probing the mysteries shoop site probing the mysteries shoop site

Volume 26, Number 1 Center for the Study of the First Americans Department of Anthropology

January, 2011 Texas A&M University, 4352 TAMU, College Station, TX 77843-4352 ISSN 8755-6898

World Wide Web site http://centerfirstamericans.org and http://anthropology.tamu.edu

4 Jousting over the coprolitesfrom Paisley CavesExperts tell discoverer DennisJenkins that his pre-Clovis humanexcreta probably are neither. Thesecond of our 2-part series.

7 Mapping an unchartedPleistocene landscape, theLGM coastline of FloridaTowing a sonar array, a boatcrew captained by Mercyhurstanthropologists Adovasio andHemmings is plotting ancienttopography. Dolphins supervise.

12 What’s so hard about makingpottery?East Asia boasted ceramics know-how before its first emigrantspeopled the New World. Why didit take nearly another 9,000 yearsfor pottery to appear in NorthAmerica?

15 The premiere of a series onClovis that’s worth saving!Lithics expert Charlotte Pevnywalks us through what we knowabout the culture that so brieflydominated North America, mostof it inferred from their handiworkin stone.

N A LONELY, HIGH RIDGE in cen-tral Pennsylvania, nine miles east ofthe Susquehanna River, lies an an-

Aerial view of the Shoopsite facing southwest.

Probing the MysteriesShoop Site

Probing the MysteriesShoop Siteof theof the

High above the SusquehannaHigh above the Susquehanna

cient archaeological site that shouldn’teven be there. Traditional logic dictatesthat the Shoop site (36DA20) should neverhave been occupied by Paleoamericans, atleast not to any significant extent: The loca-tion is difficult to reach, there’s no perma-nent water source within 300 m, and thereare no easily extractable natural resourcesthat might otherwisemake the site attrac-tive. “This is an out-of-the-way area, kindof out in the middleof nowhere,” notestoday’s expert onShoop, Kurt Carr ofthe State Museum ofPennsylvania. Yetthere it sits, mute testimony that largegroups of Pleistocene hunters repeatedly

saw fit to make the site their temporaryhome.

It’s reasonable to expect the occa-sional lithic scatter on a high ridge, wherehunters might stop briefly to look forgame and sharpen their tools, but Shoopis no mere hunting camp. In the 60-plusyears since its discovery, the site hasyielded a perplexing lithic assemblageconsisting of approximately 7,000 arti-facts scattered over at least 37 acres. Thecollection is characterized by a large num-

ber of “astoundinglyreworked” flutedprojectile pointsand endscrapers,

and fully 98% of theartifacts are made

from a lithic mate-rial that originates

hundreds of milesaway. Why, when suitable toolstoneswere readily available nearby?

This aura of mystery is one of thethings that make Shoop so very inter-esting—which is why researcherskeep revisiting it and its unique as-

Pittsburgh

PhiladelphiaHarrisburg

Scranton

✪Shoop

site

ALL

PH

OTO

S: S

TATE

MU

SEU

M O

F PE

NN

SYLV

AN

IA

OO

2 Volume 25 ■ Number 4

Mammoth Trumpet, Statement of Our PolicyMany years may pass between the time an important discovery is made and the acceptance of researchresults by the scientific community. To facilitate communication among all parties interested in stayingabreast of breaking news in First Americans studies, the Mammoth Trumpet, a science news magazine,provides a forum for reporting and discussing new and potentially controversial information importantto understanding the peopling of the Americas. We encourage submission of articles to the ManagingEditor and letters to the Editor. Views published in the Mammoth Trumpet are the views ofcontributors, and do not reflect the views of the editor or Center personnel.

–Michael R. Waters, Director

The Mammoth Trumpet (ISSN 8755-6898) is published quarterly by the Center forthe Study of the First Americans, Department of Anthropology, Texas A&M University,College Station, TX 77843-4352. Phone (979) 845-4046; fax (979) 845-4070; [email protected]. Periodical postage paid at College Station, TX 77843-4352 and atadditional mailing offices.

POSTMASTER: Send address changes to:Mammoth TrumpetDepartment of Anthropology, Texas A&M University4352 TAMUCollege Station, TX 77843-4352

Copyright © 2011 Center for the Study of the First Americans. Permission is herebygiven to any non-profit or educational organization or institution to reproduce withoutcost any materials from the Mammoth Trumpet so long as they are then distributed atno more than actual cost. The Center further requests that notification of reproductionof materials under these conditions be sent to the Center. Address correspondence to theeditor of Mammoth Trumpet, 2122 Scout Road, Lenoir, NC 28645.

Michael R. Waters Director and General Editore-mail: [email protected]

Ted Goebel Associate Director and Editor, Current Research inthe Pleistocenee-mail: [email protected]

James M. Chandler Editor, Mammoth Trumpete-mail: [email protected]

Laurie Lind Office ManagerC & C Wordsmiths Layout and DesignTops Printing, Inc. Printing and mailing

Web site: www.topsprinting.comWorld Wide Web site http://centerfirstamericans.com

The Center for the Study of the First Americans is a non-profit organization. Sub-scription to the Mammoth Trumpet is by membership in the Center.

semblage, trying to tease its story out ofthe stone and soil. Thanks to researcherslike Dr. Carr, we may yet come to anunderstanding of why those early Ameri-cans chose Shoop, who they were, andwhere they came from.

Early and influentialAllowing that there are still some signifi-cant gaps in our understanding of Shoop,the site has had a surprisingly largeimpact on the field of Americanarchaeology . . . and not always in a posi-tive way. In fact, researchers’ imperfectunderstanding of Shoop led them astrayfor years because they mistakenly as-sumed it was typical of its type. But thatwas almost inevitable: When JohnWitthoft published the first article onShoop in the Proceedings of the AmericanPhilosophical Society in 1952, he was work-ing in an archaeological vacuum. His wasonly the third major publication on aPaleoamerican site in the eastern UnitedStates—and the other two had appearedjust the year before. It wasn’t even clear atthe time that the fluted-point technologyexpressed at Shoop was a variant ofClovis. Who could have imagined, half acentury ago, a continent-wide fluted-pointtradition?

Witthoft characterized Shoop as anoverlook hunting camp (which it probablywas). Unfortunately, his definition estab-lished the model for Paleoamerican settle-ments for the northeastern U.S. fordecades to come. “People started lookingfor sites on the uplands,” explains Carr. “Itwasn’t until later that we realized theywere more likely to occur in the flood-plains.”

Although Witthoft was the first profes-sional to systematically study Shoop, thesite had been well known to local collec-tors since the 1930s, beginning with agentleman named George Gordon.Edgar Howard, of the University of Penn-sylvania, did some minor excavations in1942 and concluded that the site had novisible stratigraphy. Indeed, there isn’tmuch possibility of finding stratificationat Shoop, because the soil is very thinand almost completely disturbed by agri-cultural practices. In many places thelandowners are literally plowing bed-rock.

Knowing this, Witthoft didn’t bother toexcavate when he arrived in 1950. Instead,

he conducted a surface reconnaissanceon an area of over 20 acres and collected alarge and varied assemblage of lithic arti-facts. He identified 11 artifact concentra-tions, though it’s uncertain whether theywere actual Paleoamerican individual ac-tivity areas or simply the result of decadesof plowing. Witthoft later analyzed 53 pro-jectile points, 400 tools, and 500 pieces ofdebitage from his and other collections.All but a tiny percentage proved to be

made of an imported raw material, whichseemed odd, considering that native jas-pers were routinely used in other localsites. Witthoft identified the material asOnondaga chert, a dolomitic, cryptocrys-talline toolstone that outcrops in westernNew York. Later research confirmed anorigin at Divers Lake, 350 km northwestof Shoop.

Based on his analysis, Witthoft pro-posed a blade-based lithic tradition that he

October ■ 2010 2

called the Enterline Chert Industry, arguing that it was amongthe oldest in the New World and certainly “prior to Clovis.” Histhesis was based partly on the fact that Shoop tools, especiallyfluted points, seemed quite primitive; in reality, they’re justextensively reworked. He estimated the age of the occupationsomewhere in the broad range of 8000–18,000 RCYBP. Eventhough no radiocarbon ages have yet been obtained for Shoop,it’s clear today that Enterline is related to Clovis, and it has sincebeen subsumed into that cul-tural continuum.

Post-Witthoft ShoopShoop, particularly its lithicassemblage, attracted agood deal of attention in theyears following Witthoft’s in-vestigations. Alex Krieger(1954), Edwin Wilmsen(1970), Stephen Cox (1972and 1986), Gary Fogelman(1986), and Kurt Carr (1985)have reexamined the lithicassemblage. Cox’s analysisof the Paleoamerican toolswas particularly extensive;both he and Wilmsen fo-cused on the edge angles ofthe tools, especially the 500 or so endscrapers. They concludedthat the scrapers were mostly used for woodworking, a hypoth-esis that was recently confirmed by microscopic use-wearanalysis of 14 endscrapers.

Witthoft’s definition of Enterline as a blade-production in-dustry didn’t withstand scrutiny. Carr con-firmed in analyses published in 1989 and2008 that too many of the flakes recoveredclearly aren’t blades at all, just blade-likeexpedient flakes, and the assemblage oth-erwise bears none of the hallmarks ofblade-production technology. Cox, DennisStanford, William Gardner, and othershave concluded that Witthoft’s pre-Clovisage estimate was also incorrect; they ar-gue, for example, that the fluting tech-nique dates the points to about10,700–10,900 RCYBP. These recent revi-sions of Witthoft’s conclusions shouldn’tbe interpreted as a criticism of his analyti-cal skills. They simply reflect the benefit ofdecades of subsequent discoveries aboutwhat constitutes Clovis and the age of itsvarious technological aspects.

Carr and his colleagues from the State Museum, Mercy-hurst College, and Clarion University of Pennsylvania revisitedthe site in 2008, conducting both an extensive surface collec-tion and excavating a number of test units. “We spent 12 weeksexcavating there,” he recalls. “When we were there, we werecelebrities. . . . Collectors and other people came to see us,including people who were with Witthoft in the ’50s.” They also

attracted the attention of film crews; excerpts of the filmedreports can be seen on YouTube.

These investigations, the first professional, systematic workconducted at Shoop since Witthoft’s investigations, yielded 800new artifacts, of which all but about a dozen were microdebitage.Although a few large artifacts may survive in unplowed areas,Carr concludes that most of the large tools have already beencollected from the site. “After Witthoft, there was lots of collec-

tion at the site,” he notes, “and people were paying moneyearly on for artifacts from Shoop.” Carr is quick to pointout, however, that all four of the current landowners areexcellent archaeological stewards, and are doing all theycan to protect Shoop by not allowing any additional artifactcollecting.

Why there?Various theories have been postulated to explain the oddlocation of the Shoop site. Witthoft’s assessment of thesite as an upland hunting camp is probably spot on;based on paleoenvironmental research, the site was ap-parently open grassland during the days of its initialhuman occupation, and then as now it offered “a com-manding view” of the nearby Armstrong Creek valley.

Fluted points (top row), endscrapers (middle row),and sidescrapers (bottom row) from the Shoop site.

continued on page 11

Fluted points from the Shoop site.Extensive reworking accounts forthe great variation in morphology.

Furthermore, there’s a natural cul-de-sac east of the site thatwould have been useful for trapping game.

“We have close to 90 fluted points from the site, and there are500 endscrapers and 1,500 other tools,” Carr reflects. “Thesewere tools broken in use, not in production. It’s hard for me to

imagine the site didn’t involve someform of hunting and processing.” Hesuspects that caribou was the pre-ferred prey, though elk is also a possi-bility. Deer wouldn’t congregate inlarge enough groups to be worth suchan effort, and although nearbyswampy areas might have supportedmoose, they tend to be solitary ani-mals. Even though no one has foundfossil remains of caribou in Pennsylva-nia, Carr assumes they were presentabout 10,000 RCYBP, when they were

definitely in neighboring New York.Which begs the question, How old is Shoop? The answer to thatis still up in the air. About a third of the tools display potliddingconsistent with burning in fires, but poor archaeological con-text and lack of charcoal preclude attempts at radiocarbondating. The fluted points are diagnostic of late Clovis, but that’sas close as archaeologists can place the time of occupation.

0 5cm

0 5cm

0 5cm


NY CLAIM of a pre-Clovis component at an archaeologi-cal site is going to be controversial, and the PaisleyFive-Mile Point Caves site is no exception. From the

rainwater, could have contaminated animal coprolites withhuman DNA. Contamination by excavators and lab workersalso cannot be ruled completely out. For Poinar, the paucity ofassociated lithic artifacts suggests that the AMS dates ob-tained on the coprolites are unreliable: The coprolites prob-ably aren’t human and are probably Clovis or post-Clovis inage. The debitage, like the DNA, could have been bioturbateddown from layers higher up.

What they need, Poinar says, is “more samples and muchmore DNA data at quantitative levels that would argue for the

Paisley CavesPaisley CavesPaisley Caves

Cave 5excavations, 2003.

Part II:The Real Deal orScat-illogical Nonsense?




UN

LESS

NO

TED

, A

LL P

HO

TOS

CO

URT

ESY

UO

NO

RTH

ERN

GRE

AT

BASI

N F

IELD

SC

HO

OL

analyze the coprolites are uneasy about the findings. Tworebuttals followed from the article in Science, one critiquing thegenetic results, the other questioning whether the coprolitesare from humans at all.

Questions about methodology from a masterOne researcher who isn’t uncomfortable with the technologyis McMaster University geneticist Hendrick Poinar (MT 20-3, “Ancient DNA: A Tough Nut to Crack”). Dr. Poinar hasmade a distinguished career analyzing ancient coprolitesfrom humans and animals from all over the world. So, notsurprisingly, his first response to the article in Science was“Cool story. I’m glad they’re using feces!” But, in the end, hewasn’t swayed by the evidence. In a comment on the PaisleyCaves story for Science, Poinar and his colleagues argue thatthe methods used to analyze the coprolites weren’t rigorousenough to rule out a nonhuman source: Human genetic mate-rial from a later date, percolating down from above via urine or

moment Dennis Jenkins, a Senior Staff Archaeologist at theUniversity of Oregon Museum of Natural and Cultural History,and his colleagues announced in the 2008 issue of Science thatthey had recovered pre-Clovis human coprolites (mummifiedfeces) from the caves, a firestorm erupted. Every aspect of theoriginal study has come under scrutiny, and many archaeolo-gists unfamiliar with the complex genetic technology used to

AN

DRE

W C

URR

Y

Jenkins in Cave 2 during filming of the HistoryChannel documentary “All About Dung.”

DNA being endogenous and not due to possible leaching” fromabove.

In their rebuttal, Dr. Jenkins’s team rejects these criticisms:The laboratory protocols used were the same employed byPoinar in his own research; multiple lines of evidence suggestthat leaching of more modern mtDNA into the coprolites is

AA

October ■ 2010 4

“unlikely”; and the original genetic tests were sensitive enoughto pick up contaminating DNA, so the absence of such DNA isnot due to error but is an attribute of the coprolites. Further-more, they point out, the AMS dates were run on fibers in thecoprolites, not on bulk samples, so the problems with the AMSdates raised by Poinar’s group don’t apply.

Are the coprolites even of human origin?The second critique published in Science in 2008 focuses on thehuman attribution of the coprolites. Boston Universitygeoarchaeologists Paul Goldberg and Francesco Berna, whoworked on samples fromone of Jenkins’s coprolites,argue forcefully that thecoprolite samples theyhave examined aren’t hu-man. To these research-ers, who have workedextensively with herbivorecoprolites, the stainedouter surface and high fi-brous plant content in thetwo samples they analyzedwere more consistent withsamples from herbivoresrather than humans. Inthin section, the coprolite samples are poor in calcium phos-phate and undigested celluose is absent, conditions consistentwith herbivore dung and the reverse of what would be expectedin the scat of carnivores or omnivores such as humans.

Dr. Berna is emphatic and uncompromising: “The composi-tion and morphology [of the coprolites] is not compatible withhuman excrement. We have comparative reference samples ofcoprolites from carnivores, humans andherbivores. When you compare, it is obvi-ous. The coprolites that we analyzed arecomposed of degraded plant materials—grasses. We don’t have seeds or grains orpollen, just chewed grass. This is compat-ible with the diets of grazers and herbi-vores, but not plants humans would eat.”

Dr. Goldberg agrees, saying that “todemonstrate that the coprolites are hu-man, Jenkins would have to prove the DNAis not contaminated. Also, they wouldhave to explain why people are eatingfeathers and fibers and nothing else. Ifthey were hunting, you’d expect to findother things in the coprolites.”

However, Jenkins’s team counters that human coproliteswould be expected to be similar to herbivore dung if the humandiet is rich in fibrous foods, as it was in the prehistoric GreatBasin and other regions of the world. In their rebuttal toGoldberg et al. (2008), they present data on coprolites recovereddirectly from Nubian mummies dating from A.D. 550–1450 col-lected by Jenkins’s collaborator Dr. Linda Scott Cummings(Paleo Research Institute). Texas A&M University paleobotanistand palynologist Vaughn Bryant verifies that a similarly high

fiber content, including large quantities of phytoliths, is commonin human coprolites. The researchers argue that low levels ofcalcium phosphate have been observed in other Great Basinhuman coprolites. Finally, they argue that the genetic data cansettle the issue: If the Paisley Cave coprolites are of herbivoreorigin, human mtDNA shouldn’t be present, but herbivore andplant mtDNA should be. However, as Rasmussen, Cummings,and others show (2008), the coprolite that they examined for thisstudy contained only human and plant mtDNA, indicating that thecoprolites were not deposited by herbivores such as bison,horse, or camel. Plant DNA (cpDNA) in the coprolite matches to

lodgepole pine (Pinus contorta) andneedle-and-thread grass (Hesperostipacomata). These plant species were alsofound as pollen, macrofossils, orphytoliths in the coprolite.

Like many archaeologists, perennialskeptic Stuart Fiedel is hungry formore data to help evaluate the claims

The protocol for collecting coprolitesamples borrows techniques fromhandling hazard materials, excepthere the aim is to protect thecoprolites from the person.

Mt. Mazama volcanic ash(7640 CALYBP) near top ofstrata in Cave 2.

that the Paisley Cave coprolites are human: “I’d like to see adetailed description of their contents and dates, and a compari-son with indisputable human turd samples from the region.You would expect some considerable difference between dietsof the first people and of people adapting to early-Holoceneconditions in the region. There’s something suspicious if theylook like 7000 B.C. turds.”

None of these criti-cisms sways University ofWashington zooarchae-ologist Don Grayson. Dr.Grayson has studied fau-nal remains from cavesites throughout theGreat Basin. When hesaw the coprolites for thefirst time, he immediatelythought “yes [these are

from humans], there’s nothing that I can think of that wouldlook like that—like humans on a high-fiber diet.”

Even iron-clad stratigraphy doesn’t satisfy criticsAlthough researchers are still divided over the coprolite ori-gins, almost no one questions the site stratigraphy. Photos ofthe strata show horizontally bedded sediments in the lowerlevels with good lateral and vertical integrity. Extensive AMSdating of organic materials in the lower levels of several of the


caves by geochronologist Dr. Tom Stafford (Stafford ResearchLaboratories) validates this relatively undisturbed strati-graphic sequence. A total of 76 dates have been obtained thatare older than 10,000 CALYBP. The transitions from the earlyHolocene to the Clovis-age deposits and from these to stratacontaining pre-Clovis artifacts are firmly established. Jenkinshappily observes: “Not only do I have a zillion dates, they arethe best I can pay for.”

Absent publications providing detailed stratigraphic data,however, Goldberg is cautious, stating that it is difficult toevaluate the site. “In the future, I would want them to publish alot more detailed stratigraphic data” than the team has to date.He would particularly like to see close-up photos that documentexactly which strata the coprolites are coming from, accompa-nied by micromorphological data detailing the origins of thesediments from which the coprolites have been retrieved.

But, as Texas A&M University geoarchaeologist and CSFADirector Mike Waters points out, well-dated strata are only the beginning pointfor assessing site integrity. The role ofpackrats in the formation of the archaeo-logical site remains unresolved: “I vis-ited the site and looked at the evidence.There is stratigraphic integrity there.The deposits are intact, but is the siteintact? Were the coprolites brought intothe deposits by packrats? Did people usethe shelter as a latrine, or did they re-lieve themselves outside and thepackrats retrieved the materials and putthem in their nests?”

Dr. Waters is also very cautious aboutthe lithic assemblages, concerned aboutthe possibility that the lithics entered thepre-Clovis strata through bioturbation.Recognizing that local obsidian abundant

of site is it if it’s just coprolites? Does it represent two or threehunters on an expedition, holed up in the shelter for an evening?Did they camp adjacent to the shelter?” In his characteristicallyblunt way, Dr. Fiedel echoes this concern: “Behaviorally, thedata suggest that the rockshelter is occupied by people who donothing but relieve themselves in it and then move on.”

“The thing that amazes me most about Paisley Caves” con-tinues Waters, “is the preservation of an ephemeral occupation.Because there are only a few coprolites and artifacts in the pre-Clovis level, it suggests that people were there for a shortperiod of time, perhaps only a few days. To have this ephemeraluse preserved for thousands of years, and then to find it is mind-boggling. What are the odds of preserving and detecting suchan ephemeral occupation?”

Jenkins is tightening up, but not giving upMore excavation always seems to be the panacea for criticism.

New excavations in 2009 and 2010 un-covered pre-Clovis-age deposits in Cave2 and expanded the excavations of thesestrata in Cave 5, yielding more coprolitesfor analysis as well as additional faunalremains and an expanded collection oflithic debitage. One excavation unit pro-duced horse and camel bones in thesame level as a handful of lithic debitage.

To amass the most convincing evi-dence of human occupation from hissite, Jenkins has teamed up with a hostof researchers who are recognized ex-perts in their fields: Geneticist EskeWillerslev of the University of Copen-hagen is heading up the coprolite DNAresearch; Dr. Bryant is looking for pol-len, phytoliths and other botanical re-mains from the reconstituted fraction;Dr. Karl J. Reinhard of the University ofNebraska is looking for human-specificparasites; Brian Hockett of the Bureau ofLand Management is analyzing thefauna; Tom Stafford is producing a long

series of additional radiocarbon dates at the site (94 dates andcounting!); and Oregon State University geoarchaeologist Dr.Loren Davis is collaborating with Dr. Stafford on the stratigra-phy.

In response to criticisms that the DNA in the coprolites ana-lyzed so far has come from modern contamination, Jenkins hasextensively modified his methods for handling coprolites todramatically reduce the likelihood of modern contamination:Coprolites, now handled only by individuals clad in Tyvek suits,are immediately sealed in sterile containers and shipped directlyto Willerslev’s genetics lab. Willerslev handles the split, sendinga fraction on to Bryant for reconstitution and further analysis.

To counter criticisms surrounding the authenticity of thehuman attribution of the coprolites, and to settle matters relatingto diet, Jenkins is expanding his analysis of the coprolite con-tents. He is sending portions of the recently excavated coprolites

Student Cassie Albush holding aCamelops vertebra found in

deposits dating to 14,300 CALYBP.

in pluvial lake gravels would have been a source of toolstonethroughout site occupation, Waters suggests refitting. “If refit-ting could be documented within the pre-Clovis assemblage, andthere were no refits between the pre-Clovis assemblage andyounger assemblages, this would be very compelling” evidencethat bioturbation is not bringing younger lithics down into olderstrata. This would nicely complement the obsidian hydrationdata, which show consistent but overlapping increases in hydra-tion rind thickness with depth.

Not all archaeologists feel the issue of site formation pro-cesses is critical to establishing the antiquity of the humancoprolites. As Grayson points out, “You don’t have to worryabout the stratigraphy since you have clearly human coproliteswith direct radiocarbon dates. There is nothing wrong with thedates and nothing in print that questions the dates.”

Perhaps the driving undercurrent of concern about the Pais-ley Caves centers on the site contents. Waters asks, “What kind continued on page 10

October ■ 2010 6

Jacksonville

DaytonaBeach

Tampa

Florida MiddleGrounds

Modern shoreline

Late-Pleistoceneshoreline 20,000 years ago

Miami

0 100

Miles

HEN YOU GET RIGHT DOWN TO IT, all archaeology is akind of detective work: we dig up clues (often literally),then piece them together in an effort to understand what

Enter NOAAResearchers have been itching to study the submerged conti-nental shelf since at least the 1960s, when it was first suggestedthat the earliest Americans must have utilized it heavily when itwas exposed. At the time, however, the cost would have beenastronomical—and the technology simply wasn’t there. That’sno longer the case; it’s now possible for scientific desire andtechnological feasibility to meet somewhere in the middle, andnot surprisingly, they have begun to do so.

Perhaps best known for his work at MeadowcroftRockshelter, a pre-Clovis site in southwestern Pennsylvania,Dr. Adovasio has stood at the vanguard of First Americans

studies for decades, steadily working to advanceour knowledge of these elusive people (MT 25-1,-2, “Paleo Woman: Lost to History”). The opportu-nity to extend his reach from the terrestrial to thenautical occurred at a recent meeting of the Societyfor American Archaeology, when he struck up aconversation with NOAA (National Oceanographicand Atmospheric Administration) archaeologistFrank Cantelas, who is associated with their under-water grants program. Adovasio knew that NOAAfunded shipwreck studies, but was surprised tolearn that they wanted to look in other directions aswell. No, the NOAA archaeologist told him, they hadnever sponsored a systematic scrutiny of Pleis-tocene underwater landforms. “So in 2007,”Adovasio recalls, “we put together a proposal toscrutinize an area west of St. Petersburg in order to

PrehistoricFlorida

Submerged

PrehistoricFlorida

Submerged

PrehistoricFlorida

Submerged

PrehistoricFlorida

SubmergedFinding Cultural Clues

on the Continental Shelf

life was like in bygone ages. For the sleuths investigating theinitial peopling of the Americas, one of the biggest obstacles toprogress has always been the fact that not only is the trail stonecold, but much of it also lies underwater. Until recently, techno-logical limitations have thwarted our efforts to overcome thischallenge; but fortunately, that’s beginning to change. For thepast few summers a team of dedicatedresearchers, led by James M. Adovasioand C. Andrew Hemmings of Penn-sylvania’s Mercyhurst College, has com-bined new technology with plenty ofold-fashioned hard work to coax the seainto giving up some of its secrets.

What they’ve found is still prelimi-nary—but as the technology maturesand the project evolves, it has the poten-tial to stand American archaeology on itshead.

Water, water everywhereBy the height of the Last Glacial Maximum(LGM), about 22,000 RCYBP, the intense coldhad locked up much of the Earth’s seawaterin miles-thick continental ice sheets. As aresult, oceanic coastlines lay as much asseveral hundred kilometers farther afieldthan they do today, exposing a much greaterland area to exploitation by both wildlife andhumans. It stands to reason that at leastsome of the earliest American cultureswould have exploited the landscape and re-sources along the ocean’s edge . . . but theevidence for these occupations is almost im-possible to find, given the geologicalchanges that have occurred since.

As the ice retreated far to the north, it reshaped the geogra-phy of the post-Pleistocene world in startling ways. Glacialscars became the Finger Lakes of upstate New York and theSeven Rila Lakes of Bulgaria; fine silt from glacial outflowsblanketed vast tracts ofEurasia and North America;and the land itself rose up inisostatic rebound, no longercrushed under countless tonsof ice. Perhaps the most wide-spread change, and arguablythe most frustrating one forFirst Americans researchers,was the 75- to 120-m rise insea level that resulted asthe ice melted.

Today, many of themost interesting Pleis-tocene landscapes lie be-neath the sea.

WW


topography than either can alone. “There was a steep learningcurve in determining which methods and hardware would pro-duce the best results,” Adovasio admits, “and we’re still in theprocess of finding them. We haven’t created a revolutionarypiece of machinery or anything like that, but I think the waywe’re working with the current technology and pushing itslimits, we’re getting the kinds of insights into prehistoric sub-merged sites that we’d hoped to get when we proposed thisenterprise.”

Their effectiveness took a step forward in 2009, when theywere able to tweak the technology in ways that not only in-creased resolution, but also collected data in half the time.“This lets us see in real time what the seafloor is like, and whatis immediately below it,” notes Andy Hemmings, Adovasio’s co-Principal Investigator. “It essentially cuts an electronic back-hoe trench directly below the boat, in the middle of a side-scansonar swath 100 to 300 yards wide.”

Adding to the rigors always associated with undersea ar-

look at everything from the LGM beach to the modern shore-line, to study what they might look like under their modernbiomantle, and to look for things like prehistoric river chan-nels, sinkholes, and the like.”

When fieldwork began in July 2008, Adovasio’s team focusedon three areas: the edge of the continental shelf; the Florida

Adovasio preparing to hand an underwatercamera to diver Jessi Halligan.

NO

AA

JAM

ES M

. A

DO

VA

SIO

Above, the towfish; below, a chert nodule from aninundated outcrop off the coast of Florida.

NO

AA

Middle Grounds (FMG) and, to the west of that, the ancientClovis shoreline; and the edge of the FMG, which corresponds tothe end of Clovis. The team looked at all those areas the firstyear, quickly locating the remains of the LGM shore (whichretains a series of remarkably intact near-shore features), as wellas what they interpret as karstic sinkholes or cavities that maycontain archaeological materials. They also identified the PaleoSuwannee—that is, an extension of the existing Suwannee Riverchannel across the continental shelf. In 2005, other researchershad traced the channel out as far as 20 km; Adovasio’s teampicked it up about 25 km out, and by the end of the 2009 fieldseason had traced it as far as 160 km into the Gulf. In addition,they found the ancient channel of the modern St. Marks River,and located several submerged chert outcrops that might haveserved early humans as toolstone quarries.

Smart uses for new gadgetsSince the beginning of the project, the team has creativelycombined sub-bottom profiling and side-scan sonar technologyin ways that provide a better view of the ancient undersea

chaeological work, logistical issues required the team to workon a grueling schedule. “Our crew [in 2009] consisted of oneresearch diver and eight archaeologists: three Ph.D.s, fourgraduate students, and one undergraduate student,” explainsTexas A&M grad student Jessi Halligan. “To maximize expen-sive boat time and equipment rentals, we were working on a 24-hour schedule. This meant that every day, everyone would takeat least one shift monitoring the remote sensing gear duringthe night, while we would dive during most days.” Halligan was

October ■ 2010 8

NO

AA

JAM

ES M

. A

DO

VA

SIO

JAM

ES M

. A

DO

VA

SIO

a diver, and also one of two crew members charged withmanaging data and maintaining the towfish, the electronic datacollector towed behind theship.

Unexpectedly, the towfishacquired a retinue of under-sea admirers: inquisitive dol-phins, which interfered withthe painstaking work of map-ping the ocean floor. “We hadas many as 45 dolphinsaround it at times, so welearned a new hazard to archaeological fieldwork,” Hemmingsreports. Meddling dolphins eventually put the towfish out ofcommission, and the team had to dive to investigate interestingspots.

The team first dove in 80–105 ft of water, and found onlylimestone outcroppings; but at sites 45–65 ft deep, they hitchert bedrock immediately adjacent to the river channels.Hemmings estimates that those locations last saw the light ofday at least 8,000 years ago. One of the most interesting sites,located where the intersection of the Paleo Aucilla and an-other river form nearly a rightangle, is a feature that suggestschert-rich subsurface deposits.The team calls the site Thor’s El-bow. “In the few places we know oflike that in Florida,” Hemmingspoints out, “there are massivePaleo sites.” Early occupationsites on shore were almost invari-ably chosen with ready access tofreshwater and toolstone, and it’slikely the same criteria would ap-ply to occupation sites out on thecontinental shelf.

The way forwardPrehistoric inundated archaeology has never been easy, andthe complications involved in deep dives are particularly frus-trating. In some cases, limited air supplies and the need toavoid decompression diving severely limit the time divers can

spend on the bottom. At Thor’s El-bow, for example, Hemmings wasable to stay down only nine minutes,just long enough to pound a core, capit, and hand it off to Halligan to take tothe surface. Furthermore, the dailyexpenses of this type of fieldwork arestaggering; hence the tight 24-hourschedule and field seasons of just a

few weeks.Scientifically, these are early days yet. Nevertheless, data

collected so far make it possible to form amodel that will narrow the team’s focus infuture seasons. As Adovasio explains, “Ithink we’re homing in on three kinds ofsite setting situations: sinkholes and

A B

A, side-scan sonar image of asegment of the Paleo SuwanneeRiver valley; nothing is visible.B, sub-bottom profile image ofthe same area, clearly showingthe river channel.

Record of bottom-imaging scan showingsubsurface sand ridges preserved nearthe LGM beach.

Dolphins off the starboard bow.

karstic features, which we know accumulated late-Pleistoceneand Holocene materials in terrestrial locations; river-edge loca-tions, which might contain materials in overbank deposits; andchert outcrops. In the three places where we thought we mightfind chert, we did find chert. We’re comfortable with the accu-racy of where we think such sites ought to be; now it’s a matterof examining them thoroughly enough that we’re reasonablycomfortable we can find cultural materials, and then being ableto determine what those cultural materials tell us.”

In addition to honing their technological edge with sub-bottom profiling and side-scansonar equipment, the teammembers are working to extendthe limits of their diving capabil-ity. According to Hemmings,they’re slowly but surely be-coming technical divers. Theirnext goal is to reach as deep as165 ft without decompression,then 225-plus ft with mixedgases—before going evendeeper. “Where the LGM coastwould have been is about 365

feet deep,” Hemmings points out. “That’s awful deep for diving.Do we really expect to find people there? No, because we don’thave people in the Americas 22,000 years ago . . . but thenagain, how many people have looked at soils that age? Our work


has a tremendous potential to push the earlyend of when humans entered the NewWorld, simply by looking at areas people

Hemmings (left) and Adovasio at seaduring the 2009 field season.

Crew members with bottom-imagingdisplays during the 2008 field season.

Hemmings is second from left, JessiHalligan is at the extreme right.

cal sites. They intend to goback to some of their high-potential locations with adredge, to remove the biologi-cal cover and look for intactlate-Pleistocene sedimentsand associated archaeologicalmaterials. They’ve targetedseveral sites for the summerof 2010; if one pans out, theyplan to excavate the followingsummer.

Ultimately, Adovasio asserts, NOAA needsto interface with other agencies like theNational Science Foundation to underwriteactivities on a scale that aren’t just “pre-liminary shots in the dark.” That’s why, hesays, publications like Mammoth Trumpetare important. “One of the things you do isbring continuing archaeological research toa wide and diversified audience, as it’s beingdone. What we’re trying to do is not only toreach the scholarly community, but to trans-mit this to the lay public . . . actually pursu-ing the educational moment in the broad

perspective.“I think the most important vehicle for doing that is sitting

down and talking to people like you, the way I’m doing now.”–Floyd Largent

BOTH

: N

OA

A

How to contact the principals of this article:C. Andrew HemmingsMercyhurst Archaeological InstituteMercyhurst College501 East 38th StreetErie, PA 16546e-mail: [email protected]

Jessi HalliganDepartment of AnthropologyTexas A&M University, 4352 TAMU

to labs that specialize in analyzing zooarchaeological materialssuch as bone fragments, insects, and fish scales; macrobotanicalremains such as seeds, pollen and phytoliths; and the presenceof human-specific and other parasites, bacteria and viruses.

In spite of this, Berna is not convinced that contaminationfrom later or modern DNA could ever be completely ruled out. Herecommends that Jenkins “keep excavating in nice contexts.Tools would be stronger than the genetics. Contamination isalways there with genetics. Obsidian hydration works in someplaces, not others. But there is such good organic preservation in

Paisley Caves

continued from page 6

the caves, Jenkins should rely on dates on organic materials” tobest establish a pre-Clovis presence in the Americas.

The recent seasons have been productive, and visitors tothe site have been impressed. New excavations have in-creased the number of tools in the pre-Clovis strata. Shouldthey prove to have some sort of integrity as an assemblage,concerns about the coprolites-only nature of the site shouldwane. Jenkins continues to invite friends and skeptics to visitthe site. It’s perhaps worth noting that his strongest critics—Poinar, Goldberg, Berna, Fiedel, and anthropologist Dr. GaryHaynes of the University of Nevada, Reno—have yet to makethe trip.

Hamilton College Paleoindian researcher Dr. G. Tom Jonesremarks that “if my understanding is correct, Jenkins is nowrecovering materials from another of the caves that are equally

College Station, TX 77843-4352e-mail: [email protected]

James M. Adovasio, ProvostSenior Counselor to the PresidentDean, The Zurn School of Natural Sciences & MathematicsDirector, Mercyhurst Archaeological InstituteMercyhurst College501 East 38th StreetErie, PA 16546e-mail: [email protected]

could have occupied thathaven’t been previously ac-cessible.”

Adovasio agrees, pointingto two discoveries that havebeen especially significant:verifying that existing riversextend well out onto the conti-nental shelf, and finding andsampling chert outcroppings.“If we’d been permitted tohave single-malt scotch onboard,” he muses, “that wouldhave been an occasion to sample it!”

In the upcoming field season, Adovasio and his team in-tend to study those locations in greater detail and therebyestimate the potential for identifying preserved archaeologi-

▲

▲

October ■ 2010 10

old as the original coprolites. They have also recovered a smallassemblage of artifacts. Mostly flakes, I believe, but still a niceadjunct to the fecal remains. This material is coming from well-stratified deposits of fine-textured materials where there seemsto be little disturbance and where stratigraphic mixture ispretty easy to detect when it occurs. I think it’s getting harderand harder to find problems with this record.”

Waters probably speaks for the majority of archaeologistswhen he comments that “Dennis is doing a good job. I like thefact that he has brought in a solid scientific team. The rightpeople are working on the site. But before we can pass final

judgment on the site, we need a good site report. It is incum-bent on Dennis and his team to get the data out there so peoplecan decide about it.”

Jenkins agrees, saying, “I just want this to be science.There’s no room for this to be personal.”

–Ariane Oberling PinsonRenaissance Science Consulting

Departments of Anthropology and Geography,University of New Mexico

Departments of Anthropology and Geography,Central New Mexico Community College

A downside of Shoop, compared with other regional Paleosites like Vail, Debert, and Bull Brook, is that any hearthfeatures would have been obliterated by almost two centuriesof plowing. Carr holds hope for a small swale that apparentlyhas escaped the plow, but he hasn’t been able to examine itsince his last visit in 2008. “I think there’s charcoal at Shoop,”he says, “it’s just a matter of finding it.”

A compelling possibilityGiven the evidence, Carr has formed an intriguing hypothesisthat neatly encapsulates the mysteries surrounding Shoop. ThePleistocene occupants, he believes, entered Pennsylvania fromthe Great Lakes region while following migrating caribou, mostlikely during the YoungerDryas climatic interval. He’sstill unsure whether one groupor successive groups campedat Shoop as they drifted south.“I think the multiple concen-trations suggest it was occu-pied by groups of 25 to 30people, possibly separategroups, over a period of aslittle as 50 years,” he says. “Wedon’t have good horizontalcontrol of the artifact assem-blages, but when we comparedifferent concentrations, wedo see some noticeable differ-ences.”

Most of the lithics are made of Onondaga chert simply be-cause the occupants came to the site carrying that material.“They probably arrived from the north, which is where the stonecame from, so that’s what they used,” Carr theorizes. “Theyweren’t familiar with the local materials.” This is another pointthat argues for large migrations across a relatively short timeframe: Occupants familiar with the area would have had a betterknowledge of local lithic sources. As it is, all the tools from Shoopwere resharpened and reused until they couldn’t be usedagain—“characteristic of what happens to a tool assemblagethat’s been moved 200 miles from its source,” notes Carr.

Where do we go from here?Although the Shoop site has been well known since the 1950sand studied repeatedly since, its origins remain enigmatic. Whydid Paleoamericans visit the site, especially in such numbers,and why did they reuse foreign toolstone to such an extent whenlocal lithic sources were readily available? Carr’s theory of theirorigins is logical and compelling, but verifying what was happen-ing there requires further work, especially in areas of the sitethat haven’t been disturbed by plowing. Carr is confident this willeventually occur; and when it does, we may finally have someanswers to our questions about this unusual occupation.

Despite its obscure origins, Shoop influences modern think-ing about Paleoamerican settlement patterns in the Mid-Atlan-tic. “One of the interesting things about Shoop is that it seemsthey adapted to the environment very quickly,” says Carr. Allacross the continent, in fact, we see local adaptations develop-ing rapidly in the late Pleistocene. A prime example is the

adoption of fluting technology: Re-gional knappers from coast to coastcame up with unique variations offluted points from the word go, andthat in just a few decades.

“This rapid adaptation may be whatwas happening with Shoop, too,” Carrsuggests. “They were moving southto exploit something, possibly migra-tory animals. The Younger Dryasshift probably happened in someone’s

Probing the Mysteries of the Shoop Site


Carr (left) at the Shoop sitewith Lloyd Chapman of theNational Park Service, 2008.

lifetime. How did local cultures reactto that? If we could get some better dates on Shoop, I believe itwould be very important to understanding human cultural adap-tation over the very short term.”

–Floyd Largent

How to contact the principal of this article:Kurt W. CarrSenior CuratorState Museum of Pennsylvania300 North StreetHarrisburg, PA 17120-0e-mail: [email protected]


HE MOST ANCIENT CERAMICS in the world are found inAsia, appearing in the archaeological record possibly asearly as 18,000 CALYBP. Three thousand years later

grate all possible information to increase the accuracy andprecision of the 14C date from the analytical and from thearchaeological context point of view.” The result is a trustwor-thy date on the oldest pottery known to man.

China isn’t alone when it comes to ancient pottery. TheJapanese archipelago boasts a number of sites containingdated ancient ceramics. In fact, Yaroslav V. Kuzmin of the

Institute of Geology andMineralogy at Novo-sibirsk, Russia, believesthat Japan has the mostabundant records of theearliest pottery in EastAsia. Japan’s earliest ce-ramic type, from the In-cipient Jomon period,dates to 15,600–16,800CALYBP. The authenticityof these dates was ini-tially disputed on ac-count of “old carbon”emissions from volcaniceruptions; dendrochro-nology, however, putthese suspicions to rest.The results from datingHolocene tree rings cor-responded with the C-14dates, confirming that al-though Japan may nothave the oldest pottery in

the world, it isn’t far behind China.Dr. Kuzmin also investigated ceramics from Initial

Neolithic sites in the Russian Far East 3600 km from theBering Strait excavated in the1990s and 2000s. Here he foundevidence of ceramic technologyassociated with charcoal datesas old as 14,100–16,500 CALYBP.When ceramic artifacts were di-rectly dated instead of associ-ated charcoal, they returneddates even a few hundred yearsolder.

But hold the phone. Directlydated? Yes, such experimentsare in the works. Kuzmin ex-plains that direct dating of pot-

tery is necessary due tothe poor stratigraphy atthe locality. So he andhis colleagues have at-tempted to date potterytempered with organics(organic material mixedwith the clay to fire thepottery) by using theaccelerator mass spec-

Vessel masterfully reconstructed from ceramic fragments found inYuchanyan Cave before excavation.

ELIS

ABE

TTA

BO

ARE

TTO

The Ceramic Gap

Tdescendants of Asian people appear in the Americas, yetceramic artifacts don’t grace the American archaeologicalrecord for nearly another nine millennia.

What caused this ceramic gap? Was pottery positioned atthe wrong time or place tomigrate to the Americas?Could the technology havemade the trip, but latersomehow got lost? Or wasthis valuable technologysimply left behind? And ifso, why? All these subplotsare possible, some moreprobable than others, butthey all have one thing incommon. They all havetheir origin in Asia.

What happens in Asia,stays in AsiaThe subject of ceramic his-tory is so broad and deepthat to explore it fullywould take lifetimes. Ex-amples from its technologi-cal summit stir images ofgoliath Greek amphorae,Basketmaker pottery, per-haps even Venus of Willendorf–like statuettes. Our interest,though, is in the very first ceramics, which appeared on thecontinent of Asia. The world’s most ancient pottery, dating to16,000–18,000 CALYBP, wasfound in Yuchanyan Cave inthe Yangzi River basin insouthern China by a team ofarchaeologists that includedElisabetta Boaretto of Bar-Ilan University in Israel andother scientists.

Dating ancient artifactscan be tricky, and ceramicremains pose additional

Excavating Yuchanyan Cave;inset, location of site in China.

ELIS

ABE

TTA

BO

ARE

TTO

complications. Dr. Boarettoand collaborators were consequently painstakingly thoroughin their methods. They dated both bone and charcoalsamples, in strata where pottery was discovered and also inlayers above and below the pottery to obtain a high level ofprecision. Sediment mineralogy was checked for any distur-bances that might introduce errors into the dates. Sampleswere prescreened in the field and again in the lab to eliminatesuspicious specimens. “In short,” Boaretto sums up, “we inte-

✪Yuchanyan

Cave

km0 1000

October ■ 2010 12

0 3cm

0

5

cm

trometer (AMS) method. The process involves dividing thesherds into interior and exterior parts, then combustingthem. Using different temperatures and oxidants for combus-tion, they obtained resultsin the neighborhood ofthose obtained datingcharcoal in close relationto ceramic sherds.

So can pottery be di-rectly dated? SuzanneEckert of Texas A&M Uni-versity gives her affirmingnod that dating this way isin the realm of possibility.She should know; she’s ex-perimenting with a proce-dure to radiocarbon datesoot and residues on ce-ramics herself. Yet shewarns that you “cannotsimply grab any particularsherd and date it.” She cau-tions that the potentialproblem with Kuzmin’stechnique is that, eventhough he is attempting to date the plant fibers added byhumans at the time of manufacture, these plant fibers can becontaminated by carbon that occurs naturally in the clay.Although not a specialist in Russian Far East ceramics, Dr.Eckert suspects that these vessels were not fired to a tempera-ture high enough to burn off naturally occurring organics.These organics would skew the radiocarbon date toward theage of the clay, not of the manufactured pottery. It wouldn’t

make the sherd look . . . well, as old as dirt, but somewherebetween the age of the sediment and the age of the ceramic.

In their paper “Direct Radiocarbon Accelerator Mass Spec-trometric Dating of the Earliest Pottery from the Russian FarEast and Transbaikal,” O’Malley, Kuzmin, and colleagues

discuss the difficulties of separating the plant fibers from thenaturally occurring organics. Sometimes it isn’t possible.Nonetheless, they believe their technique accomplishes just

such a separation, so they date only theplant fibers. This new method of datingpottery isn’t universally accepted amongarchaeologists concerned with directlyradiocarbon dating pottery, but some au-thorities suggest it holds interesting pos-sibilities for future research. It’s encour-aging that charcoal samples fromprimary contexts at the Gromatukha sitein the Russian Far East appear to con-firm 14C dates of pottery from the samecontents.

New World, no pottery?Despite evidence that ceramics tech-

Boaretto with the children of the familythat housed her during the excavationof Yuchanyan Cave.

ELIS

ABE

TTA

BO

ARE

TTO

nology was in use in Asia well before thefirst migrants crossed over to the

Americas, no whisper of pottery pops up in the Americanarchaeological record until 5,500 years ago. And as Dr. TomDillehay of Vanderbilt University points out, that was inSouth America. It took North America another 2,000 yearsafter that to climb on the band wagon; the earliest evidenceof pottery on this continent is a late-Archaic site on StallingsIsland in Georgia, an island in the Savannah River northof Augusta, that dates to 3500–3800 RCYBP (3700–4200

CALYBP). That’s a gap that spansmillennia. Perhaps “gap” isn’tthe right word—canyon, gulf,or chasm might be a betterchoice.

The point is, pottery showsup late in the game with NewWorld players. Is it possiblethat ceramics technologymade the migration and onlylater got lost somewhere alongthe way? Such occurrences areunusual, Eckert says, thoughnot entirely unheard of. In fact,

she believes exactly such a case happened in Polynesia. Sothe question now becomes, Could evidence of early NewWorld pottery, say in Alaska, have gone unnoticed in thearchaeological record? Dr. Steven Simms from Utah StateUniversity and Dr. Andrew Ugan of the Smithsonian Tropical

Pottery of the Osipovkaculture (7800–13,300 RCYBP)from the Gasya site in theRussian Far East.

AFT

ER D

EREV

IAN

KI

& M

EDV

EDEV

199

5


Research Institutedon’t think this is thecase. They, along withJason Bright of the Uni-versity of Utah, haveput a great deal of ef-fort into discoveringthe relationship be-tween mobility andmanufacturing ceram-ics in the Great Basin.If ceramic implementswere in use, surelysome trace of themwould have been dis-

the trip, then neither did their knowledge. Throughout his-tory it’s the case that individuals possessing unique hard-learned skills are absent in the first wave of migrants to arrivein a new land. Some artisans are left behind, some come alonglater. Except in this case, when the sea level rose and the LandBridge was submerged, the option to follow was lost.

Taking a step back and looking from a different view at aNew World bereft of pottery, is it possible that ceramic tech-nology was purposely left behind? Perhaps these originalhunter-gatherer travelers didn’t find it practical to tote theirceramics. After all, a clay pot is neither the lightest nor the

easiest item tocarry. Why dragalong a bulky potthat will mostlikely break any-way? Mobility andpottery aren’t thehappiest bedfel-lows.

This is whereUgan and Simmsstep in. They pointout that, contraryto what we mightthink, many mo-

bile groups did make use of ceramic technology. By examin-ing how much time and care a potter invests in creating avessel, Ugan and Simms tell us, we can predict what kind ofhandling the item is expected to endure. Pottery made bysomeone on the go isn’t made with the detail and attention

invested in a vessel that isexpected to last years. Afterall, who takes fine china ona picnic? To assess thepotter’s intent, Ugan andSimms consider such fac-tors as wall thickness, tem-per, size, and surface finish,but they caution againstgeneralizing. Some factors,wall thickness, for example,may swing either way. Thepotter in a sedentary societymay take care to thin thewalls of a pot because he ex-pects the pot to last a longtime and considers it worththe effort; on the otherhand, the potter in a highlymobile group may do thesame thing to make the potlighter and therefore easierto carry.

Ugan and Simms’s experi-ments with this technique inthe Great Basin on Fremont

Examples of the Stallings Potterytradition from sites in the Savan-

nah River valley. This is the earliestdated pottery in North America. K

ENN

ETH

E.

SASS

AM

AN

STEV

EN R

. SI

MM

S

TOM

D.

DIL

LEH

AY

Above, Simms in Argentina, 2010;right, Dillehay in the lab.

continued on page 20

covered by now in early sites. It’s appearing morelikely that although the technology existed in Asia,the New World simply missed the ceramics bus.

That won’t fit in my carry-onIt seems odd that early migrants would intentionallyleave behind a valuable technology. We know thatpottery was being manufactured in the Russian Far East at theright time to be carried with other camp goods by the firstmigrants. What we don’t know, however, is how widely thetechnology was spread among the population. Did all theseancient Siberian groups possess the knowledge of how toshape and fire pottery, or was it askill known only to a select few?Eckert points out that althoughsome ancient Siberians clearlyhad ceramic technology, it’s pos-sible that migrants to NorthAmerica came from groups thatlacked it. After all, the technologyto carry limitless music and bookson pocket-size devices exists to-day, but not everyone owns aniPod or a Blackberry. We can’t ex-pect that everyone of those an-cient groups was up to date on thelatest fad.

Eckert gives another spin onthe same theory. Perhaps, shesuggests, ceramics manufactur-ing was a responsibility assignedto specific members of the group,such as certain women or the eld-erly. If these people didn’t make

October ■ 2010 14

Aznick, MT

Blackwater Draw, NM

Colby, WY

Dent, CO

Domebo, OK

East Wenatchee, WA

Gault, TX

Jake Bluff, OK

Lehner, AZMurray Springs, AZ

Aubrey, TXTopper, SC

Sloth Hole, FL

Paleo Crossing, OH

Shawnee-Minisink, PA

Lange-Ferguson, SD

HE WEALTH OF KNOWLEDGE we have amassed onthe Clovis culture would stun any archaeologist 10years ago and utterly dumfound a top-notch authority

information comes to light. CSFA was founded on this prin-ciple and today remains determined to keep an open mindabout how the Americas were settled. Who got here first?Where did they come from? When did they arrive? Thesequestions drive the Clovis-First vs. pre-Clovis debate, whichchallenges a cornerstone of North American anthropology.

The case for pre-Clovis today rests on shaky evidence—it’slean, varies widely from site to site, and either hasn’t beenformally published or was published so recently that the ar-chaeological community hasn’t had time to assess it and passjudgment. Any newly discovered site that possesses acceptable

bona fides, meaning good stratigraphic con-text and reliable geochronological dates,is certain to be recruited by one side orthe other in this contest of opposing hy-

potheses. Paradoxically, the case for pre-Clovis has to be pleaded in terms of Clovis

evidence—in this contest the Clovis-Firstteam has a clear home-field advantage.

Part I: Simply ClovisPart I: Simply Clovis

Clovis sites. All these sites either wererecently redated by Mike Waters and

Tom Stafford or have diagnostic Clovisartifacts from secure stratigraphic contexts.

They all date within the remarkably narrowinterval of 10,765–11,125 RCYBP.

One obvious fact trumps all arguments in thisdebate: We can’t agree on what isn’t Clovis unlesswe agree on what is Clovis.

50 years ago. Data from recently discovered sites, reanalysis ofmaterials from known sites, and even a possible extraterrestrialevent (a comet now demands consideration) have shaken upthe scientific community and stirred archaeologists to discussin a big way what is Clovis and what isn’t.

But this is what scientific inquiry is all about, a willingness toquestion existing beliefs and explore fresh possibilities as new


Keeping up with an ever-expanding databaseNew scientific techniques and solid scholarship have so en-larged our knowledge of Clovis technology and lifeways thatthe conventional wisdom of 40 years ago is today permanentlyconsigned to the rubbish heap of history.

For instance, identifying the iconic fluted projectile pointwith extinct Pleistocene megafauna once dominated our per-ception of Clovis hunter-gatherer subsistence; our fixation onClovis the mammoth hunter overshadowed the “gatherer”aspects of Clovis diet, which we now know included smallgame and plant resources (MT 25-4, “The PaleoindianMenu: Subsistence and Diet”). Or consider how evidencecollected in the last decade from the south-central and south-eastern United States demonstrates that blades and bladecores aren’t an “occasional” occurrence at Clovis sites; bladeproduction was a principal component of the Clovis techno-logical repertoire, particularly where abundant, high-qualitytoolstone was easily available (MT 20-1, -2, “Assault onGault”). And whereas we once considered Clovis caches

Clovis bifaces from the Gault site, Texas. Note the largeflake scars that travel across the midline of the biface

and the removal of endthinning, or fluting, flakes. Clovisknappers fluted early and often! The preform at lower

right was nearly finished when it broke.

merely static finds, which we studied because they helpedus understand Clovis tool production, we now comprehendthe behavioral aspect of caching; it was a strategy used byClovis folks who were familiar enough with their surround-ings to risk storing implements on the landscape for futureuse—a real-world solution to known environmental condi-tions, not just a collection of “lost” tools (MT 22-2, “Snap-shots in Time: New Insights from Clovis Lithic Caches”).

The point is that our understanding of Clovis is fluid,continually changing as new evidence comes to light. Keep-ing in mind past perceptions (and misconceptions) and howour knowledge has progressed, let’s move forward on theClovis-First vs. pre-Clovis debate by simply asking, What isClovis?

The organization of Clovis technologyThe first question to ask is, Why is Clovis technology impor-tant? How can material remains contribute to our under-standing of who first settled the Americas?

Archaeologists study the technological organization of aculture to learn how its members adapt available resourcesfor use in their strategies for survival. These strategies haveeconomic dimensions (for example, obtaining and storingfood, acquiring materials for clothing, shelter, and tools)and social dimensions (division of labor, planning migra-tions and hunts, establishing and maintaining authority,disposing of the deceased). The environment sets the stage;

Clovis blades come in a variety of shapes and sizes.

it presents a complex of conditions and defines the problemsthat must be solved by creating and applying technological

FRA

NK

WIE

RSC

OTT

MIN

CH

AK

October ■ 2010 16

strategies. The business of the artifact analyst is to identifytechnologies, understand how these technologies were em-ployed in social and economic strategies, and theorize howtechnological change relates to long-term cultural change.

In the case of Clovis, the artifact analyst’s principal interestis lithic technology because of the near-total absence of per-ishable artifacts, except for a few of bone, after 13,000 years.Clovis lithic technology is the manifestation of a strategy thatorchestrated the tasks of acquiring toolstone, manufacturingtools, maintaining them, and discarding them when they wereno longer usable.

If Clovis tools and the technology used to create them canbe precisely defined, then archaeologists have a standard forcomparing artifacts from potential pre-Clovis sites. Coupledwith radiocarbon dates for all known Clovis sites in NorthAmerica (MT 22-3, -4, “Clovis Dethroned: A New Perspec-tive on the First Americans”) and genetic data (Goebel et al. in“Suggested Readings”), these multiple lines of evidence giveus the clearest template of what constitutes Clovis technologyand paint the most complete picture of what life was likeduring the Clovis era.

This article is the first installment of a series that reviews themost recent evidence of Clovis technology—think stone andbone tools! Our focus will be on sites with diagnostic Clovis

Clovis blades and flakes were made into toolsdesigned to get the job done—cutting, scraping, orboring holes. Use-wear research shows that the toolspictured here were used on a variety of hard and softmaterials, like wood, bone, hide, and meat.

artifacts, backed up with solid stratigraphic context and cred-ible radiocarbon or OSL dates. Of course, compelling evidencefrom sites of less certain provenance also begs to be addressed.Each installment of this series will address a different class ofartifacts of the Clovis toolkit: points and bifaces, blades andtools made from blades, and bone tools. We’ll review what weknow about each component and how recent discoveries haveexpanded or altered the sum of our knowledge.

A brief review . . .If you’re a regular reader of Mammoth Trumpet, then youknow that Clovis is identified by a distinctive type of flutedprojectile point that is found widely distributed across NorthAmerica. This projectile point, and the cultural phenomenonassociated with its makers, was named after the type sitediscovered in the late 1920s near Clovis, New Mexico. ButClovis has come to mean so much more. It is synonymouswith a lifestyle practiced by mobile hunter-gatherer peoplewho burst upon this continent at the end of the last Ice Ageabout 13,000 years ago during a time of rapid environmentalchange, and after no more than 400 years surrendered stew-

Clovis blade cores. These typical examplesare from south central Texas.

ardship of the land to new cultures. Clovis is typified (andparodied) in the movies and on television in scenes depictingFR

AN

K W

IER

CH

ARL

OTT

E PE

VN

Y


scraggly, fur-wearing hunters with spears tipped with stonepoints stalking the obligatory woolly mammoth or saber-toothed tiger.

Archaeologists use the term Clovis to refer to substantialitems such as tools, diet, even the people themselves. Not-withstanding continent-wide similarities in certain aspectsof Clovis life, we can’t say with certainty that these peopleshared precisely the same mores, that is, the same religiousbeliefs, morals, and mortuary practices, the more ephem-eral aspects of human life that are nearly impossible toidentify archaeologically. University of Nevada anthropolo-gist Gary Haynes in The Early Settlement of North America:The Clovis Era, a comprehensive review of all things Clovis(see “Suggested Readings”), uses “Clovis era” to character-ize the traits of fluted-point assemblages that date to aninterval around 11,000 RCYBP. CSFA Director Mike Watersand geochronologist Tom Stafford discovered that this inter-val is very small indeed, spanning as little as 250 years.

Haynes cites no fewer than 14 traits that he considerstypical of Clovis artifact assemblages, including:■ the presence of bifacial, fluted projectile points often

made from nonlocal stone■ bifaces used as multipurpose tools and preforms■ blades and blade cores■ cutting and scraping tools made on flakes or blades■ multifunctional tools such as scrapers with graver spurs■ the occasional presence of bone and ivory tools■ the occasional presence of hearths■ infrequent presence of plant foods and small animals■ tools made of high-quality stone, which may originate

from sources over 250 km away■ sites associated with water sources or water near areas

where stone is available to make tools.

Clovis points. Though these iconic projectile points varyin size and the type of toolstone they were made from,

they all have the characteristic lanceolate shape, distinc-tive fluting flakes removed from the basal edges, and

grinding on the lower edges and bases. The large hori-zontal point at the bottom is a bit longer than 20 cm.

North American continent and in practically every conceivableenvironmental setting, not every tool type is present at every site.To date, archaeologists haven’t recovered the complete Clovistoolkit from any single site. This is because different types of sitesrepresent different activities or groups of activities. For example,what is often referred to as a Clovis “kill” site is a location wheregame, usually the remains of mammoth or less often bison, wasacquired and processed. What might that look likearchaeologically? Artifacts would likely include fairly specializedtools like fluted projectile points, as well as a few tools used to skinand butcher the animals. There may also be small pieces of stonedebris (debitage) detached when tools were resharpened. Andwith a little luck there may be some portion of the skeleton left.

Kill sites are most often found in the Southwest and Plains

regions of the U.S. In the early years of research, what we learnedabout the Clovis era was based in large part on evidence recoveredfrom these sites (hence the monolithic Big-Game Hunting Theorythat dominated our thinking for so many years). Clovis kills in-clude such well-known sites as Blackwater Draw Locality No. 1 inNew Mexico, Colby in Wyoming, Dent in Colorado, Domebo inOklahoma, Lehner in Arizona, and Murray Springs in Arizona.

Making toolsWhere did Clovis hunter-gatherers obtain the stone to make theirtools? At quarries, or toolstone procurement sites, which mayhave been located some distance from other resources such asfood and water. The overwhelming majority of the artifact assem-

Notice how many things on that list have to do with rocks?Obviously, it’s because rock is so enduring. Common sensetells us that Clovis folks wore clothing, probably made ropeand cordage from plant fibers, and most likely fashionedbags and other containers of hide or leather. Unfortunately,with just a few exceptions these perishable organic materi-als haven’t stood the test of time. Osseous (bone) tools havebeen recovered from a mere handful of sites like MurraySprings in Arizona, Blackwater Draw in New Mexico, andthe Anzick cache in Montana.

How did Clovis people use the landscape?Archaeologists use artifacts and their association with eachother to understand the purpose sites served. Despite thepervasive presence of Clovis projectile points across the

CH

ARL

OTT

E PE

VN

Y

October ■ 2010 18

Suggested ReadingsGoebel, T., M. R. Waters, and D. H. O’Rourke 2008 The Late

Pleistocene Dispersal of Modern Humans in the Americas. Sci-ence 319: 1497–1502.

Haynes, G. 2002 The Early Settlement of North America: TheClovis Era. Cambridge University Press.

Mallouf, R. J. 1989 A Clovis Quarry Workshop in the CallahanDivide: The Yellow Hawk Site, Taylor County, Texas. Plains An-thropologist 34:81–103.

blage from a quarry consists of debris remaining when stonewas reduced. Tested nodules or cobbles, expended cores,manufacturing rejects such as broken bifaces, and knappingtools including hammerstones are the kinds of artifacts recov-ered from quarries. Diagnostic artifacts like finished Clovispoints usually aren’t found at quarries. Clovis quarry sites arefew in number. The Yellow Hawk site in Texas is a potential

candidate, though contextually there are problems with therecovered assemblage.

Life and work at a campsiteCampsites were seasonally occupied locations where inhabit-ants remained for days or perhaps weeks. A campsite usuallydisplays a diversified artifact assemblage and may retain semi-permanent evidence of occupation, perhaps hearths or evendwelling structures, because folks resided in the area for aconsiderable spell or regularly returned to the same location.Campsites were workshops where food and hides were pro-cessed and all stages of tool manufacturing were performed,including knapping tools from blanks or cobbles, heat treating,recycling and resharpening, and producing expedient tools(general-purpose throwaway implements). We can find almostany stone tool in every stage of manufacture at a campsite—expedient tools, scrapers, drills, burins and gravers, repaired orreworked tools, and finished projectile points. Clovis campsiteslike Aubrey in Texas and Shawnee-Minisink in Pennsylvaniaalso have yielded the remains of foods drawn from a variety ofanimals and plants, proof that Clovis folks had a more well-rounded diet than we earlier supposed.

The types of site differ according to the particular activitiesthe Clovis inhabitants were engaged in. A campsite associatedwith procuring game (kill-camp) is structured differently than acamp site associated with a quarry (quarry-camp). Clovis kill-camps are located adjacent to kill sites at Blackwater Draw inNew Mexico and Murray Springs in Arizona. These campsitesappear to have been occupied briefly during and after kills; theartifact assemblages comprise tools used to process prey andsome evidence of tool manufacture and maintenance. In the

case of a large-scale Clovis camp, its inhabitants may have beenengaged in several industries. The Gault site in central Texas,for example, a quarry-camp that was occupied over a consider-able time or many times, shows evidence of a wider range ofactivities than simply quarrying. The Topper site in SouthCarolina, which boasts an easily accessible outcropping ofAllendale chert, a fine toolstone, is a textbook example of a

Clovis quarry-related site.Clovis caches are a unique kind of site consist-

ing of closely clustered, apparently “stored,”implements or materials. Tools in caches aren’t

Tankersley, K. B. 2004 The Concept of Clovis and the Peopling ofthe New World. In The Settlement of the American Continents: AMultidisciplinary Approach to Human Biogeography, edited by C.M. Barton, G. A. Clark, D. R. Yesner, and G. A. Pearson, pp. 49–63.The University of Arizona Press, Tucson.

Waters, M. R., and T. W. Stafford, Jr. 2007 Redefining the Age ofClovis: Implications for the Peopling of the Americas. Science315:1122–26.

About the author Charlotte Donald Pevny received herPh.D. from Texas A&M University in 2009. She was a post-doctoral Fellow at the Center forthe Study of the First Americansfrom 2008 to 2009. Dr. Pevny’s dis-sertation focused on Clovis lithictechnology and tool use at theGault site in central Texas. Shealso has worked at the nearbyFriedkin site, which has Clovisand pre-Clovis components. Herresearch interests include usewear and experimental archaeol-ogy. She is particularly interestedin site-formation processes andunderstanding the effects of post-depositional processes on lithic assemblages from early sites.Pevny currently works in the field of cultural resources man-agement as a project manager for R. Christopher Goodwin &Associates, Inc., in New Orleans, Louisiana.

Clovis beveled bone rod withincised lines. Finds of Clovis toolsmade from bone or other organicmaterials are few and far between.

worn out like the broken and discarded tools found in kill andcamp sites, but show little use and sometimes are even inpristine condition. Caches were created for either utilitarian orceremonial motives. Clovis caches discovered across the U.S.vary widely in their settings and contents.

Archaeology is nothing without context. Now we’ve set thestage for our next installment, where we’ll look in detail at aspecific component of the Clovis toolkit and tell why it’s impor-tant to the Clovis-First vs. pre-Clovis debate.

–Charlotte Pevny

CH

ARL

OTT

E PE

VN

Y


and later cultures show promise. They are confident that thetechnique can be applied elsewhere, with the caveat thatdifferent ceramics made by different cultures require specialattention. Could it be used in the Russian Far East to shedlight on the mystery of why ceramics stopped short of cross-ing continents all those years ago? Maybe. Ugan suspectsthat the archaeological record in Siberia may be too meagerfor detailed investigation. As an alternative he suggests look-ing for clues at other cultures during periods of ceramicdevelopment, such as the New World during the late Ho-locene.

“Pottery,” says Ugan, puzzling over the ceramic gap, “isone solution to a problem of transporting, cooking, and stor-ing.” It’s only one solution of many, some more costly thanceramics, others less. Boiling stones, storage pits, and bas-ketry, he points out, are alternatives that cultures around theworld have used as effective substitutes for pottery. Nonethe-less we’re left with the question, Why would Asian peoplediscard ceramics for something else before hitting the Ameri-cas? Perhaps, Ugan proposes, there was some sort of shift inthis people’s way of living their highly mobile lives. When thisoccurred, “pottery was not the right tool for that kind oflifestyle.”

One last possible theory . . .There’s another possible explanation to account for the mys-tery of why ceramics technology, developed in the RussianFar East early enough to board the Bering Strait Express,hasn’t showed up in the oldest New World sites dating to15,000 calendar years ago. But this assumes that those sites,Clovis sites, are the oldest sites in North America. Couldthere be older sites that haven’t yet been discovered? Arethere as yet undiscovered sites with evidence of pottery mak-ing that date to the time of its infancy in the Russian Far East?For the moment it’s a question that can’t be answered, but it’sanother reason to keep our eyes peeled for what turns up inthe pre-Clovis bag of surprises.

–K. Hill

How to contact the principals of this article:Suzanne EckertAssistant ProfessorDepartment of AnthropologyTexas A&M Universitye-mail: [email protected] D. DillehayRebecca Webb Wilson University Distinguished Professor

of Anthropology, Religion, and Culture and Professor ofAnthropology

Department of AnthropologyVanderbilt UniversityNashville, TN 37325e-mail: [email protected] BoarettoThe Martin (Szusz) Department of Land of Israel and

Studies of ArchaeologyBar-Ilan UniversityTel Aviv, Israele-mail: [email protected] R. SimmsProfessor of AnthropologyUtah State UniversityLogan, UT 84322-0730e-mail: [email protected] V. KuzminInstitute of Geology and MineralogySiberian Branch of the Russian Academy of SciencesKoptyug Ave. 3Novosibirsk 630090, Russiae-mail: [email protected] UganSmithsonian Tropical Research Institutee-mail: [email protected] E. SassamanHyatt and Cici Brown Professor of Florida ArchaeologyDepartment of Anthropology1112 Turlington HallP.O. Box 117305University of FloridaGainesville, FL 32611-7305e-mail: [email protected]

Suggested ReadingsBoaretto, E., X. Wu, J. Yuan, O. Bar-Yosef, V. Chu, Y. Pan, K. Liu,

David Cohen, T. Jiaoh, S. Li, H. Gu, P. Goldberg, and S. Weiner2009 Radiocarbon Dating of Charcoal and Bone CollagenAssociated with Early Pottery at Yuchanyan Cave, Hunan Prov-ince, China. Proceedings of the National Academy of Sciences106(24):9595–9600.

Bright, J., S. R. Simms, and A. Ugan 2005 Ceramics from CamelsBack Cave and Mobility in Farmer-Forager Systems in the EasternGreat Basin. In The Archaeology of Camels Back Cave, by D.Schmitt and D. Madsen. University of Utah Anthropological Pa-pers 125. University of Utah Press, Salt Lake City.

Kuzmin, Y. V., 2006. Chronology of the Earliest Pottery in East Asia:Progress and Pitfalls. Antiquity 80(308):362–71.

Nesterov, S. P., M. Sakamoto, M. Imamura, and Y. V. Kuzmin 2006The Late-Glacial Neolithic Complex of the Gromatukha Site, Rus-sian Far East: New Results and Interpretations. Current Researchin the Pleistocene 23:46–49. Center for the Study of the FirstAmericans, Texas A&M University, College Station.

O’Malley, J. M., Y. V. Kuzmin, G. S. Burr, D. J. Donahue, and A. J. T.Jull 1999 Direct Radiocarbon Accelerator Mass SpectrometerDating of the Earliest Carbon Dating of the Russian Far East andTransbaikal. Mémoires de la Société Préhistorique Française, TomeXXVI [Supplément 1999 de la Revue d’Archéométrie].

Simms, S. R., 1997 Plain-Ware Ceramics and Residential Miobility:A Case Study from the Great Basin. Journal of ArchaeologicalScience 24:779–92.

The Ceramic Gap


probing the mysteries shoop site probing the mysteries shoop site

Documents