The dubious association of Anzick 1 with Clovis artifacts

In their paper on the Anzick 1 genome, Eske Willerslev and his associates state several things as facts, which are actually not facts.

For example, toward the end they triumphantly declare that

In agreement with previous archaeological and genetic studies, our genome analysis refutes the possibility that Clovis originated via a European (Solutrean) migration to the Americas.

It certainly doesn’t.

Another example can be found in this statement:

We find that the data are compatible with the Anzick-1 individual belonging to a population that is directly ancestral to the two South American Karitiana samples, as is the case for the Mayan, after masking the latter for recent European admixture.

Willerslev should tell everybody which recent Europeans the Caucasoid admixture in the Maya is from, because it certainly isn’t from the post-Columbian Spanish.

Even their very first sentences are examples:

The only known Clovis burial and associated mortuary assemblage was found in the Americas at the Anzick site, Montana, in 1968. Here, approximately 100 stone tools and 15 osseous tool fragments that are technologically consistent with artefacts of the Clovis complex were found in direct association with the partial fragmentary remains of an infant child (Anzick-1). The human remains were found directly below the Clovis artefacts and were covered with red ochre. Bone from the skeleton was directly dated using XAD-collagen to 10,705 ± 35 14C years BP or 12,707–12,556 calendar years BP, close to the end of the Clovis time period.

The truth is that the association between the Anzick 1 remains and the Clovis artifacts has always been a matter of dispute, although you would never know that from reading the Anzick 1 paper.

And the dates for the Anzick 1 remains aren’t “close to the end of the Clovis time period”, they’re after the end of the Clovis time period.

The artifacts and remains at the Anzick site were discovered by accident in 1968 by Ben Hargis, who was using a front-end loader to load rock debris into a dump truck. After seeing a couple of the stone tools in some of the debris, he stopped working in the area, and later that day he and some friends removed all of the artifacts and human remains by hand.

Later in 1968 the site was professionally investigated by Dee Taylor of the University of Montana. In his report titled “The Wilsall Excavations: An Exercise in Frustration”, Taylor wrote

Unfortunately, the Wilsall (Anzick) material was unearthed in such a way that data from several levels could have become thoroughly mixed… It is unfortunate, too, that our amateur diggers were so thorough and succeeded in finding almost everything that was there, leaving nothing in situ…the potential importance of the site cannot be overemphasized…Had it been possible to make a definite association of the human bones with Clovis materials, it would have given archaeologists their first glimpse of the actual bones of one of these ancient hunters.

In 1999 Larry Lahren interviewed two of the discoverers at the site. The belief that the human remains were located beneath the Clovis artifacts is based entirely on these recollections of events that had taken place 31 years earlier.

One of the most frequently cited papers on the Anzick site is this paper from 2001. Dennis Stanford, one of the two originators of the Solutrean hypothesis, is mentioned in the acknowledgments as having reviewed drafts of the manuscript. The paper states that

The original ledge was capped and sealed by weathered, decomposed fine sandstone, and the bones, reportedly, were found below the lithic artifacts.

Note that this statement is qualified with the word “reportedly”.

Even if the human remains really were found beneath the Clovis artifacts, that still leaves the possibility that non-Clovis people came across a Clovis cache that had been left there hundreds of years earlier, and used it as part of one of their burials. Indeed, the radiocarbon dating evidence indicates that may be exactly what happened.

The 2012 book on the Solutrean hypothesis by Dennis Stanford and Bruce Bradley, Across Atlantic Ice, mentions the Anzick site on page 60:

Another Clovis cache site, Anzick in southwestern Montana, contained more than 132 artifacts and bone tools buried in a collapsed rock shelter. This site was similar to the East Wenatchee Site, as it contained a large number of bifacially flaked stone artifacts, including Clovis points, a few unifacially retouched tools, and bone rods. A notable difference from East Wenatchee was the presence of the partial remains of two subadult human skeletons. The remains of a toddler consisted primarily of skull fragments, and the artifacts were all stained with red ocher. The other individual has been shown to be unassociated with the Clovis materials.

A note for this paragraph on page 265 says that

There are some real problems with these two skeletons relative to their association with the Clovis Cache. The bone foreshafts have a good, solid Clovis age radiocarbon date of 11,040±35 RCYBP. The toddler has a date of 10,680±50 RCYBP, while the other child has a date of 8,600±90 RCYBP. The toddler’s bones are covered with red ocher, but the other child’s remains are not stained. Unfortunately, the entire site, including the artifacts and remains, was disturbed by earthmoving equipment, and the exact locations of the burials relative to the cache are unknown. It may be that they were not associated with the Clovis Cache but were incidentally buried nearby and the red ocher staining the toddler’s bones is purely coincidental.

The Anzick site is also mentioned on page 180:

While it is difficult to use lack of evidence as evidence of a relationship, we must point out that no Solutrean or Clovis burials have been found. Unlike with other Paleolithic European, Siberian, and later North American cultures, which all have evidence of mortuary customs, we have no idea how the earliest Paleo-Americans or Solutrean people buried their dead. Both groups apparently used a form of burial that did not preserve the human remains. Their customs may have been similar to the historic scaffold type of burial, wherein the body was left in an open environment to facilitate transformation to another dimension. Some investigators think the remains of two individuals recovered at Anzick Site in Montana are associated with the Clovis artifacts there. However, radiocarbon dating of the human bones indicates that they were placed near the cache 400 or more years later. The lack of human remains from both cultures renders it impossible to assess their paleo-genetic relationships.

The Anzick 1 paper states that

Clovis, with its distinctive biface, blade and osseous technologies, is the oldest widespread archaeological complex defined in North America, dating from 11,100 to 10,700 14C years before present (BP) (13,000 to 12,600 calendar years BP).

It gives this 2007 paper and this 2008 paper as references, but these date ranges for Clovis don’t appear in either paper.

The supplementary information for the Anzick 1 paper says that

The date on Anzick-1 cranial bone (10,705±35) differs from the average date for the rods (11,035±45). Their respective calibrated age ranges are 12,722 to 12,590 calendar years BP (Anzick-1) and 13,039 to 12,763 calendar years BP (antler rods) — values that do not overlap at the 95.4% confidence interval.

It goes on to say that

The ages of both the ochre-stained cranial fragments and the osseous tools are within the accepted age range of Clovis.

It gives the 2007 paper as a reference for this statement.

The 2007 paper gives a radiocarbon date range for Clovis of 11,050 to 10,800 years BP, and, using the most accurate calibration for the Clovis time period, it gives a maximum range of 13,250 to 12,800 calendar years BP, and a minimum range of 13,125 to 12,925 calendar years BP.

So the date range for the antler rods overlaps the Clovis date range, but the date range for the Anzick 1 remains is after the Clovis date range, contrary to what the Anzick 1 paper claims.

The supporting online material for the 2007 paper states

The Anzick site in Montana is reported to be a Clovis burial and cache. At Anzick, 12 radiocarbon dates were obtained from the cranial elements of a purported Clovis infant skeleton and 2 dates on associated bone foreshafts. Collagen extracted from the foreshafts yielded an average age of 11,040 ± 35 14C yr B.P. The human skeletal remains were dated during three separate research programs. The first batch of seven dates on bone comprise five chemical fractions that were considered reliable and averaged to 10,680 ± 50 14C yr B.P. Later, a single purified collagen sample yielded a date of 11,550 ± 60 14C yr B.P. This measurement is rejected because subsequent dating of the same XAD fraction and preceding fractions from newly sampled bone did not replicate the 11,550 14C yr B.P. result. The source of the contaminating 14C-depleted carbon is unknown. A more recent series of dates from a single cranial fragment provided four new radiocarbon ages. These fractions confirm previous date estimates for the skeleton of 10,705 ± 35 14C yr B.P. The 14C dates on the skeleton versus the dates on the bone foreshafts suggest that the skeletal remains and Clovis artifacts may not be related and that the foreshaft ages more accurately date the site. The 10,700 year old human remains could post-date the Clovis cache, but additional research is needed to resolve this issue. A more recent, late Paleoindian or early Archaic human skeleton was also found at the site. The association of any of the human remains with the Clovis cache is problematic because the site had been excavated accidentally with heavy machinery before the human bones and artifacts were recognized and later recovered at some distance from the actual site. Thus, the directly dated Clovis artifacts—the foreshafts—appear to accurately date the site.

The 2008 paper, citing the 2007 paper, says that

Radiocarbon dates obtained over the last 40 years from Clovis sites across North America suggested that the complex ranged in age from 13.6 to 13 ka; however, evaluation of the existing dates and new 14C assays reveals that Clovis more precisely dates from 13.2–13.1 to 12.9–12.8 ka, a shorter and younger time span for Clovis than earlier thought. The current evidence suggests Clovis flourished during the late Allerød interstadial and quickly disappeared at the start of the Younger Dryas stadial.

This contradicts the date range for Clovis given by the Anzick 1 paper, which even gave the 2008 paper as a reference.

Posted in Uncategorized
9 comments on “The dubious association of Anzick 1 with Clovis artifacts
  1. anonymous says:

    Nice work, Genetiker. I had a suspicion something about the dates was fudged, and you just confirmed it. The date estimates I’ve heard for the opening of the Beringian land bridge are later than Clovis.

  2. genetiker says:

    There’s an animation of the Beringian coastline here. It shows that the land bridge existed up until around 10,000 cal years BP, when the Bering Strait formed.

    And there’s a video of the inundation of the land bridge and the melting of the Cordilleran and Laurentide ice sheets here. It shows that the ice-free corridor had formed by around 14,500 years ago, which is around 1,500 years before Clovis.

    But Stanford and Bradley have long argued that the ice-free corridor was inhospitable to life for a long period of time after its formation. In this paper from 2004 they say

    “After deglaciation, the ice-free corridor may have been barren of plant and animal life for a long period of time, rendering the area unsuitable for human habitation (Burns 1995; Mandryk 1990). In fact, recent research indicates that passage may not have been biologically viable until after 11,000 years ago, too late for use by Clovis ancestors (Jackson and Duk-Rodkin 1996; Wilson and Burns 1999). This being the case, a land route to the Americas was essentially blocked between 22,000 and 11,000 years ago.”

    They maintain this argument in Across Atlantic Ice.

    This page has this map of North American vegetation 13,000 radiocarbon years ago, or about 2,000 radiocarbon years before Clovis, and also this map of vegetation 11,000 radiocarbon years ago, which is within the Clovis date range.

    In any case, the Mongoloid-Caucasoid hybrids from Asia probably moved down the Pacific coasts of North and South America around 15,000 years ago, 2,000 years before Clovis.

    The coprolites at the Paisley Caves in Oregon were dated to 14,300 cal yr BP, and the DNA from them was found to belong to mitochondrial haplogroups A2 and B2, which are of Mongoloid origin.

    The kinds of stone tools found in western North America and western South America are distinct from the proto-Clovis and Clovis tools that are concentrated in eastern North America. In the west one finds thick-bodied, unfluted points, as opposed to the thin-bodied and fluted Clovis points. One of the non-Clovis point types is the Western Stemmed type, which is what was found at the Paisley Caves. Clovis tools are similar to those found at earlier times in Western Europe, while the non-Clovis tools of the western Americas are similar to those found at earlier times in Northeast Asia.

    For information on American tool types, see this page. It shows that the points found at the Monte Verde site in Chile, which is dated to 14,800 cal yr BP, are of the El Jobo type, which is similar to the other western non-Clovis types.

    • anonymous says:

      Thanks for the informative response. I wasn’t aware of the mtdna from the Paisley Caves in Oregon being A2 and B2. That changes my schemas of things, and may explain the Monte Verde site as well. I was wondering what you thought of the following scenario: Seeing as Q-L804 is the closest clade to Q-M3, and Q-L804 has been found exclusively in NorthWest European countries, what if Q-L54 came from both directions, i.e. the Q-L54 that spawned Q-M3 came over with the Solutreans (carrying mtdna C1 and lots of MA-1 ANE), while the Q-L54 that spawned Q-Z780 (carrying mtdna A, B & D and lots of Anzick-1) came from Siberia? But then, that leaves many things still to be explained, such as many Cherokees claiming to be R1b, etc.

  3. genetiker says:

    I’ll respond after responding to the criticism that I received here.

  4. genetiker says:

    There are several problems with this scenario.

    Amerindians south of the US border are almost identical to Anzick 1 autosomally. But Q-Z780 makes up only a very small fraction of their Y chromosomes. This paper looked at a sample of 436 Amerindians from Mexico to South America, and only 2.1% of them were Q-Z780 (subclade Q-L191).

    Q-M3 is overwhelmingly dominant in South America, and there is also a lot of mt C in South America. But I’m only aware of one Clovis site in South America, found on the Caribbean coast of Venezuela. The vast majority of Clovis sites are in the United States, and of those, most are in the eastern United States.

    Tools of the El Jobo type were found at the Taima-Taima site in Venezuela and at the Monte Verde site in Chile, and both of these sites are dated to 14,800 cal BP, 1,600 years before Clovis. So the Mongoloid-Caucasoid hybrids from Asia were definitely the first people in South America.

    If the Solutreans were Q-L54, then we would expect to see Q-L54 in France, Spain, and Portugal, but no Q-L54 has been found in these countries.

    C is a Mongoloid mitochondrial haplogroup. There is no evidence of any kind that Mongoloids were present anywhere in Europe during the Paleolithic.

    The mtDNA from Paleolithic and Mesolithic Europe is almost entirely U. No Mongoloid haplogroups have been found in samples from Paleolithic Europe, and no Mongoloid haplogroups have been found in samples from Mesolithic Europe west of Russia.

    I think that the first Mongoloids to intrude upon European soil were the Y N1c Uralic-speaking Mongoloids of the Mesolithic Kunda culture.

    I think that the Cro-Magnons of the Aurignacian were I and that the Gravettians were R1 Indo-Europeans. I think that the Solutreans were either I, or R1, or both.

  5. […] via a European (Solutrean) migration to the Americas”, which is complete nonsense. In this post I exposed the questionable nature of the association between the Anzick 1 remains and Clovis […]

Leave a Reply to genetiker Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: