Thursday, November 29, 2018

14,000 Year Old Fishing Village Unearthed In British Columbia

The New Archaeological Site In British Columbia

While it has received prominent mention in recent years, it is still possible to gain valuable insights into human prehistory by means other than genetics. Sometimes old school archaeological digs and carbon dating can still be a source of important discoveries. 

An archaeological site in British Columbia sheds light on the lives of members of this Founding Population at a time close to their primary expansion out of Beringia. Among other things, it corroborated the hypothesis that these people had relatively long term settlements in some places, and relied on a mix of fishing and terrestrial hunting and gathering for subsistence. 
CTV reports that a team of students from the University of Victoria’s archeology department have uncovered the oldest settlement in North America. This ancient village was discovered when researchers were searching Triquet Island, an island located about 300 miles north of Victoria, British Columbia. 
The team found ancient fish hooks and spears, as well as tools for making fires. However, they really hit the jackpot when they found an ancient cooking hearth, from which they were able to obtain flakes of charcoal burnt by prehistoric Canadians.

Using carbon dating on the charcoal flakes, the researchers were able to determine that the settlement dates back 14,000 years ago[.] . . . 
Alisha Gauvreau, a Ph.D student who helped discover this site. . . and her team began investigating the area for ancient settlements after hearing the oral history of the indigenous Heiltsuk people, which told of a sliver of land that never froze during the last ice age. 
William Housty, a member of the Heiltsuk First Nation, said, “To think about how these stories survived only to be supported by this archeological evidence is just amazing. This find is very important because it reaffirms a lot of the history that our people have been talking about for thousands of years.” 
But, one quote from the PhD student in the story is mostly wrong:
“What this is doing, is changing our idea of the way in which North America was first peopled, said Gauvreau.”
In fact, while this find is important, it is important mostly because it confirms and corroborates the existing paradigm regarding the peopling of the Americas, not because it "is changing our idea" of how this happened. It is notable not because it changes our ideas about the peopling of the Americas, but because it is some of the most clear and concrete evidence to date confirming the existing paradigm.

But, it is understandable and forgivable that an investigator selling a story about her discovery to the press stretched the truth a little on this score. Paradigm changing discoveries are hot news. And, while this particular paradigm affirming find actually is important, paradigm affirming results are rarely news (imagine how dull the nightly news would be if it ran a big news story every time that the Large Hadron Collider had a result consistent with the Standard Model of Particle Physics).

Background: Why Does The "Founding Population" Of The Americas Matter?

When it comes to the prehistory of the Americas, one of the central questions is to understand is how people arrived in the Americas, and one of the central players in the answer is the "Founding Population" of the Americas.

The Founding Population was a group of people with a quite small effective population size (a few hundred at most) who rapidly expanded from Beringia into essentially all of the "virgin territory" of North America and South America over a period of a couple of thousand years or so as the last great ice age (which peaked at the Last Glacial Maximum about 20,000 years ago) retreated, starting more than a thousand years before the Young Dryas climate event (ca. 12,900 to 11,700 years years ago, which was a return to glacial conditions which temporarily reversed the gradual climatic warming after the Last Glacial Maximum started receding).

There is a growing community of investigators and observers of the prehistory of the Americas who give credence to the scattered bits of evidence for one or more older hominin populations in the Americas (either modern human or archaic hominid) who migrated into the Americans from Beringia before or during the Last Glacial Maximum, rather than only starting when the vast North American glacier started to melt and recede. But, we know that any earlier hominins in the Americas (modern human or otherwise) never thrived and were either almost entirely wiped out by the later waves of modern human migration, or were so similar genetically similar to the founding population of the Americas they are indistinguishable from them genetically. Because there is no distinguishable trace of them in any modern or ancient DNA samples from the Americas with the possible exception of some minor "paleo-Asian" ancestry in a few tribes in the Amazon, whose origins are a mystery.

But, even if you find that evidence to be credible, there is overwhelming modern and archaic genetic evidence that 99.99% or more of the ancestry of the pre-Columbian residents of the Americas is derived from a single "Founding Population" which started to expand in earnest not many centuries earlier than 14,000 years ago, subject to two exceptions: (1) Inuits in the Arctic and sub-Arctic, and (2) some select tribes in Alaska, the Pacific Northwest and the American Southwest with Na-Dene ancestry. 

The Inuits derive from a migration wave from Northeast Asia within the last two thousand years and replaced earlier "paleo-Eskimo" populations in Northern Canada. The Na-Dene derive from a migration wave from Northeast Asia around the time of the European Bronze Age and then admixed with descendants of the Founding Population who were already present in North America. 

But, apart from a small component of some cryptic "paleo-Asian" ancestry in a handful of hunter-gatherer tribes in jungles in the Amazon River basin near the northeastern foothills of the Andes Mountains, all other pre-Columbia genetic ancestry in the Americas derives from the Founding Population. Founding Population ancestry was the predominant source of ancestry in almost every non-Inuit indigenous person in North America and South America in 1492, and was the only source of ancestry in the lion's share of those millions of people.

So, given their central role as the primary ancestors of all of the indigenous people of the Americas, except the Inuits, knowing more about this quite small community of people from around 14,000 years ago, is obviously a matter of great importance.

41 comments:

DDeden said...

See message by GHorvat:

https://groups.yahoo.com/neo/groups/HumanMigrations/conversations/messages/6344
-

I interpret this as: The (coastal bark-boating) Andaman-like males were killed off by early southward-trecking Paleo-Americans/Beringians with clovis tech., while a few surviving Andaman-like females mated with them, keeping the MtDNA lineage alive only in their surviving descendants.

andrew said...

The only problem with that analysis is that the no Andaman-like mtDNA lineages survived, even in the populations where Andaman-like autosomal genetics persisted. Now, generally speaking, if there are literally just a few such women, losing their mtDNA lineages sometime before the present due to random genetic drift wouldn't be too surprising. But, the Americas in the early founding period was growing wildly because there was endless virgin territory that they didn't have to compete for and abundant megafauna which wasn't used to humans. And, it is quite unusual to loose any ancestry component, either autosomal or uniparental due t random genetic drift in a rapidly expanding population. You lose ancestral genetics due to genetic drift when populations are stable or hit a bottleneck during which they decline in population. So, to lose the mtDNA lineages entirely due to random drift the women have to, by chance, have no daughters with surviving female descendants in very favorable conditions in the first few generations, and the number of admixing women needs to be very small in absolute number. Basically, there needs to be 100% genocidal slaughter until one small tribe remains in the Amazon, and then the conditions there need to be population stagnant and geographically immobile, and only a few women need to admix at this final brink, which would cause the likely loss of their mtDNA due to genetic drift and the survival of only a modest amount of their autosomal DNA which doesn't spread into neighboring populations a couple hundred miles away.

The main Native American mtDNA clades are A2, B2, C1 (including C1b, C1c and C1d) and D1. In addition X2a, D2 and D3 are found only in North America. Two of these are probably late arrivals: D2 is found in Inuits and the Na-Dene and D3 is found only in Inuits. But, X2a was probably a founding clade that just wasn't in the subset of people who went to South America. C1a and A2a in Asia are probably back migrations from the Americas.

As of 2007, "Haplogroup D4h3 ranges from Alaska to Tierra del Fuego and has recently been identified in Alaskan skeletal remains (10,300 ybp) [27]. We identified haplogroup C4c in two Ijka-speakers from Colombia, but its distribution in the Americas remains poorly characterized." Both are deeply rooted Asian clades.

Another study from 2012 states that C4c has a similar geographic distribution to X2a although it is found in one individual in Columbia, South America in addition to 15 in North America, while X2a is not and C4c is similar in age to or younger than X2a https://www.ncbi.nlm.nih.gov/pubmed/22024980 and a third study from 2009 says that D4h3 has a North and South American coastal distribution https://www.ncbi.nlm.nih.gov/pubmed/19135370 D4h3 is mostly from Thailand and would be the only mtDNA candidate viable for Paleo-Indians really. https://en.wikipedia.org/wiki/Haplogroup_D_(mtDNA)


andrew said...

In addition to having a similar distribution within the Americas to X2a excluding two identical outlier individuals in Columbia (which is admittedly where we would want Paleo-Asian mtDNA to be) and a younger age, there is also the issue that C4c is rooted in North Asia. Per this paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3006427/

"The C4 branch shows a coalescence time of 20–22 kya, implying that it began to expand before the LGM. Inside haplogroup C4 a new subclade, C4e, specific for Altai region populations has been revealed (Figure S1). It is defined by transitions at nps 151, 152, 7307, 15479 and, together with Russian individual (Rus_184), characterized by lack of adenine insertion at np 2232, which is thought to be diagnostic for a whole subclade C4a'b'c [20]. This subclade represents a major fraction of C4 mtDNAs and can be further subdivided into C4a, C4b and Native American-specific branch C4c identified so far only in two Ijka-speakers from Colombia [4] and one Shuswap individual from British Columbia [21]. Cluster C4a dates to 19–25 kya, demonstrating the pre-LGM time of divergence, in contrast to C4b, which is characterized by younger coalescence time estimated as 6–7 kya."

Since this paper from the year 2010 there have been 13 more C4c individuals have been found in North America, all east of the Rocky Mountains and north of the U.S. Southeast. So, C4c just isn't a good candidate for a Paleo-Asian as opposed to founding population mtDNA lineage.

More C4 data from https://dna-explained.com/2013/09/18/native-american-mitochondrial-haplogroups/:

"C4

2007 Tamm
Anzick Provisional Extract, Estes January 2015 – (4 C4 with no subgroup)
Chippewa – White Earth Reservation, Minnesota – private test at 23andMe
Inupiat people from Alaska North Slope – Raff 2015
New Native American Mitochondrial DNA Haplogroups, Estes, 2017
C4a

Native American and Siberian, Kumar 2011
New Native American Mitochondrial DNA Haplogroups, Estes, 2017
C4a1

Anzick Provisional Extract, Estes, September 2014, kits F999912 and F999913
Buryat and Eleeut – Derenko
India and Russia and Turkey – FTDNA
India in the Haplogroup C project at Family Tree DNA
C4b

Kumar 2011
New Native American Mitochondrial DNA Haplogroups, Estes, 2017
C4c

Beringian Founder Haplogroup – 2008 Achilli

2007 Tamm – found in only 2 samples, an Ijka sample from South America and a Shuswap speaker from North America,
Shuswap Speaker, North America – Malhi 2010 (Suswap is now C4c1)
New Native American Mitochondrial DNA Haplogroups, Estes, 2017
C4c1

Anzick Provisional Extract, Estes, September 2014, kits F999912 and F999913
Suswap – Malhi
N American – Kashani
Anzick Provisional Extract, Estes January 2015 – (1 C4c1)
Chippewa Cree, Jasper House, Alberta, Canada (1835), FTDNA American Indian Project
New Native American Mitochondrial DNA Haplogroups, Estes, 2017
C4c1a

North American – Kashani, Cherokee in the Haplogroup C project at Family Tree DNA
New Native American Mitochondrial DNA Haplogroups, Estes, 2017
C4c1b

North American – Kashani
New Native American Mitochondrial DNA Haplogroups, Estes, 2017
C4c2

North American – Kashani
New Native American Mitochondrial DNA Haplogroups, Estes, 2017
C4e

Uika – Tamm 2007
Shor and Teleut – Derenko
New Native American Mitochondrial DNA Haplogroups, Estes, 2017"

andrew said...

While the same paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3006427/ describes mtDNA D as mostly northern Asian, it has a broader distribution and is much older than C4a:

"Haplogroup D has a likely pre-LGM time depth characterized by an overall coalescence time estimate of 35–37 kya, depending on the mutation rate used. Two of its major subclades, D4 and D6, have a similar age of 24–28 kya and 23–42 kya, respectively, whereas subclade D5 has an older coalescence time estimated as 32–37 kya (Table S4). Haplogroup D4, the most represented of D clades, is further subdivided into fifteen principal subclades (D4a–D4j, D4k'o'p', D4l–D4n, D4q), which range from ∼6 to ∼28 kya when using the sequence variation of the entire genome and from ∼3 to ∼42 kya when only synonymous mutations are counted. Some of these subclades have a very distinctive geographic distribution, which is highly informative about the demographic history of the northern Asia. Whilst all subclades are found in eastern Asia, so that eastern Asian lineages occur throughout the tree, few of them are specific for northern Asian populations."

mtDNA D4b1a2a1a is found in Inuits in Greenland and Canada with some frequency and also in Arctic and near Arctic Asia. It is of a piece with mtDNA D3 in distribution.

mtDNA D4h is 20.1kya to 19.4kya in age, considerably older than mtDNA C4c. Specifically with differing assumptions about mutation rates:

>>D4h 11 20.06 (12.51; 27.87) 19.35 ± 6.04
>>>D4h1 7 17.38 (8.5; 26.67) 18.02 ± 7.96

mtDNA D4h in general is found mostly in southern Siberian – in green (a small portion), and in eastern Asian – in red (the predominant part).

mtDNA D4h3 which is rooted in SE Asia (and Thailand in particular) where Andaman-like Paleo-Asian ancestry would be expected, and at odds with the root location of other Founding population lineages and in doesn't have a distribution similar to any of the other founding mtDNA clades of the Americas:

D4h - Thailand/Laos
D4h1
D4h1a
D4h1a1 - Japan
D4h1a2 - Japan
D4h1b - Hunan (Han), Japan
D4h1c - China (incl. Tu), Tibet
D4h1c1 - Japan
D4h1d - Bargut
D4h2 - Ulchi
D4h3 - Thailand
D4h3a - South America (Peru, Ecuador, Argentina, Bolivia, Brazil), Mexico, USA
D4h3a1 - Chile
D4h3a1a - Chile
D4h3a1a1 - Chile
D4h3a1a2 - Chile
D4h3a2 - Chile, Argentina
D4h3a3 - Chile
D4h3a3a - Mexico, USA
D4h3a4 - Peru
D4h3a5 - Chile, Peru, Argentina
D4h3a6 - Peru, Ecuador
D4h3a7 - ancient Canada
D4h3a8 - Mexico
D4h3a9 - Peru
D4h3b - China
D4h4 - Uyghur, Tibet, Japan
D4h4a - Kyrgyz (Tashkurgan), Buryat, Bargut

The distribution in the Americas gets tolerably close to the target area where Paleo-Asian autosomal DNA has been found.

andrew said...

Here is another distribution list (limited to the Americas) from https://dna-explained.com/2013/09/18/native-american-mitochondrial-haplogroups/

"D4
2007 Tamm
Cayapa, Ecuador – Fagundes 2008
Anzick Provisional Extract, Estes January 2015 – (2 D4)
Chumash – Breschini and Haversat 2008
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4b1

Anzick Provisional Extract, Estes January 2015 – (1 D4b1)
D4b1a

Inupiat people from Alaska North Slope (noted as formerly D3), ancient Neo-Eskimos – Raff 2015
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4b2a2

Anzick Provisional Extract, Estes January 2015 – (1 D4b2a2)
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4e1

Mexican American – Kumar 2011
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4e1a1

Anzick Provisional Extract, Estes January 2015 – (1 D4e1a1)
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4e1c

Kumar 2011 – found in Mexican Americans (2 sequences only)
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4g1

Anzick Provisional Extract, Estes, September 2014, kits F999912 and F999913
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h1a

Anzick Provisional Extract, Estes, September 2014, kits F999912 and F999913
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h1a1

Anzick Provisional Extract, Estes, September 2014, kits F999912 and F999913
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h1a2

Anzick Provisional Extract, Estes, September 2014, kits F999912 and F999913
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h3

Beringian Founder Haplogroup – 2008 Achilli
2007 Tamm
Anzick Provisional Extract, Estes January 2015 – (1 D4h3)
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h3a

Veracruz, Mexico, Arequipa, Peru, Loreto, Peru, Ancash, Peru, San Luis Potosi, Mexico, Maranhao, Brazil – Perego 2009
Mexican American – Kumar 2011
Anzick Provisional Extract, Estes, September 2014, kits F999912 and F999913
Anzick Provisional Extract, Estes January 2015 – (2 D4h3a)
Raff and Bolnick, Nature February 2014 – Anzick’s haplogroup
Remains from On Your Knees Cave in Alaska, Chatters, 2015
Gran Chaco, Argentina – Sevini 2014
Aymara, Mapuche, Huilliche, Kawesqar, Tehuelche, Yamana in Chile and Argentina, South America – de Saint Pierre, 2012
Native American Mitochondrial
DNA Haplogroups, Estes, 2017
On Your Knees Cave, Alaska, 10,300 YPB – Lindo 2017
D4h3a1

Coquimbo, Chile, O’Higgins, Chile, Coquimbo, Chile, Santiago, Chile, Los Lagos, Chile, Bio-Bio, Chile – Perego 2009
D4h3a1a

Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h3a1a1

Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h3a2

Gran Chaco, Argentina – Sevini 2014
D4h3a3

Chihuahua, Mexico, Tarahumara, Mexico, Nuevo Leon, Mexico – Perego 2009
D4h3a4

Arequipa, Peru – Perego 2009
D4h3a5

Maule, Chile, Los Lagos, Chile, Santiago, Chile – Perego 2009
D4h3a6

Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h3a7

British Columbia ancient sample 939, may be extinct – Ciu 2013
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4h3a8

Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4j

Anzick Provisional Extract, Estes January 2015 – (2 D4j)
Native American Mitochondrial DNA Haplogroups, Estes, 2017
D4j8

Gran Chaco, Argentina – Sevini

andrew said...


https://dna-explained.com/2017/03/02/new-native-american-mitochondrial-dna-haplogroups/ shows a D1 hotspot right in the Amazon near where Paleo-Asian autosomal DNA traces were found.

The autosomal info is In August, 2015, in an article published in Science, Genomic evidence for the Pleistocene and recent population history of Native Americans by Raghaven et al suggested that a secondary migration occurred from further south in Asia, specifically the Australo-Melanesians.
A second paper, published in Nature in September 2015 titled Genetic evidence for two founding populations of the Americas by Skoglund et al says that South Americans share ancestry with Australasian populations that is not seen in Mesoamericans or North Americans.

andrew said...

There is D4h in India and Nepal. https://haplogroup.org/mtdna/rsrs/l123456/l23456/l2346/l346/l34/l3/m/m80d/d/d4/d4h/

Austin Whittal discusses D4h3 at http://patagoniamonsters.blogspot.com/2014/01/mtdna-d4h3a-haplogroup.html and http://patagoniamonsters.blogspot.com/2014/01/mtdna-d4h3a-continued.html He notes in this 2014 post: "InNative American Groups its frequency is high in some groups, low in others and absent in most:

21.67% Cayapa (Ecuador)
16.00% Chumash (Ca. US)
10.26% Fuegians (Chile)
2.70% Mapuche (Chile)
2.56% Klunk Mound
1.82% Tarahumara (Mexico)
1.28% Nahua (Mexico)
0.40% Mixtek - Mixe - Zapotec (Mexico)
0.95% Quechua (Peru)

It is absent in all other Native Americans sampled. Clearly the canoe people (Cayapa, Chumash and Fuegians) have the highest frequencies. Mapuche very likely got it thorugh admixture with the now extinct Chono (who also carried it), under Spanish rule in northern Patagonia, at the ChiloƩ Island outpost. The other groups in Mexico and Peru carry it at a frequency below 2%, all the North American native people except the Chumash and the Klunk Mound skeletal remains do not carry it."

From here: http://etheses.whiterose.ac.uk/7872/1/KK%20Eng%20PhD%202014.pdf
"D4h has four subclades, D4h1 is almost entirely found in Japan and at much
lower levels in China. D4h2, D4h3 and D4h4 are each represented by one complete sequence,
and they are found respectively in North Asia, Brazil and Japan."

"Figure E.18 shows D4h dates to ~24 ka, and includes subclades D4h1, D4h2, D4h3 and D4h4. However, D4h2, D4h3 and D4h4 are represented each by an instance from Russian Far East (Starikovskaya et al., 2005), Brazil (Hartmann et al., 2009) and Japan (Tanaka et al., 2004). Tamm et al. (2007) reported that haplogroup D4h3 is found between Alaska to Tierra del Fuego, and has been recently identified in Alaskan skeletal remains dating to ~10.3 ka (Kemp et al., 2007). D4h1 dates to ~20 ka and found in Japan (Tanaka et al., 2004) and North China (Zheng et al., 2011). The subclades of D4h1 are all found in Japan only (Tanaka et al., 2004). On the other hand, the HVS-I data shows the root type is widely distributed across East Asia albeit at a minor frequency (Kong et al., 2006). The overall haplogroup D4h suggests early settlements in Japan during the LGM and a likely southern source in China."

andrew said...

There was a Holocene migration to Columbia from North America which could have brought the mtDNA C4c individual there: http://www.scielo.org.co/pdf/unsc/v20n2/v20n2a09.pdf particularly Chibchan-speaking peoples. " Here, it is significant to note the high frequency of C within the Arawakan-speaking Wayuu population, despite their close geographic proximity to the Chibchans peaking populations of the SNSM (consisting of the Ijka, Kogui, and Arsario) in which haplogroup A predominates (Melton et al. 2007). The prevalence of C amongst the Wayuu may also support the notion that they share genetic affinity with other indigenous groups from the Amazonian region, where haplogroup C likewise predominates and in (or near) which the Arawakan languages are thought to have originated (Aikhenvald 1999, 2006; Walker & Ribeiro 2011)."

It looks like the two Columbias spoke a Chichan language. So that outlier is well explained by Holocene migration.

Bottom line: the only mtDNA clade that could be associated with Paleo-Asian origins is mtDNA D4h3a and that isn't a horrible fit.

andrew said...

The Chichan may be Mayan related. They came from Meso-America which is still closer to C4c land than Columbia.

andrew said...

Corroborating the Thai case of D4h3: https://www.biorxiv.org/content/biorxiv/suppl/2016/07/11/063172.DC1/063172-1.pdf

Consider "About 12,600 years ago, when ice sheets still covered parts of North America, a baby boy lived, died and was buried in a rocky grave in a field in western Montana. A new whole genome sequencing of this infant — the oldest genome sequence of an American individual — identifies his community as ancestors of Native Americans who live on the continent today. (As always, this type of statement is contradicted elsewhere).

In addition, analysis of the child’s mitochondrial DNA indicated Anzick-1 belonged to what’s known as the D4h3a haplogroup, or lineage. The finding is important — and surprising, according to researchers — because the D4h3a line is considered to be a “founder” lineage, belonging to the first people to arrive in the Americas. Although rare in most Native Americans in the U.S. and Canada today, D4h3a genes are found more commonly in native people of South America, far from the Montana cliff beneath which Anzick-1 was laid to rest." http://realhistoryww.com/world_history/ancient/Misc/Americas/Pre_America.htm

But, we know that Anzick-1 has no Paleo-Asian autosomal ancestry, so this makes linking it with D4h3 rather suspect. The bearers of Paleo-Asian ancestry themselves also seem to lack D4h3a. While not an impossible result, the linkage seems very thin and implausible.

"Native Americans living in the Amazon bear an unexpected genetic connection to indigenous people in Australasia, suggesting a previously unknown wave of migration to the Americas thousands of years ago, a new study has found.
The new study, published July 21 in Nature, indicates that there's more to the story. Pontus Skoglund, first author of the paper and a postdoctoral researcher in the Reich lab, was studying genetic data gathered as part of the 2012 study when he noticed a strange similarity between one or two Native American groups in Brazil and indigenous groups in Australia, New Guinea and the Andaman Islands.

The Tupƭ-speaking Suruƭ and Karitiana and the Ge-speaking Xavante of the Amazon had a genetic ancestor more closely related to indigenous Australasians than to any other present-day population. This ancestor doesn't appear to have left measurable traces in other Native American groups in South, Central or North America. The genetic markers from this ancestor don't match any population known to have contributed ancestry to Native Americans, and the geographic pattern can't be explained by post-Columbian European, African or Polynesian mixture with Native Americans, the authors said. They believe the ancestry is much older--perhaps as old as the First Americans. In the ensuing millennia, the ancestral group has disappeared. "We've done a lot of sampling in East Asia and nobody looks like this," said Skoglund. "It's an unknown group that doesn't exist anymore." The team named the mysterious ancestor Population Y, after the Tupƭ word for ancestor, "YpykuƩra."" https://www.nature.com/articles/nature14895

andrew said...

The Supplements to Soglund (2015) is at https://media.nature.com/original/nature-assets/nature/journal/v525/n7567/extref/nature14895-s1.pdf

"The evidence for two ancestral populations is driven by Amazonian and Australasian populations To determine which outgroup and Native American populations contributed the most to the rejection of rank 0, we examined the weight coefficients assigned to each population by qpWave. We find that the greatest coefficient among the outgroups is assigned to the Onge from the Andaman Islands. The next strongest coefficients are assigned to South Asians and Oceanians. The greatest coefficients among Native Americans are assigned to the Amazonian SuruĆ­ and Karitiana, whereas the lowest are assigned to the Central American Mixe and Pima (Extended Data Figure 2). "

"A genetic affinity between native Amazonians and native Oceanians We find evidence for a significant excess in shared derived allele frequencies between Amazonians and five populations: Onge, Papuans, New Guinean highlanders, Australians from Arnhem Land, and Mamanwa Negritos from the Philippines (Z > 3). In addition, we find positive statistics for other Oceanian populations such as Bougainville Papuans, Tongans, and Ami from Taiwan, and some Indian populations such as Kharia and Bengali (Extended Data Table 2). "

"Comparison with the ancient Clovis-associated Anzick individual
We investigated the evidence for differential relatedness to Australasians using
ancient Native American samples overlapped with San and Yoruba ascertained SNPs
on the Human Origins array. We use Dinka as outgroup, as they are less differentiated
from non-Africans than Yoruba are. This minimizes the effects of errors in the ancient
DNA causing an attraction to the outgroup. We computed f4(Dinka,
Onge/New_Guinea/Papuan; Ancient, Mixe/Surui). We find no evidence for a greater
affinity to Onge/New_Guinea/Papuan in any ancient sample than is found in the
Mixe. In contrast, we find that the SuruĆ­ show an affinity to
Onge/New_Guinea/Papuan compared to the ancient Anzick individual (Table S3.6). "

andrew said...

"we tested different proportions of this
ancestry in the SuruĆ­, assuming that the proportion in Mixe is 0, and determined the
proportions for which the maximum Z-score between predicted and observed f4-
statistics is less than 3. We find that the plausible range of Australasian-related
ancestry in the SuruĆ­ is 0.7-9.8% and that the lowest maximum Z-score is obtained for
an admixture proportion of 2.3% (Extended Data Figure 7). If we instead find the
proportion for which no Z-score is greater than 2, we obtain a range of 1.5-5.7%."

Marnie said...

@andrew, November 30, 2018 at 7:02 AM

"The only problem with that analysis is that the no Andaman-like mtDNA lineages survived"

It could be inferred from the Matsumura paper that from 100,000 to about 20,000 years ago, the entire populations of southern India, Nicobar Islands, Andaman Islands, Southern China, Southeast Asia, Papua New Guinea and Australia had an "Andaman" like characteristic.

Japan also appears to initially have had a somewhat Andaman like characteristic (given that the Ainu retain a trace of this).

See Matsumura et al.

link: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0198689

Papuans and Aboriginal Australians Y-chromosome haplogroups are primarily K and C lineages:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4819516/

Mitochondrial haplogroups of Aboriginal Australians are M and N derived:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5347126/

Given that mtdna M and N, and Y-chromosome K and C were well established in Southeast Asia by at least 50,000 years ago (according to the above papers) it is possible that some of the Y-chromosome haplogroups in the Americas (K derived and Q derived) and mtdna (M derived and N derived) initially reached the Americas on the order of 50,000 years ago.

In that case, it would not be true that "no Andaman-like mtDNA lineages survived".

Other support for this possibility:

-Many aspects of the material culture of various groups of Native Americans are similar to the material culture of Southeast Asian indigenous cultures, including Papuans and Aboriginal Australians

-Some Plains Native Americans share, in high frequency, an uncommon blood group with Aboriginal Australians (type A)

-Na-Dene language speakers, including Chipewyans, have a high degree of Y-chromosome haplogroup Haplogroup C-M217. Yet, Na-Dene do not appear to have been a late arriving population from Asia and are not separable as a population from other Native North Americans.

See Raghevan: https://www.ncbi.nlm.nih.gov/pubmed/26198033

Therefore, it is possible that the Y-chromosome Haplogroup C-M217 in the Americas emerged in North America, Beringia, Sakhalin, Aleutian Islands, and Japan before the last Ice Age.

Marnie said...

@andrew, November 30, 2018 at 7:02 AM

"third study from 2009 says that D4h3 has a North and South American coastal distribution https://www.ncbi.nlm.nih.gov/pubmed/19135370 D4h3 is mostly from Thailand and would be the only mtDNA candidate viable for Paleo-Indians really. https://en.wikipedia.org/wiki/Haplogroup_D_(mtDNA)"

Mitochondrial haplogroup D lineages emerge about 40,000 to 60,000 years ago. Their distribution indicates an expansion into the Americas before 40,000 years ago from Japan or Southeast Asia.

Marnie said...

@andrew November 30, 2018 at 10:04 AM

"We find that the plausible range of Australasian-related
ancestry in the SuruĆ­ is 0.7-9.8% and that the lowest maximum Z-score is obtained for an admixture proportion of 2.3% (Extended Data Figure 7). If we instead find the proportion for which no Z-score is greater than 2, we obtain a range of 1.5-5.7%."

When I see analysis like this, I assume that I am looking at a very noisy result, with some confounding factor, like the time since the two populations split (>50,000 years ?), and admixture.

Marnie said...

@andrew November 30, 2018 at 8:27 AM

""The C4 branch shows a coalescence time of 20–22 kya, implying that it began to expand before the LGM. Inside haplogroup C4 a new subclade, C4e, specific for Altai region populations has been revealed (Figure S1). It is defined by transitions at nps 151, 152, 7307, 15479 and, together with Russian individual (Rus_184), characterized by lack of adenine insertion at np 2232, which is thought to be diagnostic for a whole subclade C4a'b'c [20]. This subclade represents a major fraction of C4 mtDNAs and can be further subdivided into C4a, C4b and Native American-specific branch C4c identified so far only in two Ijka-speakers from Colombia [4] and one Shuswap individual from British Columbia [21]. Cluster C4a dates to 19–25 kya, demonstrating the pre-LGM time of divergence, in contrast to C4b, which is characterized by younger coalescence time estimated as 6–7 kya.""

"Since this paper from the year 2010 there have been 13 more C4c individuals have been found in North America, all east of the Rocky Mountains and north of the U.S. Southeast. So, C4c just isn't a good candidate for a Paleo-Asian as opposed to founding population mtDNA lineage."

The alternative possibility for C4 is that it represents a highly mobile people who ranged from North America (east of the Rockies) to the Altai (when Beringia was passible).

Have a look at the distribution of Cervus canadensis:

http://linearpopulationmodel.blogspot.com/2016/04/distribution-of-cervus-canadensis-wapiti.html

And the distribution for caribou reindeer:

http://linearpopulationmodel.blogspot.com/2013/11/caribou-and-reindeer-map.html

The pattern for C4 reflects the pattern of mobile caribou/reindeer hunters in North America and Eurasia.



Marnie said...

@andrew November 30, 2018 at 8:49 AM

"The autosomal info is In August, 2015, in an article published in Science, Genomic evidence for the Pleistocene and recent population history of Native Americans by Raghaven et al suggested that a secondary migration occurred from further south in Asia, specifically the Australo-Melanesians."

When I read a statement like this in a paper that is published in Science, it makes me wonder if the editors at Science have their heads in the clouds.

Yes, the authors found an Australo-Melanesian trace signature in their results. OK. But to say it is the result of "a secondary migration occurred from further south in Asia" is unscientific and misleading. The fact is that the authors don't have the slightest idea why this "Australo-Melanesian" trace signature appears. Their data is still quite limited in terms of time and distribution sampling. So I think they should just say that they see a trace Australo-Melanesian signature and leave it at that, without the phony, confused embellishments.

I realize that it is just a sentence, but I see these kinds of wild, over stated, over confident statements in ancient DNA studies all the time. Hey, I'm really sick of it. And I think I am not the only one.

Marnie said...

@andrew November 30, 2018 at 9:50 AM

"But, we know that Anzick-1 has no Paleo-Asian autosomal ancestry, so this makes linking it with D4h3 rather suspect. The bearers of Paleo-Asian ancestry themselves also seem to lack D4h3a. While not an impossible result, the linkage seems very thin and implausible."

It is possible that given proximity of Anzick-1 to the passes between Montana and the Pacific West Coast, that these people living along the east side of the Rocky Mountains gradually intermarried with people further west (some living along the Pacific West Coast, who may have had D4h mitochondrial DNA).

G Horvat said...

Andrew wrote - "The only problem with that analysis is that the no Andaman-like mtDNA lineages survived..."

This is true... but nearby Nicobarese are about 30% haplogroup B. The branch is B5, however, which is not especially close to Native American B2. B2 is closest to the 'B4b' sequences of the Bunun and Hainan Islanders.

Haplogroup D4h2 was also detected in an ancient Jomon sample (Adachi AJPA 2011). The first Asian D4h3 sequence that was found was from the Shandong peninsula.

andrew said...

"Yes, the authors found an Australo-Melanesian trace signature in their results. OK. But to say it is the result of "a secondary migration occurred from further south in Asia" is unscientific and misleading. The fact is that the authors don't have the slightest idea why this "Australo-Melanesian" trace signature appears. Their data is still quite limited in terms of time and distribution sampling. So I think they should just say that they see a trace Australo-Melanesian signature and leave it at that, without the phony, confused embellishments.

I realize that it is just a sentence, but I see these kinds of wild, over stated, over confident statements in ancient DNA studies all the time. Hey, I'm really sick of it. And I think I am not the only one."

I don't disagree. There are lots of things that ancient and modern DNA researchers are doing right. But, they are both insufficiently humble about what their results mean, and grossly insufficiently interdisciplinary. To get a convincing result all of the pieces need to come together. The DNA data needs to fit to a plausible narrative of how it got that way, and the archaeological and linguistic data need to confirm that.

For example, most of the mtDNA C4c papers note that there are two individuals from a particular language group who have this in Columbia, but fail to not that this population was a secondary wave arrival that appeared in the Holocene in Columbia and has linguistic and genetic similarities to the Mayans, which makes the presence of this mtDNA type near the outlier autosomal DNA populations far less interesting.

andrew said...

I've been collecting data in comments as I've gotten going responding to the first one, but I will write this up more coherently and less disjointedly as a proper blog post fairly soon.

DDeden said...

The DNA discussion is way beyond me. I'm grateful for it, despite my ignorance.

Chippewa, Yahgan(Fuego), PirahĆ£(Amazon), Tasmanians & some Australians used bark canoes. Inuits(walrus; umiak, kayak) & Tibetans(yak; khudru) used skin boats, some Central Asians also did.

Tibetan plateau inhabited 30-40ka
http://feedproxy.google.com/~r/TheArchaeologyNewsNetwork/~3/nWdBHsA1yuQ/new-archaeological-site-revises-human.html?utm_source=feedburner&utm_medium=email

Khudru@Tibet: greased yak-hide coracle

Men can pass on mtDNA

https://archaeologynewsnetwork.blogspot.com/2018/11/study-shows-mitochondrial-dna-can-be.html?utm_source=feedburner&utm_medium=email&utm_campaign=Feed:+TheArchaeologyNewsNetwork+(The+Archaeology+News+Network)&m=1#qYilYJuqVrFE3Hds.97

DDeden said...

YpykuƩra@Tupi: ancestor
Horim@Hebrew: ancestors
Ebu@Sinama/Bajau: ancestors
Ebu@Flores/(Rampasa?): ancestors
"* E(m)buhuara" embear(forebears)

Possible (very remote) ling. link between group names: Kharia (India), Surui, Xavante & Karitiana (Brazil)? "*Xyua(t)Ʀ"

Marnie said...

@andrew November 30, 2018 at 3:59 PM

"most of the mtDNA C4c papers note that there are two individuals from a particular language group who have this in Columbia, but fail to not that this population was a secondary wave arrival that appeared in the Holocene in Columbia and has linguistic and genetic similarities to the Mayans, which makes the presence of this mtDNA type near the outlier autosomal DNA populations far less interesting."

The SAA conference (held in Washington this year) had a very good session on Western Hemisphere lithic cultures, and their distributions. They were looking at not just North America, but also, Central and South America and are trying to discern, using lithic morphometrics, the relationship of these lithic cultures over time and geography:

[120] Symposium źž FLUTED POINT TECHNOLOGY: VARIATION ACROSS A HEMISPHERE The transmission of Clovis fluted point technology, the earliest well-dated fluted point industry, is considered to have spawned regional variants of point forms as a result of cultural drift and founder effects as Paleoindian groups dispersed throughout the Western Hemisphere and adapted to local ecological settings. Recent research on fluted point technology in North and South America has highlighted such variability in technology and morphology across both geographic space and time. Now more than ever, researchers across the Western Hemisphere focus on understanding regional variants in fluted point technology in terms of their evolutionary development, historical relationships to earlier forms, and functional advantage in specific ecological settings. Traditional methods of technological and morphological comparison are met with digital analyses of morphology and shape using geometric morphometrics, phylogenetic analyses using cladistics, experimental analyses of function and manufacture, and technological analyses of manufacture sequence and reduction protocols. The papers in this session discuss morphology and technology of fluted point variants, such as Folsom, Cumberland, Barnes, Northern Fluted, Gainey, Suwannee, Fishtail, and their relationships to neighboring forms, both spatially and chronologically, as well as evolutionary evidence of their historical development and origin.

I think these researchers will continue this approach, and I am hoping to see another great session on this at next year's SAA.

andrew said...

"Men can pass on mtDNA"

I has happened, but it is a one in a ten million fluke. Odysseus or Sinbad sailing across the Indian Ocean, kidnapping someone in the Andaman Islands, then sailing to South American, climbing over some mountains and depositing their abductee there would probably be a more likely explanation.

Jaap said...

Andrew; I'm glad you've been doing your homework. So have Marnie and DD. So there's little for me to add. Except to say that I'm sick and tired of numbskulls only looking to adhere to the 'existing paradigm'. US academia has an awful record in this respect. Must I remind you of Cinque Mars and the treatment he got? Or even Louis Leakey who apparently didn't know the difference between an artifact and a geofact! Think of all those highly competent people worling away at their finds and no longer bothering knocking at the door of US-academia! Good-year, Nuidon ... They run into people taking the shit out of you, suggesting collobos monkeys may have made fires in stone-rings within caves. Prove otherwise! It's kids' stuff! Like this stupid mantra of 'when modern humans first dispersed out of Africa 60 kya ago ...' When the evidence is in your face! The aborigenes had been down under for ages! With the finest of industries!
To my mind it's just common sense: the thinking of an honest human. If you don't want to see what's in front of you, you may revert to a host of arguments that aren't really arguments. If you want to believe natural forces may well have put up those tusks erect, and smashed those bones, then fine! Have your way! No point in us having a conversation. Making a case that questions these finds - like where are the scratchmarks? How did the meat come off? - there we'd be on the same page again. But still the knocking on the door would persist: it so much looks like a human trace! And as such it should be treasured! Only to be shelved when the counter-narritive gets too convincing.
The acrimony with which the Solutrean Hypothesis was received is another example of how I was trained not to engage in science. I'm an academic and nobody's fool. And this was just ghastly! US-academia got their political correctness completely down-side-up! Them was blue-eyed niggers, for chrissakes! What's ya problem? This idea is still on my table, pending further evidence ...
Who it was survived in the Americas is no secret: there was an ANE-population in Northern Siberia around 25 kya. These very same were in Siberia at that very same time! So Northern Siberia, Alaska, Yukon. HG-groups with a link to other groups, some of whom had already ventured south of the Gordilderan Ice-sheat. Lost contact in the IGM, but no worries, others came after.
Point is: did these folks meet anyone already there? I bet they did! But where? And when exactly? And how? And whom? There's such a lot to figure out here that I can imagine people saying: 'Nah, don't want to go into that. Just can't further my academic career with that question!' And there you are.
Still there are plenty of people to get their hands dirty. Also those whose work has not been according to the paradigm will probably have beeen keeping at it, if only out of self-respect ...
Evidence says there's a group here who won. Is that all that matters? How? When? Why?
Here I want a good story. To reconnect me and assert myself up-against and/or together with you. Not very scientific, I'll admit ...

andrew said...

@Jaap

"I'm sick and tired of numbskulls only looking to adhere to the 'existing paradigm'. . . . Point is: did these folks meet anyone already there? I bet they did! But where? And when exactly? And how? And whom? There's such a lot to figure out here that I can imagine people saying: 'Nah, don't want to go into that. Just can't further my academic career with that question!'"

Unlike the situation is cosmology, where theorists are making beauty based estimates of the likelihood of things existing (like fundamental constants having particular values) with no meaningful Baysean prior distribution to guide them, in the fields of archaeology and paleo-genetics we do have priors based on data to guide us, which we summarize in what is called the "paradigm." The Baysean prior distribution isn't based on unlimited evidence, but it is based upon a significant volume of data which provides robust expectations because data from lots of different methodologies taken in isolation, and integrated together, tends to point to the same conclusions.

For example, given the genetic unity of the vast majority of New World population genetics and the low effective population of the Founding Population and the probable most recent date of major expansion (all of which can be determined even from modern population genetics alone and which are further calibrated and enhanced with ancient DNA), we can say with considerable confidence that there were at most only a few languages in use in the Founding Population and that there could have been just one or two of them. And, we can also say, based upon linguistic data, that given the diversity of New World languages as of 1492 inferred from all available data, that the original languages of the Founding Population have been diversifying and evolving for a much longer period of time than the Indo-European languages (probably not less than 5,000-6,000 years old even giving any expectation of youth the benefit of the doubt). So, we know that the Founding Population must have started to expand from a bottleneck effective population size >> than 6,000 years ago, which is very consistent with estimates from other sources of 14,000ish years ago. Genetic data and linguistic data are corroborated by carbon-14 dates of the oldest human remains and undeniable obvious signs of human presence (e.g. structures and carved bone tools and petroglyphs).

andrew said...

And, we also have corroboration from African, North Asian and East Asian archaeology, which puts a floor on the time frame that modern humans could have, respectively, existed at all anywhere (ca. 250,000 years ago), been respectively Out of Africa (ca. 125,000 years ago), into South Asia and beyond by the Southern Route (ca. 100,000 to 80,000 years ago), or North Asia (ca. 50,000 years ago), and finally, in turn, in NE Asia adjacent to what is now the Bering Strait (ca. 40,000 to 30,000 years ago). And, since we have multiple thresholds, even if one fails that doesn't bring us back to ground zero, so the further back in time we look, the more powerful exclusions based upon Old World Archaeology (which are corroborated by modern and ancient DNA become. And, we can use genetic data (especially pre-Columbian ancient DNA) to largely rule out, for example, the Solutrean Hypothesis, which makes the time frame constraints posed by the earliest human presence in NE Asia stronger than they would be otherwise. And, with paleoclimate data from a variety of sources that help make those estimates robust, we can determine during which time frames crossing from Asia to the Americas via a land bridge and/or short distance maritime activity was possible, because we have good reason to believe that long distance maritime travel had not been invented in that time frame.

So, we have an empirically based methods of determining the extent to which a conclusion is extraordinary or expected. And, in that situation, it isn't unreasonable or unscientific. Also, even to the extent that we can't rule out entirely, for example, modern human occupation of North America prior to the paradigmatic expansion of the Founding Population, we can reach quite a few constraints about what any deviation form the paradigm could have looked like.

andrew said...

For example, "[a] reconstruction of the sea-level history of the region indicated that a seaway existed from c.  135,000 – c.  70,000 BP, a land bridge from c.  70,000 – c.  60,000 BP, intermittent connection from c.  60,000 – c.  30,000 BP, a land bridge from c.  30,000 – c.  11,000 BP, followed by a Holocene sea-level rise that reopened the strait.", and the earliest archaeological evidence of any modern human population in NE Asia is from ca. 40,000 to 30,000 years ago. So, any purportedly modern human presence in the Americas significantly before 30,000 years ago would be very extraordinary. Similarly, we know from genetics that any such population either made a negligible contribution to the existing population of the Americas through almost complete replacement of existing populations (despite the fact that all known long term significant interactions between hominins of evolutionarily adjacent species has always left evidence of admixture even across species lines (human-Denisovan, human-Neanderthal, Denviosan-Neanderthal, Denisovan-preDenisovan and preNeanderthal hominin). So it is highly improbable that the Founding Population of hunter gatherers encountering another hunter-gather population, would have admixed with any discernibly genetically different existing population of the Americas unless its numbers were truly tiny relative to that of the rapidly expanding Founding Population (in which any early introgression of pre-existing residents would have almost certainly been preserved and amplified due to Founder Effects since gene loss due to random genetic drift is greatly inhibited in expanding populations). And, we know that any prior modern human population in the Americas did not, for whatever reason, lead to mass extinctions of megafauna as it did in North Asia and Europe and Australia and New Zealand. And, we know that any prior modern human population didn't build significant structures or any tools more elaborate than unadorned stone tools. And, we know that it didn't introduce or domesticate in situ, any domesticated plants or animals as that would leave lots of archaeological evidence.

andrew said...




Basically, any modern human population significantly prior to 16,500 years ago or so, almost had to have been a very small, very primitive population that was not well adapted enough to thrive and expand in virgin territory or make very distinctively human artifacts (unlike, for example, contemporaneous Cro-Magnon people in Europe or first wave settlers of Tibet, who left behind distinctive human art, intentional grave sites, middens, ecological impacts, etc.) and almost had to have no significant impact on their successor population. So, even "paradigm breaking" finds in the 30,000 to 16,5000 years ago time frame would be little more than an academically interesting footnote in prehistory, and "paradigm breaking" finds purportedly prior to 30,000 years ago would be exceptionally implausible unless the hominins in question weren't modern humans. And, since there is absolute no evidence of H. Erectus anywhere nearly far north enough to cross the land bridge and no evidence that H. Erectus had long distance maritime capabilities, those archaic hominins still have to be within the last 500,000 years and really based upon the absence of North Asian evidence of them in the relevant time frames, they would be much less likely in the pre-135,000 years ago time frame for which we have no North Asian evidence of them than in the 70,000 to 60,000 years ago time frame when there was a land bridge but no modern humans nearby, which case Neanderthals or Denisovans might have crossed it.

All of these background parameters that collectively produce the synthesis of all of the robust evidence confirming itself by multiple means that we call the paradigm, mean that claims of finds that aren't absolutely unequivocal that are outside the paradigm need to not be given the benefit of the doubt if there are other plausible but unlikely possibilities (the likelihood of improbable things happening anywhere is much greater than the likelihood of any given improbable event due to the look elsewhere effect). In contrast, claims consistent with the paradigm, because they are consistent with all other evidence, rather than contradicting multiple other kinds of significant evidence, which should be viewed with skepticism. This isn't just a sociology of science issue.

andrew said...

Does this mean that we should ignore evidence that in contrary to or significantly modifies the paradigm? Of course not. But, if the evidence could potentially have other explanations (e.g. evidence of fire that wasn't a wildfire or very primitive lithic tools that arguably could have been made by hominins or evidence of what looks like butchering of bones), in a time period contrary to the paradigm, we should be very careful to look for any possible methodological defects and any possible alternative explanations.

If we find elaborate cave art and bone cavings in Peru that can be unequivocally dated from carbon or uranium dating of flakes of the paint and fragments of the bone carvings to 70,000 years ago, accompanied by complex fabricated iron tools which can be chemically matched by impurity proportions in the metal to local sources, yeah, that is totally going to blow the paradigm out of the water. Suddenly, a lot of otherwise nutty theories about a very ancient lost civilization of Atlantis or ancient aliens is going to start sounding way, way more plausible than they do now. But, I am very confident that nobody is going to discovery anything out there like that and that it doesn't exist.

Marnie said...

@andrew

"The Baysean prior distribution isn't based on unlimited evidence, but it is based upon a significant volume of data which provides robust expectations because data from lots of different methodologies taken in isolation, and integrated together, tends to point to the same conclusions."

Having been to some sight digs in Alberta, and having talked to some of the archaeologists there, I can categorically say that there is very little effort or research funding to do the kind of archaeology that would allow archaeologists to find sites there below the Clovis level.

And given the deplorable state of underfunded sights on the East Coast that I heard Dennis Stanford describe at the SAA last year, some of which are tentatively dated to before 20,000 years ago, and I do not think the "paradigm" you describe above is constructed from a sincere search for evidence. Hypotheses have to be based on evidence. Yet, archaeologists such as Goodyear, who do dig below the Clovis level, often do find stone tools or bone that show evidence to human intervention. They have not as yet found Clovis points, but associating modernity with Clovis, if that is to be the measure for humans in North America in your paradigm, does not make sense.

Many plains Native Americans never constructed large built structures. Their structures were mobile in order to allow them to take advantage of the geography in which they lived. And they also often hunted without needing stone points. They ran animals over cliffs. They did not need stone points to do this.

Other Native Americans in the Southeast extensively used bamboo for weapons, houses, fishing, etc. They did not need stone points either. By comparison, ethnographies show that Papuans and Australians did not always construct their points from stone.

So, in short, the notion of attaching modernity to built, immobile structures and Clovis/Folsom stone points, is ill constructed, from a scientific standpoint, because it includes only certain types of evidence, and this inclusion (and exclusion) of evidence is biased.


Marnie said...

@andrew

Regarding this comment:

"Genetic data and linguistic data are corroborated by carbon-14 dates of the oldest human remains and undeniable obvious signs of human presence (e.g. structures and carved bone tools and petroglyphs)"

For the Americas, Siberia and Beringia, please state the samples and related dates you are referring to that unilaterally indicate that humans did not arrive south of the Cordilleran-Laurentide Ice Sheet before 16,000 years ago.

Here is the ancient DNA genetic evidence, that I am aware of, so far, related to Siberia and North America, in the time frame of interest:

Yana RHS
https://www.biorxiv.org/content/biorxiv/early/2018/10/22/448829.full.pdf
date: 31.5 kya

Anzick-1
https://en.wikipedia.org/wiki/Anzick-1
date: 12,707–12,556 years BP

Alaska
https://news.nationalgeographic.com/2018/01/alaska-dna-ancient-beringia-genome/
date: 11,500 BP

Mal'ta
https://en.wikipedia.org/wiki/Mal%27ta–Buret%27_culture
date: 24,000 years BP

Kolyma
https://www.biorxiv.org/content/biorxiv/early/2018/10/22/448829.full.pdf
date: 9.8 BP

Tianyuan
https://en.wikipedia.org/wiki/Tianyuan_man
date: 39,000 to 42,000 BP

If you account for the fact that Beringia was unflooded until 11,000 years ago, you would expect that there would continue gene flow between Siberia and North America until 11,000 ybp. Kolyma, in fact, shows this. But this gene flow would have over written the genetic signature of earlier arrivals to North America. Therefore, the correct experiments would look at the structure and gene flow of populations between Asia and the Americas using contemporaneous in time samples between the Americas and Asia.

Since we do not as yet have contemporaneous in time samples in the Americas that could be compared to Mal'ta, Yana RHS, and Tianyuan, I don't think we can say that we have properly run the experiment to look for how and when humans moved between Asia and America in the last 50,000 years. Looking objectively at the last 50,000 years with contemporaneous comparisons between continents would be a good start for these geneticists. So far, I haven't seen that for dates before 13,000 years ago.

What I do see are lots of "eight-by-ten colour glossy pictures with circles and arrows" showing a big jump from East Asians 26,000 years ago to Native Americans 13,000 years ago, but with a complete absence of data for Beringia and the Americas earlier than 13,000 years ago.

https://news.nationalgeographic.com/2018/01/alaska-dna-ancient-beringia-genome/

That's a 13,000 year gap where we don't know where anyone was. And no data for the Americas, not even Alaska and the Yukon, before 13,000 years ago. Yet we know humans were there for more than 10,000 years, based on the Bluefish Cave site.

Also, the proposed models do not account for the obvious back migrations that would have continuously occurred between Siberia and the Americas.

So no, I do not think the current genetic evidence indicates with any degree of confidence that humans reached the Americas south of the Ice Sheets only after 16,000 years ago.

andrew said...

@Marnie

Thanks for that convenient set of links and references.

In a nutshell, you are arguing that absence of evidence isn't evidence of absence, and that absence of evidence can be a self-fulfilling prophecy when it influences what kind of evidence scholarly investigators invest funds to try to discover. And, those are fair points.

"For the Americas, Siberia and Beringia, please state the samples and related dates you are referring to that unilaterally indicate that humans did not arrive south of the Cordilleran-Laurentide Ice Sheet before 16,000 years ago."

I agree with you that we can't entirely rule out a human presence in that time frame, although the older you get, the less plausible a human or hominin presence becomes. For example, we can be very comfortable that any trace of a hominin in the Americas from 200 kya is, at a minimum, not a modern human and not H. Erectus or a more archaic kind of hominin.

What we do have is genetic data that overwhelmingly supports that hypothesis that 99%+ of non-Inuit, non-Na-Dene, pre-Columbia ancestry in both North America and South America derives from a single, roughly contemporaneous wave of migration from a single (possibly mildly structured) source population with a low effective population size.

We can be comfortable, although not quite so overwhelmingly certain, that there was a very low rate of loss of genetic ancestry due to random genetic drift in the Founding population of the Americas during the pre-Clovis time period when it was expanding into (at least nearly) virgin territory at an exponential rate that was pretty much as fast as it was possible to expand in a terrestrial hunter-gather society.

We can be comfortable, although not quite so overwhelmingly certain, that any pre-16000 years ago population that was significantly genetically distinct from the Founding population in a discernible way either didn't exist or was extremely small in population relative to the rapidly expanding Founding population size.

We can be comfortable, although not quite so overwhelmingly certain, that any pre-16000 years ago population was small enough or had a lifestyle of a type, that it did not leave a great ecological impact beyond the impact already arising from the LGM ice age.

We can be comfortable, although not quite some overwhelmingly certain, that the combination of our data points regarding the timing of a modern human presence in Siberia and the timing of the Bearing land bridge being open and the lack of maritime technological then, make any modern human presence in the Americas prior to 30,000 years ago highly unlikely.

We can be comfortable, although not quite so overwhelmingly certain, that any pre-16000 years ago population did not have a very significant or large scale material culture in any medium that would have been distinctive and well preserved over long time periods.

andrew said...

So, while it is true that absence of evidence isn't evidence of absence, I think that there are also some quite meaningful constraints on the size, genetic distinctiveness, and character of any pre-16000 years ago modern human population in the Americans south of the big North American glacier that came into being there during the LGM. A small population in North America that came from a common gene poll with the Founding population within a few thousand years that didn't have a well preservable material culture and didn't really thrive and exponentially expand in virgin territory in the time period from 30,000 years ago to 16,000 years ago s possible, and the closer you get to 16,000 years ago, the less tightly the parameters of what we think we know constrain the possibilities.

For example, it could be that the population structure we see in the Founding population actually reflects two waves of Founding populations, with a most recent common ancestor ca. 20,000 years ago, each with small effective population sizes, one of which arrived south of the ice sheet ca. 18,000 years ago in North America but only started to really thrive around the time of the Clovis culture when improved climate conditions improved food supplies and an improved toolkit allowed them to better exploit improved conditions, and the other of which started expanding 16,000 years ago. A modest tweak to the paradigm like that one is very possible.

But, the bigger the pre-Founding population gets, the more technologically advance it becomes, and the more distinct it is from the Founding population genetically it is, the less plausible it becomes, because any time two hominin species capable of producing fertile offspring together meet there is almost always some discernible level of admixture which we don't see at levels that are detectible outside of a tribes with a combined populations in the 10s of thousand in the far reaches of the Amazon (the mysterious Paleo-Asia DNA signature), the seemingly Paleo-Indian individuals (based on their bones) who have been DNA tested are ordinary Founding population individuals genetically, and the longer a pre-Founding population is present in the Americas when it is virgin territory the less plausible it is that they didn't expand exponentially.

It is very hard indeed, for example, to find a model in which the Americas had a large Paleo-Asian population the entered North America prior to the LGM Ice Age that survived long enough in meaningfully large numbers to experience first contact with the expanding Founding population. History is full of genocides, but it is very rare that a genocide is so complete and so rape free that no genetic trace of the obliterated population is left in the population that commits the genocide. And, the larger the now extinct population was and the more genetically distinct that it was, the harder it becomes to explain why it left no discernible trace of genetic introgression outside the deepest corner of the Amazon (and a small one even there) in pre-Columbian populations of the Americas that were expanding exponentially in population at the time into effectively virgin territory.

Probably the only model that I can imagine where that happens is that this pre-Founding population of Paleo-Asians, for example, we managing to live in harmony but couldn't grow exponentially due to less than ideal conditions during the LGM ice age, and then was wiped out by the same climate conditions that wiped out American megafauna during the Younger Dryas outside of a handful of the most wet and lush refugia in the two continents, because they didn't come up with the technological and food production innovations that the Founding population did that allowed the descendants of the Founding population to adapt and survive after the Younger Dryas. But, that is a pretty grand and wild hypothesis with some very thin evidence indeed to serve as its foundations.

Marnie said...

@andrew

Here goes. I'm going to walk through some of your December 7th comments and give you my thoughts.

"For example, we can be very comfortable that any trace of a hominin in the Americas from 200 kya is, at a minimum, not a modern human and not H. Erectus or a more archaic kind of hominin."

I think a good staring point would be to consider when Middle Paleolithic tools first appeared in SE Asia and Asia. A good paper is the recent Hu et al paper:

http://linearpopulationmodel.blogspot.com/2018/11/late-middle-pleistocene-levallois-stone.html

Levallois tools appear in South China (Guanyindong Cave) site approximately 170,000–80,000 years ago. The reason I am interested in Levallois tools is that these techniques do appear in the tools at Paleoindian sites in the Americas. I don't really know who the makers of these tools are, but presumably, if the makers of these tools diffuse to South China by 170,000 years ago, they could also have made it to the Americas.

In the Hu et al. paper, by MIS 5 (130,000 to 80,000 years ago), there are several Levallois sites in Mongolia and Siberia. Now we're getting close to hominins reaching North America, especially when the sea level is low enough (between 140,000 and 130,000 years ago.)

The next probably time that humans reached North America was after 70,000 years ago. See level was again low enough. Humans are established in Australia and Sulawesi by 65,000 years ago. Sea going up to about 100 miles is established. We know that bone tools appear in Japan at about 55,000 years ago:

http://linearpopulationmodel.blogspot.com/2018/04/geology-and-quaternary-environments-of.html

I don't know if human groups reached the Americans in MIS 6,5,4,3, or 2, but since humans were definitely in Asia by 65,000 years ago, I think it is reasonable to at least look for arrivals to North America before the last Ice Age.

Furthermore, as I mentioned above, all of the DNA (Y-chromosome C and K) and mtdna (A, B, C, D, and X) in the Americans also appear in Aboriginal Australians, as does the type A blood group. And, as we now know, Aboriginal Australians have been in Australia for at least 65,000 years.



"What we do have is genetic data that overwhelmingly supports that hypothesis that 99%+ of non-Inuit, non-Na-Dene, pre-Columbia ancestry in both North America and South America derives from a single, roughly contemporaneous wave of migration from a single (possibly mildly structured) source population with a low effective population size."

But this arrival could have occurred anytime after 65,000 years ago. Beringia Standstill is not required and seems contrived to me.


"We can be comfortable, although not quite so overwhelmingly certain, that any pre-16000 years ago population that was significantly genetically distinct from the Founding population in a discernible way either didn't exist or was extremely small in population relative to the rapidly expanding Founding population size."

What if the "Founding population" you are describing (Anzick-1 like), arrived in the Americas in MIS 5, 4, 3? But had low population density for some reason (like Ice Age or some other reason) and then rebounded after the Ice Age? Why don't you think this is a possibility?

Marnie said...

"We can be comfortable, although not quite some overwhelmingly certain, that the combination of our data points regarding the timing of a modern human presence in Siberia and the timing of the Bearing land bridge being open and the lack of maritime technological then, make any modern human presence in the Americas prior to 30,000 years ago highly unlikely."

With the exception of a few short periods, Beringia was walkable after 65,000 years ago until 11,000 years ago:

http://linearpopulationmodel.blogspot.com/2017/04/similar-meltwater-contributions-to.html

In any case, it is clear that humans in SE Asia and Japan did make sea crossings on the order of 50 km 50,000 years ago.


"We can be comfortable, although not quite so overwhelmingly certain, that any pre-16000 years ago population did not have a very significant or large scale material culture in any medium that would have been distinctive and well preserved over long time periods."

I disagree with you on this one. Humans in SE Asia had rock art, refined stone axes, and boats at least 50,000 years ago.

So that is essentially the basis for our disagreement. I think that Anzick-1 like populations could be derived from a population that first reached the Americas in MIS 4 or 3, and continued to maintain contact with Asian populations up until the flooding of Beringia. And you don't think so. So that is the basis for our disagreement.

I don't think it is worth getting too hypothetical about this (regarding tool complexity, social complexity, effective population size, etc.). What is needed is DNA in the Americas before 13,000 years ago, better archaeology to look below the Clovis level at American sites, precise dating, and a more methodical approach that doesn't make it impossible for people to publish sites with archaeological dates older than 16,000 years ago.

Thanks for your consideration! (and please excuse typos as I have to run now.)

Marnie said...

Correction:

Original comment in my December 7, 2018 at 5:28 PM post:

"Furthermore, as I mentioned above, all of the DNA (Y-chromosome C and K) and mtdna (A, B, C, D, and X) in the Americans also appear in Aboriginal Australians, as does the type A blood group. And, as we now know, Aboriginal Australians have been in Australia for at least 65,000 years."

Amended comment:

"Furthermore, as I mentioned above, all of the DNA (Y-chromosome C and K derived) and mtdna (M and N derived) in the Americans also appear in Aboriginal Australians, as does the type A blood group. And, as we now know, Aboriginal Australians have been in Australia for at least 65,000 years."

andrew said...

@Marnie

I couldn't care less about typos.

""We can be comfortable, although not quite so overwhelmingly certain, that any pre-16000 years ago population did not have a very significant or large scale material culture in any medium that would have been distinctive and well preserved over long time periods."

I disagree with you on this one. Humans in SE Asia had rock art, refined stone axes, and boats at least 50,000 years ago."

My reasoning here is not the humans in SE Asia didn't have those things or that people that long ago weren't that sophisticated as a general proposition, but that if there was enough of a large scale material culture in a medium that was preserved better that we would have found at least a little of it forming a distinct type of archaeological material older than 16,000 years. The old sites that we do have prior to the oldest hominin bones are very primitive and marginal stone tool technologies. They don't have rock art. We don't see refined stone axes or little art objects like the Earth mother figures in Vinca Old Europe (Neolithic, I know, but just illustrating the kind of stuff people made). They don't have large accumulated middens. They don't have bone tools which definitely were known to all non-African H. Sapiens for as long as they were outside of Africa pretty much (certainly by 70kya).

Even if the people digging were actually digging for coal or oil or water or diamonds or gold or dinosaur bones or to make skyscraper foundations or at the bottom of sink holes or deep in caves, rather than actively looking for anthropological archaeology, if there was a lot of really striking material culture that an amateur couldn't mistake for being accidental or animal in origin, we would have found it somewhere and we haven't. Funding matters and can have the effect of blinders, but there are lots of opportunities for unexpected surprises to pop up. King Henry's bones in England were found by people working on a new car park. Anzick-1 was found by some kids in a rural area finding stuff on their property while out playing or working.

The paradigm doesn't encourage looking for human relics older than 16,000 years, but Paleo-Zoologists of the kind featured at "Twilight Beasts" while perhaps less well funded are very much interested in those strata looking for ancient sloths and saber tooth tigers and prehistoric horses (went extinct in North America after they evolved here and spread to Asia), etc.

The absence of evidence as evidence of absence argument, as I note, isn't "overwhelmingly certain" evidence, but there has been enough digging at older layers for all sorts of things that only more subtle traces of modern human presences before the paradigm time period are going to be missed. Obvious traces would have been found. Paleo-Zoologists aren't anthropologists but they aren't uneducated dummies oblivious to what human relics look like either.

andrew said...

In experimental physics, when you see an absence of evidence to contradict the paradigm, you do your best to determine based upon various other reasons you might have missed something that was really there what the biggest thing you were not 95% or more certain to miss would be.

I'm trying to do the same thing, to quantify what the biggest plausible pre-Founding population presence could be given the absence of indisputable positive evidence, albeit in a more qualitative than quantitative manner. There is no way we would have missed a civilization with as much of a material footprint as the Mayans from 25,000 years ago in North America or South America. The aboriginal population of Australia left unmistakable signs of their presence including bones and huge "wildfires" from shortly after their arrival. An unknown pre-Founding population that was light on material culture and had a modest population would be much more likely to escape our attention today than one that had an abundant material culture and a large population.

With the multiple disciplines of knowledge available to us, we can place reasonable constraints on what we could have missed if it is out there and we missed it. They aren't absolute, but they are pretty significant.

andrew said...

One more though about pre-Founding populations. They shouldn't be all that different from their contemporaries in Asia, especially the contemporaries in Asia that are closing to the Bearing Land Bridge (there because while ancient Mariners could cross 50km straights, they weren't making the kind of hundreds and thousands of km trips by sea that the Polynesians did until much, much later, and any small group thrown those kinds of distances by a freak storm wouldn't have a large enough effective population size and gender balance to survive for more than two or three generations). Almost anything we'd see in the pre-Founding period in the Americas ought to have an Asian counterpart of sufficient antiquity.