Across Atlantic Ice Read online

  Westward expansions may have occurred in different ways, including exploration along major rivers and the Gulf Coast. The river orientation of Clovis people continues in the prairies, with major concentrations of artifacts along rivers and especially at the junctions of major rivers and streams. For instance, the majority of Clovis points from Iowa have been found in counties along the Mississippi, Missouri, Iowa, and Des Moines rivers, and they are especially concentrated in the region adjacent to the confluences of these rivers.22 Moreover, the majority of recorded Iowa projectile points are made of raw materials from locations to either the east or the southeast.

  Expansion onto the prairies and plains and ultimately to the West Coast took place near the end of Clovis times, and it appears that the Clovis use of the west was by small bands of explorers rather than large groups of colonizing people. This exploration was wide ranging, covering virtually all geographic areas and ecological zones, but it did not include long-term settlement.

  That Clovis people met other peoples during these explorations is strongly hinted by the contents of the Fenn Cache in Utah.23 In this cache, along with the usual horde of typical Clovis bifaces and projectile points, was an item previously unknown with Clovis artifacts, a crescent bifacial artifact. Since this artifact form is thought to have its origin among Great Basin non-Clovis people, its presence in the Fenn Cache suggests that descendants of people who colonized the New World by expanding along the north rim of the Pacific Ocean encountered and exchanged technological ideas with Clovis peoples.24

  Once the large herbivores such as the mammoth, camel, and horse became extinct, the bison was left without effective competition for expanding grasslands, and herds flourished. As a reaction to this resource windfall, people in the Plains adapted to the new environmental conditions and focused their economy on bison. They modified their hunting equipment into more specialized and effective weapons tipped by the fully fluted projectile points we now call Folsom. The advent of Folsom technology coincides with the appearance of the first western large habitation sites, such as Adair-Steadman in western Texas, Lindenmeier in northern Colorado, and Hanson in northern Wyoming.25 These sites are comparable in size and contain tool kits that represent diverse activities such as we see in the large eastern Clovis sites.

  SUBSISTENCE

  Animal bones and other perishable specimens used to reconstruct subsistence activities are rare in their sites, but a few remarkable exceptions provide us a glimpse of the types of food Clovis people consumed. The primary animal remains found in western sites come from mammoths, and since these spectacular sites have been in the literature longer than those where a much more mundane diet is evident, the concept that Clovis people were mammoth specialists has become ingrained. The evidence even in the west is, however, beginning to change this image. At the Murray Springs Site in Arizona a small bison kill complemented a mammoth kill.26 A campsite nearby yielded broken artifacts from both the mammoth and the bison kills, leaving little doubt that the inhabitants were the perpetrators of both events. Further, there is a bison kill in western Oklahoma associated with Clovis points.27

  Camel bones and possible tools made of camel bone have been found at several Clovis sites, but there is no direct evidence that the Clovis people killed these beasts.28 At a Hell Gap Paleo-Indian bison kill on the outskirts of Casper, Wyoming, archaeologists found the remains of a camel that met his end several millennia before the bison. A Clovis point found in the vicinity lead to speculation that the hunter who lost the projectile point killed the camel.29

  Mastodon and mammoth remains along with Clovis artifacts made from ivory and horse bones have been found in submerged Florida sites, fitting the western stereotype of the mighty big-game hunter. However, western-style mammoth kill sites are rare in the east, where most complete Clovis projectile points are isolated specimens from upland settings. This suggests a procurement strategy in which an individual hunter or small group stalked small herds or even single animals, such as deer, that would have been common in the eastern woodlands. Burned fragments of caribou bone have been found in hearth features in the northeast, suggesting that late Clovis people hunted a variety of tundra and steppe animals.

  At the Shawnee Minisink Site in Pennsylvania, fish bones and hawthorn seeds were found in what appears to be a summer camp, pitched between a stone quarry and the Delaware River.30 At the Aubrey Site on the banks of the Trinity River near Dallas, Texas, evidence was found of mammoth, bison, ground sloth, and even small mammal and reptile procurement.31 Farther up the Trinity at the Lewisville Site, hearths with burned bone suggest a summer encampment where the main foods included small mammals, amphibians, reptiles and reptile eggs, and baked mud dauber larvae.32 And at the Kincaid Rockshelter in southern Texas, the Clovis menu included alligators, turtles, badgers, raccoons, and mice.33 Thus, it appears that Clovis peoples were capable of exploiting almost any environment they encountered and adapted readily to local fare. Still, it is noteworthy and unexplained why mammoth remains, with only a slight hint of smaller game resources, dominate the High Plains and western sites.

  FLAKED STONE TECHNOLOGY

  Clovis subsistence was accomplished in large part through the use of flaked stone tools. Although there are minor variations in their overall tool forms, there are nonetheless enough similarities that one can easily distinguish Clovis and descendant fluted point assemblages from all other archaeological cultures and time periods. Clovis sites exhibit very specific material culture and behaviors (table 2.1). By and large, regional differences developed in Clovis and post-Clovis tools through time, but they can all be linked through a list of related production and artifact traits that we posit have common origins.

  Much ink has been spilled over the significance of fluted point variation, but from our point of view the artifact forms recovered in most archaeological excavations usually represent the end point of a tool’s use, making stylistic comparisons difficult. After manufacture, artifacts were used, broken, repaired, broken, repaired again, and eventually discarded, only to be found millennia later by archaeologists, who try to make sense of their discoveries. Occasionally we get a glimpse of what Clovis points originally looked like, in the form of caches and hunting losses. However, given the state of the majority of artifacts we find, what is the meaning of variation? Are there common denominators we can use as rules of classification? We know of many definitions of what a classic Clovis point looks like. If the “type Clovis point” from Blackwater Draw defines the type (figure 2.6n), how does it compare to the rest of the Clovis points from the same site? There is in fact a wide range of variation in Clovis point forms from Blackwater Draw (figure 2.6a–h and k–n). A typologist would have trouble identifying several of these as Clovis, yet their context indicates that they are all part of the same flaked stone assemblage. Differences like these are evident even in kill sites, but variation is even more apparent in large habitation sites such as Gault in central Texas.34

  TABLE 2.1 Clovis Production and Artifact Traits

  Clovis people used a unique technology to produce a distinctive style of projectile point that can be distinguished even from other fluted points that are their regional successors. Even so, there is not a single form of Clovis point, and we see variations through time and space. Moreover, repairing broken points could change their length-to-width and base-to-blade proportions and affect the shape of the base or the area designed to accommodate hafting. In some cases, a point was modified and used as a knife or other type of tool. All Clovis points were probably not intended for identical uses, because Clovis hunters stalked and killed a variety of game ranging from mammoths to mice. Additional variation was introduced by differences in raw material and in the skill and experience of the knapper. Even the amount of time a flintknapper had to devote to the production of a replacement spear tip—for example, in emergency situations—could have resulted in variation from the norm.

  FIGURE 2.6.

  Clovis point variations: (a–c)
reworked and partially bifacially flaked from Blackwater Draw, New Mexico; (d–h) reworked from Blackwater Draw; (i–j) partially bifacially flaked from Queen Anne’s County (i) and Kent County (j), Maryland; (k–l) reworked and (m–n) original forms from Blackwater Draw; (o–p) partially bifacially worked originals from the Anzick Site, Montana (o), and the Drake Site, Colorado (p); (q) middle phase biface end-thinned on one face only from the Fenn Cache, Idaho; (r) finished and unfluted Clovis point from the Fenn Cache.

  Clovis projectile points were manufactured with a generalized bifacial flaking technology; that is, flakes were removed from both faces of the specimen during its manufacture (chapter 1 for more about bifaces). This is in direct contrast to unifacial technology, in which flakes are removed from one face only. While there are examples of Clovis preforms and points that were fluted on only one face (figure 2.6q) and some points that exhibit no fluting (figure 2.6r), there are quite a few that were flaked primarily on one face (figure 2.6a–c, i–j, and o–p) with the other only partly modified, occasionally by the removal of characteristic flute flakes from the base (figure 2.6i–j).35 Edge grinding normally extended from the base to the maximum width of Clovis projectile points, which is usually about a third of the length from the indented base. The indented bases were edge-ground as well, and some have slightly flared ears. The point edges are usually slightly convex and taper to a sharp but not elongated tip. For the most part, Clovis points are relatively thin and are thickest at about the location of the greatest width.

  What continues to interest us is not how much variation there is, even with the aforementioned circumstances, but how distinctive Clovis points, tool types, and manufacturing technology are across much of the Western Hemisphere over a span of several hundred years.

  The evidence from a variety of sites, including stone procurement areas, campsites, caches, and especially workshops, makes it clear that Clovis biface production was a highly developed technology. It is also evident that although Clovis shares many traits with other biface technologies, it has certain unique techniques and combinations of techniques. Clovis people used many of the by-products of their bifacial technology for tools, and indeed the technology was probably perfected in part to produce these other items. Particularly important were the large, flat biface thinning flakes, especially useful as butchering and skinning tools.

  Another distinctive aspect of the Clovis flaked stone industry was the selection of exotic materials, especially in the west. Clovis knappers went far out of their way to obtain the highest-quality and most colorful stone and acquired flake stocks and artifacts from vast distances, through exchange, direct procurement, or both. Clovis knappers had the pick of the litter at stone outcrops because they were often the first people to explore many of the geological areas they encountered. One stone type they sought out was quartz crystal (see Bradley et al. 2010, plate 1). Although high-quality optical quartz flakes reasonably well, its brittleness makes it a challenge to work. Why they chose this stone when many other, more suitable types were readily available is not clear. Perhaps it was because of its special optical or perceived spiritual characteristics. Certainly crystals have held special significance for people worldwide and throughout time. Although many non-Clovis knappers used rare stones, the degree to which Clovis people went to obtain colorful raw material is unparalleled in the prehistory of North America.

  When large pieces of stone were available, the preference was to make big bifacial flake cores, to create flakes as large and elongated as possible (figure 2.7a). These cores were part of the mobile Clovis technology and provided a portable stock of stone material.36 They were manufactured from the glassy stones obtained from local outcrops or gravel deposits. When a stone of suitable material was found, it was worked down to reduce its size and weight and to produce a desired shape. This process guaranteed the quality of the stone before it was carried away and made the core fairly compact and easy to transport. The core provided the raw material from which large, thin, flat, and straight flakes could be removed as needed.

  FIGURE 2.7.

  Large bifacial flake cores: (a) unmodified core from Fenn Cache, Idaho; (b) biface core from Anzick Cache, Montana, that was deliberately broken into two pieces (1 and 2) to make a tool from fragment 2, such as the projectile point in 3c (hachured background of c is fragment b2); (c) early stage end-thinned biface from Anzick Cache; shaded areas a'-b'on figures 1' and 2' illustrate unmodified faces of b1 and b2 that were conjoined along the break surface of b1 and b2; c and c' of 3' are unmodified faces of c.

  The bifacial cores, reduced by the removal of flakes to make tools, eventually became too thin to produce suitable flakes. Depending on the size of the biface, it would then be made into a knife or a spear point. When the core was too large for a single point, it could be broken (figure 2.7b) and the resulting pieces used to make two or more points (figure 2.7c). Pieces made from these fragments but not finished often retain a bit of the break in the form of a square edge (figure 2.7a', b', and c'). We should also point out that bifaces in various phases of production could be used as expedient tools, such as knives, when necessary. A multiple-use approach to stone use is extremely efficient for mobile groups, who only carried tested and useful stone.

  MASTERING THE ART OF OVERSHOT FLAKING

  Controlled overshot flaking is harder than it sounds. As a beginning flintknapper, I found consistent biface thinning difficult to master and a source of great frustration. This process produced the most mistakes, used up the greatest quantity of good stone, and became an obsession. For years one of my most distressing actions was driving a flake too far and removing a chunk of the other edge. I frequently threw down the piece in disgust. What has now become known as overshot flaking was to be avoided at all costs. It wasn’t until I saw the artifacts from the Sheaman Clovis site in eastern Wyoming that I got a glimmer of recognition that some overshot flaking may have been intentional. While studying the Sheaman flakes, the archaeologist George Frison recognized that some may have come from a single biface, and he set about putting them back together. He had found startling evidence that Clovis knappers not only recognized the value of overshot flaking for biface thinning but were able to control it. Once I experienced this epiphany, I tried to emulate the process. At that point I was a pretty good knapper, but I was humbled by the control and skill it takes to consistently apply overshot flaking, especially near the finish of an already flat, thin piece. After many years I still find it one of the most challenging techniques, but the idea that it is useful and can be controlled has finally achieved general acceptance among knappers, even though few have mastered it. Bruce

  Clovis biface technology is distinguished by the use of overshot thinning throughout the reduction process (figure Intro.5).37 We think this technique had two main purposes: to massively thin and flatten the biface, and to produce large, flat flakes for use as tools. It was also a handy way to remove irregularities on an opposite edge, such as squared-off areas. Some projectile points exhibit overshot flake scars, including the results of removing the small portion of edge opposite the platform side.

  Projectile point production, whether from pieces of large biface cores, unmodified raw material, or large flakes, followed the same general sequence: once a basic bifacial outline was achieved, thinning and regularizing became the main objectives. Although overshot flaking remained the primary approach, two additional but related techniques, normal thinning and full-face thinning, were also employed. Because overshot is difficult to accomplish and the outcome hard to control, platforms were individually made and carefully prepared to enhance success. The biface edge from which the flake was to be struck was turned or beveled toward the face from which the flake was to be removed. A segment of the turned edge was then isolated from the adjacent edges by the removal of small flakes from the flaking face until the striking platform protruded not only down toward the flaking face but also out from the adjacent edges. These carefully isolated platforms were centered on
a ridge between two flake scars. Isolated Clovis platforms were relatively wide and straight, as opposed to the convex isolated platforms produced by many other thinning techniques. When flaking preparation was finished, the striking surface was heavily abraded. This abrasion commonly curved over onto the adjacent flaking face. Finally, force was applied with direct percussion. The angle of the blow had to be nearly parallel to the biface surface and strong enough to produce a large flake. Even if everything went just right, it took incredible skill to control the flake termination so that it removed only a small portion of the opposite edge. In our experience, this highly distinctive platform preparation and flake spacing has resulted in flakes that are as culturally diagnostic in Clovis as the finished points.

  MORE THAN POINTS

  During the 1970s, working for George Frison on several Paleo-Indian sites in Wyoming, I first stumbled onto the observation that many Clovis biface thinning flakes had distinctive characteristics. We were working at the Agate Basin Site, in a locale where unidentified flakes were eroding from a bank. At first we found only flakes, many quite large and most of them biface. There was a range in size and form, but a large proportion had distinctive platforms and more or less parallel sides. What was noteworthy about the platforms, compared to what we were finding elsewhere, was that they were wide and straight. They were also heavily abraded. I remember being struck by their distinctiveness at the time. But it wasn’t until the next year that further excavations at the site, now known as Sheaman, revealed that it was a Clovis occupation. Since the discovery of the Sheaman flakes, I have seen many assemblages that have the same type of platform. These assemblages come from sites all over the Americas, and without exception they have been determined to be Clovis. Bruce