Across Atlantic Ice

by Dennis J. Stanford

  The Berelekh Site was relatively rich in perishable osseous tools. The collection includes four mammoth bone knives, one leaf-shaped ivory projectile point, two scrapers, and several ribs with abrasions that may indicate use as burnishing tools. Mochanov and Fedoseeva state that these bone artifacts exhibit the same fine pressure flaking seen on the stone points. This is curious and makes us wonder whether the ivory and bone were mineralized when modified. The paleontologists N. K. Vereshechagin and V. V. Ukraintseva also think that the inhabitants of the site used bones mined from fossil deposits as a raw material source for artifact manufacture.17 Although experiments to study the tool-manufacturing flaking properties of large bones—for example, from elephants—have shown that it is possible to make large flakes on dense bones, our experience is that pressure-flaking unmineralized bone is impossible because bone and ivory rapidly lose their workability, even when well preserved. If we are right, the purported pressure flaking on this bone indicates that it was mineralized and that it is not relevant to the dating of the site.

  Berelekh is also famous for yielding a fragment of mammoth bone incised with the figure of a mammoth, which suggests that the artist who made the etching saw a live mammoth or even hunted one. It must be remembered, however, that frozen mammoths are commonly found even today melting out of the Siberian permafrost, and the artist may have used a long-frozen specimen as a model. We also know that mammoths were around during Clovis times and that they survived on Wrangle Island, in the Arctic Ocean northwest of the Bering Strait, until long after Clovis times, so this engraving does not necessarily indicate the artist saw a mammoth at a Clovis date.18

  FIGURE 3.4.

  Akmak artifacts from Ushki Levels VI and VII and Bolshoi Elgakhchan: (a–c) Ushki Level VI projectile points; (d–f) Level VI microblade cores; (g–h) Level VI burins; (i) Level VI end scraper; (j–l) Bolshoi Elgakhchan projectile points; (m–p) Ushki Level VII projectile points; (q) Level VII microblade core; (r) Level VII foliate biface; (s) Level VII burin; (t) Level VII uniface knife. Arrows show direction of spall removal.

  It is difficult to assess what all of these bone tools represent culturally, since most of them were found during paleontological salvage operations carried out mainly by high-pressure water excavation, which thoroughly mixes everything. Although the fossil bone deposit is dated to around 13,000 years BP, the human occupation of the site may have been much later. Vereshchagin and Ukraintseva suggest that the occupation evidence was recovered from a depth of 1.5 meters, rather than 2.5 meters as originally reported, and that the human occupation of the site probably occurred around 10,000 years ago.19

  Stone pendants and beads are present in some western Beringian Late Dyuktai sites. Eight hundred stone beads were found in Ushki Level VII in association with a human burial (the bone was all but gone), and three stone pendants came from the same level. The younger Ushki Level VI produced several stone pendants and a flat pebble with scratches that, to those with good imaginations, could depict tents. Five stone pendants were found at Berelekh, but their provenances are also questionable. If they were associated with the cultural occupation level, they would be presumed to date to around 10,000 years BP. Two of the pendants have short incised lines or notches along their edges, and all have conical holes, drilled from opposite faces, that meet in the middle, producing an hourglass shape.

  It appears that the cultural sequence of western Beringia is relatively straightforward. The Late Paleolithic flaked stone technology is dominated by an inset microblade tradition that is totally absent from Clovis. The few examples of bifacial projectile point technology are either insecurely dated (Berelekh) or have proved to be younger than Clovis (Ushki VII). Although there are local variations within assemblages and technologies, we contend that these variations are minor and are to be expected in such a large region as technologies evolved through ten millennia.

  Early Dyuktai technology first appeared in far western Siberia during the height of the LGM. Why this happened is difficult to understand, considering the harshness of the climate, but clearly the people who used it did not move farther eastward across the Kolyma Basin until the end of the glacial period, when environmental conditions made it more feasible to settle the area. It is also possible that they were pushed eastward by early Neolithic peoples who were moving north from central Asia during a warming phase. In any case, there is little doubt that the Late Dyuktai flaking techniques evolved from the microblade techniques developed by earlier Dyuktai flintknappers, as they represent very similar traits. The current archaeological evidence indicates that Late Dyuktai people arrived on the shores of the Bering Sea more than 12,000 years ago. When they finally crossed into the region now known as Alaska, did they find an uninhabited territory, or did they encounter people already living in the area—or even moving westward out of eastern Beringia?

  EASTERN BERINGIA

  The archaeology of eastern Beringia is also difficult to assess. As in western Beringia, the number of well-dated and stratified sites is meager, but at least two technological patterns are beginning to emerge. The majority of early Alaskan sites, like those in Siberia, seem to have been occupied by people who used microblade technology at about the same time as or shortly after it appeared in northeastern Asia. We suggest this indicates that the microblade-based industry of Siberia spread eastward into Alaska near the end of the Late Dyuktai period. The regional and temporal spread of this microblade technology—and its offshoots—tells us that the people with this technology were well suited to the environment and flourished, developing into distinct groups. However, at about the time when this highly dynamic pan-Siberian technology was becoming established in eastern Beringia, people using non-microblade weaponry were also making an appearance in the area. Where did they come from?

  The settlement of Alaska (eastern Beringia) was probably a slow, complex process stimulated by climatic improvement, as the Arctic archaeologist John Hoffecker and his associates suggest.20 Milder weather conditions and nutrients released from the melting glaciers enhanced the productivity of rangelands. Increased subsistence resources no doubt affected the human fertility rate, causing the population expansion reflected in a major increase in the number of sites and the diversity of technologies at this time. Groups with different lithic traditions—and therefore unique signatures—moved into Beringia from different directions.

  The Beringia (sometimes called the American Paleo-Arctic) tradition is, in our opinion, an eastward extension of Late Dyuktai. One of the earliest assemblages of this tradition comes from the Onion Portage Site on the Kobuk River, just upstream from Kotzebue.21 This is one of the first deeply stratified sites found in Alaska, and it has provided baseline data for the cultural sequence of northwest Alaska. At the bottom of the site was an unconformity, or buried eroded surface, upon which more than a hundred artifacts were found. The group of artifacts has been named the Akmak Complex. (Complex is basically synonymous with the term archaeological culture: it describes a group of sites that share specific traits, especially those related to tool forms.) This accumulation of cultural debris is thought to have been redeposited from a cache of precores (figure 3.5a–b), wedgeshaped microblade cores (figure 3.5d), and formal tools such as burinated blades, retouched blades (figure 3.5e), blades retouched into scrapers (figure 3.5f), and bifacially edge-re-touched knives (figure 3.5c). A radiocarbon assay of a caribou bone associated with the artifacts indicates that the materials were deposited more than 9,500 years ago. Hence the Akmak tools are older, by an unknown time period. This assemblage is very similar to Late Dyuktai.

  Another northwestern interior Alaska site, known as Nogahabara 1, has a toolkit of 267 artifacts that appear to have been lost or abandoned between 12,740 and 13,850 years ago. These artifacts were exposed by the wind deflation of a sand dune. The fact that all of the artifacts were clustered and, with the exception of a single bifacial chert drill, made from the distant Alaskan Batza Tena obsidian source suggests that this is a single asse
mblage rather than a lag accumulation resulting from, for example, wind deflation of the dune. The assemblage consists of used flakes and flake blanks (figure 3.5g), pressure blades and pressure blade cores (figure 3.5h), notched and indented base lanceolate (leaf-shaped) projectile points or knives in various stages of manufacture (figure 3.5j–l), and unifacial scrapers. Of interest to our discussion is the unambiguous co-association of pressure blade technology with indented base thinned lanceolate bifaces and notched points. While it has been suggested that the notched and the indented base points are two distinct end products of the same manufacturing system, the lack of edge grinding on the concave base bifaces makes it likely that these specimens are late-stage preforms for notched points or reworked pieces, which are also basally thinned.22

  FIGURE 3.5.

  Akmak artifacts from Onion Portage (a–f) and Nogahabara 1 artifacts (g–l): (a–b) precores; (c) bifacially edge-retouched knife; (d) wedge-shaped microblade core; (e) retouched blade; (f) blade retouched into scraper; (g) retouched flake; (h) pressure microblade core; (i) projectile point preform; (j–k) lanceolate projectile points (possible preforms for notched points); (l) notched point.

  FIGURE 3.6.

  Pre-Nenana (a–c) and Denali (d–i) artifacts: (a–b) microblades from Swan Point; (c) burin and burin spall from Swan Point; (d) microblade core; (e–f) bifacial points and (g) burin from Broken Mammoth; (h–i) bifacially retouched point.

  The lowest level of the Swan Point Site near Delta, Alaska, provided a date of around 14,000 years ago, making it the earliest directly dated microblade occupation in Alaska.23 It contained microblades (figure 3.6a–b), a core tablet (figure 3.6c), and detritus from their manufacture, burins, cobble tools, and hammerstones but no bifaces or bifacial points. This occupation is unique in central Alaska since it is below Nenana Complex levels (see below). This strongly suggests that microblade technologies arrived in this part of Beringia before Nenana, a presumably non-microblade complex. This is important because it may indicate that the first people in the area were most likely related to the Siberian peoples—whereas if Nenana was first it could have led to the non-microblade technologies seen in Clovis.

  The Denali Complex, according to F. H. West, represents the first colonization of eastern Beringia, at about 12,000 years BP, and remained little changed until its disappearance around 8,000 years BP or later.24 There is little difference between Late Dyuktai and Denali artifacts, and there is general acceptance that they were closely related. Denali sites, which are widespread in Alaska, are dominated by microblade inset technology but also contain a significant component of bifacial knives and points. Microblades were made primarily on wedge-shaped bifacial cores (figure 3.6d), and bifacial points were intentionally thickened and finished with pressure flaking and lower edge grinding (figure 3.6e–f and h). There are also bifacially retouched flakes (3.6i). Pressure flaking techniques range from selective and nonpatterned to serial and diagonal. Burins are common and take a variety of forms (figure 3.6g).

  Trail Creek Cave 2 and Lime Hills Cave 1 are Denali sites. Both produced slotted antler points (figure 3.7a–f), which have bases that are beveled for hafting to a spear shaft. It is clear from these specimens that inset projectile point technology accompanied microblade technology. The remarkably preserved perishable tools from Lime Hills Cave 1 included the base of an antler point or knife and most of a slotted antler projectile point (figure 3.7f).25 The slots run along opposite sides and were designed to hold microblade segments, some of which were found in association. The artifacts from these sites date to about 9,500 years BP.

  Several fluted points were found during the 1940s and 1950s along the Utukok River in northern Alaska. Since all of the points were surface finds and not directly datable, archaeologists used the tried and true typological approach to cross-date them. Because they were fluted and looked somewhat like Clovis, they were considered evidence of an Alaskan antecedent to Clovis. The real clincher for the pre-Clovis case was that the Alaskan fluting technique was significantly different from that of Clovis, suggesting that it was an early developmental phase of the technology. And, after all, the points were found where the prevalent origin theory predicted they would be.

  This pre-Clovis notion seemed to be confirmed by further work on the North Slope during the 1960s and 1970s. First, Robert Humphrey, while a PhD student at the University of New Mexico, conducted a reinvestigation of the Utukok River area in the hope of finding more diagnostic artifacts in a datable context. He discovered a site that produced additional fluted point fragments, along with evidence of the manufacture of weapon tip replacements.26 A radiocarbon date of 17,300±800 years BP was obtained from a mammoth bone found eroding out of a nearby bank. Perhaps the date was too old, because it meant a gap of almost 6,000 years between it and Clovis, but nonetheless it was pre-Clovis.

  Additional fluted points were recovered a few years later at the Putu Site, on the North Slope of the Brooks Range, and a charcoal date suggested that their age might be 11,470±500 RCYBP (radiocarbon years before the present).27 This date was absolutely perfect, establishing fluted-point makers in the American Arctic at least 200 years before their arrival on the plains. Unfortunately, when the geologists Tom Hamilton and Steve Porter undertook geological studies in the area, they discovered that the location was beneath a glacier until around 13,000–13,500 years ago.28 According to environmental reconstructions, the site would have been, at the very best, part of an unattractive, barren pile of rocks until long after Clovis times. To clarify this situation, Mike Kuntz and Rick Reanier conducted new excavations, which confirmed a younger date for the fluted points.29

  FIGURE 3.7.

  Slotted antler points and ivory tools: (a–e) antler points from Trail Creek Cave 2 (cross-section outlines of b and e illustrate grooves for microblade insets); (f) antler points from Lime Hills Cave 1; (g–h) Nenana ivory tools from the Broken Mammoth Site, not slotted for microblade.

  NORTH TO ALASKA

  In 1966 I accompanied Robert L. Humphrey Jr. in a search for Clovis sites along the Utukok River, where an Alaskan fluted point had been found in 1949. We were fortunate to find a campsite where early people had lived and manufactured fluted points, surely evidence of the first Clovis peoples in northwest Alaska. We were also hoping to find charcoal that would date older than the radiocarbon assays from Clovis sites in the Lower Forty-Eight. Unfortunately, we found no charcoal at our campsite, but instead we dated a sample from a mammoth bone that was eroding out of the adjacent riverbank. The mammoth bone dated to more than 17,000 years old, likely too old to be associated with the fluted point campsite.

  I have spent many more field seasons over the past forty years in various parts of Alaska searching for evidence of the peopling of the Americas. It was not until the summer of 2005 that Bob Gal, National Park Service archaeologist for northwest Alaska, found the first fluted point site in Alaska with associated charcoal. The assay of the charcoal suggested that the descendants of Clovis people started living in Alaska no more than 12,000 years ago . . . long after mammoth hunters camped at the springs of Blackwater Draw. Dennis

  Although more fluted points have been found on the surface at several locations in the Brooks Range, none has been in a good geological context. Since these finds, geological mapping and controlled excavations have demonstrated that the Alaskan fluted point complex is younger than Clovis rather than being a Clovis ancestor. In fact, some archaeologists have argued that these points represent an intrusion of late Clovis people into the Arctic, or at least a northward diffusion of the post-Clovis mid-continental fluting technology.30 Current excavations at the Serpentine Hot Springs Site in western Alaska, where fluted points have been found in good context, have yielded radiocarbon dates of around 12,000 years ago.31 Support for the idea also comes from the ice-free corridor in Canada.

  The plains of Alberta and areas northward toward the ice-free corridor have yielded fluted points.32 Like the Alaskan specimens, most of these arti
facts are neither well fluted nor representative of classic Clovis technology. This situation can be interpreted in two ways. One is that the artifacts represent early Clovis technology evolving as people moved southward. Conversely, they might represent a post-Clovis fluting strategy that was spreading northward. It would be great if the former could be proved, but there are several arguments against a pre-Clovis antiquity of the Canadian specimens.

  Also like the Alaska specimens, the Canadian points are surface finds without good contextual data—with one exception. A single fluted weapon tip was recovered from a stratified site known as Charlie Lake Cave, on the very western edge of the ice-free corridor in British Columbia.33 A radiocarbon assay yielded a date of 12,000 years ago for the fluted point level. This date fits the new deglaciation chronology reasonably well and is consistent with the information coming out of Alaska. Furthermore, the dating of fossil remains from the ice-free corridor suggests that animals were not living in the corridor before 13,000 years ago. At present the Charlie Lake Cave radiocarbon date, along with the younger age estimates for the Alaskan specimens, supports the existence of a post-Clovis technological link between the Northern Plains and the Arctic. In addition, the technology of these Canadian points is more like that of some of the post-Clovis fluted points found in the northeastern and upper midwestern United States and eastern Canada. Perhaps all of these types represent Clovis-derived people who followed the retreating ice front northward through the opening corridor into Alaska.