Across Atlantic Ice Read online

  The second biface flaking technique used by Clovis flintknappers, full face, produced flakes that traveled most of the way across the face of the artifact and terminated smoothly at or near the opposite edge (figure 2.8a–c). This method was common but was done in conjunction with overshot flaking. It is actually the same procedure, just with “premature” terminations. Platform preparation remained about the same as for overshot flaking.

  The third thinning technique, diving, was practiced by Clovis knappers to a lesser extent than the other two thinning techniques and consisted of driving flakes to the middle of the biface, ending in hinge fractures. These scars were then met by flakes coming from the other edge (see chapter 1).

  FIGURE 2.8.

  Clovis biface thinning methods: (a–c) full-face flake scars; (d) early interval end thinning; (e) middle interval end thinning; (f) late interval fluting.

  Basal treatment of Clovis preforms was the same as the treatment at the sides. Bases were thinned by striking flakes from a bevel or isolated and ground platform. During the early and middle stages of preform reduction, basal thinning was part of the overall thinning process (figure 2.8d–f). The only real difference between side and base thinning was that the central basal-thinning flakes followed a low convexity formed by flake scars originating on the sides of the preform. Also, because the bases are narrow relative to the width of the removed flakes, there is little choice of where the flakes could be removed, so both faces were thinned from about the same location.

  Basal thinning brings us to the thorny issue of fluting. Along with biface thinning, probably the most sought-after skill in modern North American flintknappers is fluting. It can certainly be asked what the big deal is, if fluting is just thinning at the base. The challenge is that it incurs a much higher risk of breakage than lateral thinning. There are two reasons for this: the difficulty of striking the right spot, especially if the piece has an indented base, and the stress caused by hitting the end of a long, thin piece, which can result in a break across the middle (termed end shock).

  Archaeologists and collectors have focused attention on a single characteristic of Clovis points, a distinctive flute scar on the finished pieces. It is unclear why the final basal thinning flakes (called channel flakes) were removed on most Clovis points in spite of the high risk of breakage. Large flakes were frequently removed to thin the bases of bifaces throughout the production process and not just during the final stages, indicating that those removed near the end of the sequence were not technologically different than the earlier ones. Examination of the platforms of the basal thinning flakes shows that they were prepared in the same manner as platforms for other biface thinning flakes. Technologically, flutes were just like any other major thinning flake and not an extraordinary invention.

  Functionally, fluting thinned the base, facilitating the point’s hafting onto a spear or foreshaft. But our experimentation with hafting and using fluted and unfluted points shows no significant difference in effectiveness. If fluting was a superior way of preparing a point for hafting, why wasn’t it commonly used and reinvented many times? Clovis people themselves may have valued its symbolism as expressed by style rather than its function. Since fluting was a risky technique, perhaps the channel scars were retained to testify to the skill of the knapper. When finishing a fluted point it is easier to flake into the channel scars from the sides than it is to avoid intersecting them. An example of this is seen in the post-Clovis Dalton technology, whose pieces were fluted, sometimes to the tip, before the channel scars were almost completely removed by further flaking to finish the pieces.38 We envision that fluting might have become part of a pre-hunt ritual.39

  There is considerable variation in the way Clovis points were finished after fluting. Some have only minimal pressure retouch along the edges to remove major irregularities. Others show more extensive invasive pressure retouch. Still others have the entire circumference finished with abrupt bifacial retouch, leaving the large, flat percussion flake scars prominently visible on both faces. Finally, some Clovis points exhibit all-over pressure finishing that is well spaced and meets in a distinct midline. Examination of collections from throughout North America does not show these variations to be regional, and it is common to see all of them in collections from a single area. Thus, it is unclear whether the differences have any meaning beyond individual knapper preference. Additionally, most of the points we find have been reworked, and the finishing techniques may not represent original manufacture.

  Consistency does occur to the extent that even a group of points with different finishing techniques can usually be confidently identified as Clovis. Clovis fluting differs in concept and expression from the other fluted forms that are generally considered to appear later. Clovis channel flakes originated from the biface plane of the preform, and when both faces were successfully thinned the result was a base with a knife-like edge and both channels originating from about the same place.

  Finally, Clovis points’ lower edges were dulled by polishing or grinding. It is thought that this was a means of keeping them from cutting the lashings that held them to a spear shaft. This is not necessary for projectile points, however, and many later point types do not have edge grinding. But edge grinding might have been useful if a hafted piece was designed to be both a point and a knife. This was probably the case with Clovis points—but not Folsom points, which it’s difficult to argue were designed to be knives, even though they also have ground edges. For Folsom, edge grinding might have been a traditional technological trait, or perhaps served a different function.

  Along with bifaces, Clovis knappers had a well-developed blade technology, especially in eastern North America. Although we’ve seen a wide range of possible flaking approaches for bifaces, most North American archaeologists are unfamiliar with the equally diverse range of approaches in blade technologies in other parts of the world. In our experience, the manufacture of blades was just as complex and varied as biface technologies.

  Mike Collins from the University of Texas at Austin published the first comprehensive description of Clovis blade technology, but even he agrees that he has only scratched the surface.40 We have examined large numbers of Clovis blades, cores, and blade-making debitage (discarded pieces), and the following description is based on our impressions. More detailed analyses need to be done on the large collections from sites such as Gault in Texas and Carson-Conn-Short in Tennessee.41

  We see two different blade-making technologies in Clovis, but they share a common goal and may be variations of the same general method. One approach to initiating blade making was to select a piece of stone with one or more long natural ridges—a flat nodule or a tabular piece—to create a core around. The knapper would build a platform above one of these natural ridges, and the first blade removed would have a natural dorsal surface (figure 2.9a). The second approach was to produce a single bifacial edge vertical to the axis of the nodule to serve as the initial guiding ridge (figure 2.9b). The first piece removed in this process is known as a crested blade. When blade precores are bifacially flaked all the way around their circumference, the resulting cores have a back ridge that is often retained throughout the blade-making sequence. In Clovis there is little evidence of this, perhaps an important distinction when looking for an ancestral technology.

  FIGURE 2.9.

  Blade precore initiation methods: (a) using a natural ridge; (b) using a bifacially prepared ridge.

  After the removal of the first blades, there was a wide range of possible choices for continued blade production. The most common approaches used throughout the world were continuing with a single platform, detaching most blades in a single direction; and with opposed platforms, striking blades from both ends of a core. Although we occasionally see blade scars originating from opposite ends of Clovis blade cores, it is clear that most blades were struck from the primary (proximal) platform, while those originating from the opposite (distal) end were removed while fixing mistakes.

>   Clovis had two blade technologies: employing a core with a single flaking face, and removing blades from the entire circumference of the core.42 The first technique produced cores that are termed wedge-shaped (figure 2.10a). The discarded core remnants tend to be flat faced and have flat backs, with either a natural cortical surface or flake scars originating from the sides and meeting in the middle. A distinct aspect of Clovis blade making, especially from these wedge-shaped cores, is the apparent intent to make highly curved blades (figure 2.10b). This is unlike any blade-making tradition we are familiar with anywhere else at any time. In other blade technologies, heavily curved blades were the by-products of core face rejuvenation. This was necessary when the face of the core became too flat to remove another straight blade. A blade or blades had to be removed from one or more corners to make the face more convex, allowing for the removal of another straight blade.

  Another characteristic of Clovis blade making is that the knappers often kept removing blades even when the core became so greatly reduced in size that it could produce only small blades. Some people call these microblades, but we prefer to reserve that designation for the highly formalized technologies where small blades were the primary product rather than just the end of the line of a large-blade technology. Hence, we refer to the small Clovis blades as bladelets. This is a distinction made by Old World archaeologists as well. It is evident that Clovis people were intentionally making these bladelets, but their purpose has yet to be discerned.

  FIGURE 2.10.

  Clovis blade cores and blade: (a) top, front face, side, and rear face of wedge-shaped core from the Gault Site, Texas; (b) front, side, and rear view of curved blade, Gault Site; (c) top, front, and rear view of conical core, Gault Site.

  The second blade-making approach is different in that the cores have blades removed from around their entire circumference. The resulting cores, described as conical (figure 2.10c), are roughly circular in cross section and taper toward the distal end. Their sequence of manufacture is unclear. They were probably made from large nodules or cobbles rather than shaped bifacial precores. It is also curious that many of the blade scars on these cores indicate that relatively straight blades were made from them, even though very few such blades have been found in archaeological sites. Perhaps straight blades were the intended product and most of them became tools that were subsequently used and modified to the point that the original blade form was obliterated. A few end scrapers made on straight blades have been recovered, and many of the expended examples may have also been made on straight blades.43 A few examples of other tool types made on straight blades are also known.

  While conical cores are relatively uncommon, most are large and do not look as though they have yielded many blades. They could be blade precores that were made for export from the main quarry sites or campsites, but blade cores and debitage are rare at smaller camps and kill sites. It is also possible that some of the wedge-shaped cores were originally conical, though the lack of blade scars on their backs argues against this idea.

  There is another curious characteristic of the conical cores: on most specimens the final flaking was the rejuvenation of the platform, a step usually subsequent to the removal of the last blade. Almost no negative bulbs of percussion remain on any of the blade scars on the cores. Since this core preparation happened after shaping, the objects we have found may have been precores rather than finished and discarded blade cores. It is also possible that the large conical cores were some type of tool. We would like to see use-wear analysis of conical as well as exhausted wedge-shaped cores, some of which could make excellent adze edges thanks to the angle between their platform and core face. A lot of work needs to be done before these blade technologies are really understood.

  The remainder of the Clovis flaked stone technology is quite simple. Most blades or biface flake tools are unifacially retouched. Implements associated with kill/processing sites tend to be made on large biface flakes, with the retouching serving as both re-sharpening and tool backing. Although many of these pieces are called side scrapers, this simply indicates that the retouch is on the long side rather than the end (figure 2.11a). Microscopic examination of the edges indicates that most of these tools were used for cutting.

  Certain types of borers (figure 2.11b), denticulated blades (figure 2.11c), gravers and multiple gravers (figure 2.11d–g), and end scrapers on blades (figure 2.11h–i) are fairly diagnostic Clovis tools. End scrapers made on flakes are often sub-triangular in outline (figure 2.11j), and some examples have a slight spur on one or both corners (figure 2.11k). Concave scrapers are present but not common (figure 2. 11l). Micro–end scrapers (figure 2.11m) and retouched blades (figure 2.11n) also appear but may be undercounted in some assemblages.

  Borers are elongated drill-like tools usually made on flakes but occasionally made from broken bifaces and blades. This tool type is rare in the west and tends to be associated with large campsites. It was probably used to gouge out sockets and to drill holes in wood.

  Denticulated blades are those whose edge or edges have been unifacially flaked to produce a series of notches that form a toothed pattern. Marilyn Shoberg has reported that microscopic examination of these tools indicates they were used for cutting meat.44

  Gravers have a small, sharp point on the edge of a flake. Occasionally there is more than one point on the same tool, and these are termed multiple gravers. There is speculation that gravers were used to engrave bone and other hard materials, but they are so delicate that they could only have been used on soft matter such as hide or leather. They do work well for incising designs on leather. Another suggestion is that they were tattoo needles. A microscopic study by the archaeologists John Tomenchuk and Peter Storck found that multiple gravers were used as compasses for drawing or etching circles.45

  FIGURE 2.11.

  Clovis flaked stone tools: (a) large side scraper/knife; (b) borer; (c) denticulated blade; (d-g) multiple point gravers; (h-i), end scrapers made on blades; (j) end scraper made on flake; (k) spurred end scraper; (l) concave scraper; (m) micro scraper; (n) retouched blade.

  Adzes are an infrequently recognized tool type in Clovis assemblages (figure 2.12). These are relatively heavy tools that would have been hafted so that their working edge was perpendicular to the handle, like a hoe. They were usually highly formalized and are well described in the post-Clovis Dalton Complex in the southeast, but they have been recognized in Clovis assemblages only recently. For the most part, Clovis examples seem to be less formal than Dalton adzes, but as more of the former are identified it is likely that this impression will change. Adzes are designed primarily as heavy woodworking tools, but they may be used for other tasks as well.

  FIGURE 2.12.

  Adzes: (a) East Wenatchee; (b–d) Gault.

  Burins are flaked stone tools produced by striking away a portion of a flake, blade, or biface in such a way as to form a nearly vertical edge and a chisel-like end. This implement is common in many Old World Upper Paleolithic assemblages and in some cases is the dominant tool form. So far they are rare in Clovis tool kits, but there is evidence of burin use on the edges of broken flakes and core ridges. It is possible that flakes were intentionally broken to produce edges that would serve as small planes and possibly engravers. These tools are called pseudo-burins.

  Although eastern Clovis knappers knew about heat treatment to improve the flakability of raw material, they rarely used it, and the only evidence of this we are aware of in western Clovis assemblages is from the Anzick Site in Montana.46 The Carson-Conn-Short Site in Tennessee contained great quantities of burned chert cortex as well as color-altered tools and flakes, suggesting that heat-treatment was used at this site, but more research needs to be done to verify this interpretation.47

  ARTIFACTS MADE FROM PERISHABLE MATERIALS

  With the exception of Florida, where underwater sites have provided excellent preservation conditions, Clovis sites have produced relatively few bone, antler, ivory, or other perisha
ble material artifacts, so this significant aspect of Clovis material culture is relatively unknown.48 Nevertheless, bone and ivory artifacts that exhibit substantial intentional shaping and modification have been recovered. Expedient bone tools—those that show use but little formal shaping—are also present in many Clovis sites, especially mammoth kill locations.49

  Excavations in a site near East Wenatchee in central Washington that has been interpreted as an artifact cache have yielded an assemblage of stone artifacts and a group of bone implements. Of interest here are twelve bone tools that were recovered in direct association with large fluted bifaces and finished fluted projectile points. A thirteenth bone artifact remains in place, and there is evidence that a fourteenth was destroyed by carnivore activity. Although preservation is only fair to poor, some are well enough preserved for detailed description. Most, if not all, are slightly flattened cylindrical bone rods beveled on both ends (figure 2.13a–b). The bevels are always on the same side, but usually at slightly different angles at the opposite ends. These rods were made from mammoth limb bones or the antlers of large cervids. Some retain evidence of the cancellous tissue (the spongy interior bone), but grinding and polishing have obliterated the initial stages of manufacture. Most of the flattened, beveled surfaces exhibit shallow incisions perpendicular to the long axis of the rods. This hachuring (incised checkerboard-like cuts) is considered to be intentional roughening to aid in hafting, rather than decoration.