen, which also pushes the Bushmen/everyone else split back further than 200,000 years–from BioRxiv, “Ancient genomes from southern Africa pushes modern human divergence beyond 260,000 years ago“:

Southern Africa is consistently placed as one of the potential regions for the evolution of Homo sapiens. To examine the region’s human prehistory prior to the arrival of migrants from East and West Africa or Eurasia in the last 1,700 years, we generated and analyzed genome sequence data from seven ancient individuals from KwaZulu-Natal, South Africa. Three Stone Age hunter-gatherers date to ~2,000 years ago, and we show that they were related to current-day southern San groups such as the Karretjie People. Four Iron Age farmers (300-500 years old) have genetic signatures similar to present day Bantu-speakers. The genome sequence (13x coverage) of a juvenile boy from Ballito Bay, who lived ~2,000 years ago, demonstrates that southern African Stone Age hunter-gatherers were not impacted by recent admixture; however, we estimate that all modern-day Khoekhoe and San groups have been influenced by 9-22% genetic admixture from East African/Eurasian pastoralist groups arriving >1,000 years ago, including the Ju|’hoansi San, previously thought to have very low levels of admixture. Using traditional and new approaches, we estimate the population divergence time between the Ballito Bay boy and other groups to beyond 260,000 years ago.

260,000 years! Looks like West Hunter was correct, and we should be looking at the earlier Pygmy divergence date, too.

Two days later, a paper from the opposite end of Africa appeared in Nature which–potentially–pushes H sapiens’s emergence to 300,000 years ago, “New fossils from Jebel Irhoud, Morocco and the pan-African origin of Homo sapiens“:

Fossil evidence points to an African origin of Homo sapiens from a group called either H. heidelbergensis or H. rhodesiensis. However, the exact place and time of emergence of H. sapiens remain obscure … In particular, it is unclear whether the present day ‘modern’ morphology rapidly emerged approximately 200 thousand years ago (ka) among earlier representatives of H. sapiens1 or evolved gradually over the last 400 thousand years2. Here we report newly discovered human fossils from Jebel Irhoud, Morocco, and interpret the affinities of the hominins from this site with other archaic and recent human groups. We identified a mosaic of features including facial, mandibular and dental morphology that aligns the Jebel Irhoud material with early or recent anatomically modern humans and more primitive neurocranial and endocranial morphology. In combination with an age of 315 ± 34 thousand years (as determined by thermoluminescence dating)3, this evidence makes Jebel Irhoud the oldest and richest African Middle Stone Age hominin site that documents early stages of the H. sapiens clade in which key features of modern morphology were established.

Comparison of the skulls of a Jebel Irhoud human (left) and a modern human (right) (NHM London)

Hublin–one of the study’s coauthors–notes that between 330,000 and 300,000 years ago, the Sahara was green and animals could range freely across it.

While the Moroccan fossils do look like modern H sapiens, they also still look a lot like pre-sapiens, and the matter is still up for debate. Paleoanthropologist Chris Stringer suggests that we should consider all of our ancestors after the Neanderthals split off to be Homo sapiens, which would make our species 500,000 years old. Others would undoubtedly prefer to use a more recent date, arguing that the physical and cultural differences between 500,000 year old humans and today’s people are too large to consider them one species.

According to the Atlantic:

[The Jebel Irhoud] people had very similar faces to today’s humans, albeit with slightly more prominent brows. But the backs of their heads were very different. Our skulls are rounded globes, but theirs were lower on the top and longer at the back. If you saw them face on, they could pass for a modern human. But they turned around, you’d be looking at a skull that’s closer to extinct hominids like Homo erectus. “Today, you wouldn’t be able to find anyone with a braincase that shape,” says Gunz.

Their brains, though already as large as ours, must also have been shaped differently. It seems that the size of the human brain had already been finalized 300,000 years ago, but its structure—and perhaps its abilities—were fine-tuned over the subsequent millennia of evolution.

No matter how we split it, these are exciting days in the field!

So Why do Native Americans have so much Neanderthal DNA?

Warning: highly speculative post ahead.

As I mentioned the other day:

Worldwide distribution of B006, (from Yotova et al. “An X-Linked Haplotype of Neandertal Origin Is Present Among All Non-African Populations,” Mol. Biol. Evol. 28 (7), 2011).
Worldwide distribution of B006, (from Yotova et al. “An X-Linked Haplotype of Neandertal Origin Is Present Among All Non-African Populations,” Mol. Biol. Evol. 28 (7), 2011).
SNP PCA from Skoglund & Jakobsson’s “Archaic Human Ancestry in East Asia” (2011)
SNP PCA from Skoglund & Jakobsson’s “Archaic Human Ancestry in East Asia” (2011)

(Please note that Africans do not have chimpanzee admixture, despite the labeling on the graph–no human group has chimp admixture, because chimps and humans have different #s of chromosomes, so even if you could get a successful cross, the resulting child would be infertile, like a mule. I assume the point of the chimp node is just to represent that which has neither Neanderthal nor Denisovan admixture, though of course there is the possibility of some other form of archaic hominin admixture in Africans.)


So, Native Americans appear to have a ton of Neanderthal DNA. (Relatively speaking.)


  1. It’s all measurement error/convergent evolution/something else other than archaic admixture.

As much as I hate to say it, I still consider this very likely. There is just a ton of stuff that we don’t about the Americas–like how and when people first got here. I’m sticking here with what I think are the most scientifically-supported theories, but a lot of this is still quite disputed. In particular, all of this genetic admixture business is still kind of speculative, and when people start talking about finding admixture in the admixture, either life is totally awesome, or we’re trying too hard.

2. Survival at the Fringes theory

A lot of people seem to look at this data and respond with something like, “But Neanderthals are from Europe, not America!” But this is not a big issue; the Indians are descended from people who passed through Neanderthal-inhabited regions (the Middle East), just like everyone else with Neanderthal DNA. The migration to the Americas took place long after they acquired Neanderthal admixture.

But this doesn’t explain why they have so much of it.

My “concentration on the edge” theory states that when one population is displaced by another population, you end up with a “fringe” of the original population’s traits. Sometimes this fringe results in isolated groups, as the invading population completely surrounds or cuts off a remnant population from their former range.

The Ainu, for example, resemble certain other Oceanin groups, but not their neighbors, the Japanese. I’m speculating here, so don’t take my word for it.

But I have a much better case with the distribution of red hair: Frequency-of-red-hair-in-Europe

(So far I have found nothing explaining that dot over in Russia.)

Red hair is highly associated with the so-called Celtic fringe. It looks like it’s highly concentrated in Wales, Scotland, and parts of Ireland, but since I know a little history, and I know these aren’t areas of concentration, but just the areas that managed to escape being displaced by Anglo-Saxon invaders, just by virtue of being further away from the south-east coast of Britain.

One can imagine that the isolated dot in the middle of Russia might, at one time, have been connected with the other red-haired regions before other peoples invaded the lands between them and cut them off.

Compare to the map of blond hair:


Blond hair looks like it has been spreading steadily outward from a central source.

So what does this have to do with Neanderthal admixture in Native Americans?

It means that I think the Native Americans may have closer to original levels of Neanderthal admixture, while people in Europe and Asia have lower admixture because they mixed with later waves of people who came from Africa and had no Neanderthal admixture.

3. The Bering Strait selected for Neanderthal admixture

Alternatively, the harsh conditions of life in the Bering Strait–where some scientists think the ancestors of today’s Indians hung out (or “paused”) for thousands of years before the ice sheets opened up to let them through–may have selected for winter adaptations that came from ice-age Neanderthals. I have speculated previously that Type-2 Diabetes may actually be a winter adaptation picked up from Neanderthal DNA; now scientists think this DNA may be responsible for high rates of Type-2 Diabetes in Native Americans.

4. Western diseases selected for Western immune responses

One of the interesting things about the Neanderthal DNA hanging around in people is that it appears to code for certain immune responses. West Hunter recently had a great post (TLRs, PAMPs, and Alley Oop) detailing how they work, but for our purposes, “provide immunity” is sufficient. Austin Whittall suggests that back when smallpox, influenza, measles, and all of the other Western diseases tore through the Native Americans, killing about 90% of them, the guys who had more Neanderthal DNA were more likely to survive because they were more likely to be immune to the same stuff as Europeans. By contrast, those Indians with less Neanderthal DNA may have had less immunity to the European diseases, and so been more likely to die, leaving behind a population of high-Neanderthal DNA people.

5. One thing we can say for sure: if the data’s correct, the peopling of the Americas was more complex than previously thought.

There are four areas of particular interest:

A. The highly Neanderthal area in the Pacific Northwest

B. The highly Denisovan area in Brazil

C. The low-Neanderthal area on the South American coast

D. The low-Neanderthal and low-Denisovan area along the Baja gulf in Mexico.

I’ve got nothing on A and C; supply your own theories.

6. Speculations on the origins of high-Denisovan people in Brazil:

A couple of papers in Science and Nature recently proposed that Melanesian-related people somehow made it to the rain forest long after the other Indians got to the area. West Hunter helpfully summarizes them.

West Hunter suggests that the Melanesian-related people with their high-Denisovan DNA got to the Americas first, and were then replaced throughout the continents by later invaders, the ancestors of current Indians. The one place the Melanesian-related people managed to survive was in the depths of the rainforest, a very difficult place to conquer. Even today, there are “uncontacted tribes” living in the Amazon rainforest; if anywhere were a good spot for a group of humans to avoid getting conquered, the depths of the rainforest is a good one.

7. The low-Neanderthal and low-Denisovan area along the Baja gulf in Mexico.

So what’s up with that? As far as I know, the only people who don’t have any Neanderthal or Denisovan are Africans. (And even there, there’s a little, just due to back-migration from the rest of the world.)

Are these people descended from a totally different group that came directly from Africa?

There’s a tiny ethnic group in the area, called the Seri:

Dona Ramona of the Seri Indians of Sonora, Mexico
Dona Ramona of the Seri Indians of Sonora, Mexico

According to the Wikipedia, the Seri speak a language isolate–that is, their language, like Basque, doesn’t appear to be related to any other language on Earth–and they are not culturally connected with any of their neighbors. They’ve also held out significantly against Spanish and Mexican assimilation. In other words, they might very well be a totally isolated population that is not related at all to any of their neighbors.

Some more information on the Seris.

Then I found this interesting map:

map showing global distribution of the HLA-B*73:01 allele
maps showing global distribution of the HLA-B*73:01 allele

I found these maps over on Austin Whittall’s post on Denisovans and America.

Looks like the same spot, doesn’t it?

The two different “real” maps how different things because they come from different scientists who came up with different data, but the overall picture is similar–if you look closely, both maps show a hotspot in Israel, for example. The second map looks less detailed, (hence their miss of several Middle Eastern hotspots,) but has a wider global range, which is obviously useful for our purposes. They also help show the importance of not putting too much stock in any single study about the distribution of a particular gene or allele or whathaveyou; different scientists come up with different numbers.

At any rate, while this could be just a totally random coincidence, if it isn’t, it’s awfully interesting, isn’t it? I know the Egyptians circumnavigated Africa; the Carthaginians and Phoenicians were also noted sea-farers. Or perhaps some other group I know nothing about from the region, before folks started keeping good records. Who knows?

8. Other people’s theories: Neanderthals, Denisovans, and H erectus made it to America before we did, and H Sapiens intermixed with them after they arrived; humans evolved in American, and then migrated to the rest of the world from there.