Artifact types provide an organizational tool that allow archaeologists to group specimens into bodies which have demonstrable meaning in terms of behavior patterns. Type as a concept has variously been defined as a demonstrable clustering of attributes observed among specimens of a given class of phenomenon. (I bet you don't know who wrote that? And if you do, I bet you don't know that he studied under Coe and that he gained much of his insight during his formative years working on the Roanoke River----it was Lewis Binford.)
Type is a cluster or pattern of attributes which distinguish a group of artifacts from other units and defines them as a class, and as a set of somewhat variable attributes which can be observed to occur together in the majority of cases in such a way as to permit comparison. Numerous archaeologists and researchers have suggested that types represent the idealized form of an object from the perspective of the craftsman. Deetz has referred to this concept as a mental template where the form of an object presents a close facsimile of the template and variation is accounted for in the variance of the ideas that produce them.
Joffre Coe's pioneering work on the Yadkin, PeeDee and Roanoke Rivers in North Carolina demonstrated two important facts; (1) that stratified sites of depth and antiquity do exist in the floodplains of the Piedmont and (2) that when an occupation zone can be found that represents a relatively brief period of time, the usual hodgepodge of projectile point types are not found, only variations of one specific theme. This latter fact, called the "Coe Axiom" by Louis Brennan, clearly demonstrated the diagnostic value of projectile points, but to be useful, the types had to be defined with precision. At the time of the writing, Coe reflected that the southeast had been unusually lackadaisical in developing working types in this category. Since that time, a plethora of projectile point types have been developed, but little synthesis has occurred. Coe's typological and chronological sequence forms the basis for the Paleo-Indian, Archaic and Woodland period identifications in North Carolina and has been applied for diagnostic purposes in other regions as well. The organizers of this conference want to get back to basics to renew dialog concerning the use of rhyolite by prehistoric peoples and revisit the North Carolina cultural and chronological sequence established by Joffre Coe some thirty-five years ago. The nature of Coe's projectile point sequence has remained virtually unchanged because it was based upon superimposed and stratigraphic zones as well as projectile point types that were precisely defined.
This workbook (which we will receive in the next stage) reproduces the original type descriptions from Coe's classic monograph Formative Cultures of the North Carolina Piedmont as well as several additional types considered refinements to the sequence. An underlying theme of Coe's sequence was that the manufacture of a projectile point was not haphazard. It required transmission of appropriate knowledge from one generation to another to approximate the correct form. The knowledge of how to manufacture a point, where to obtain resource material and how to use a specimen were passed from generation to generation as traditional knowledge. Each projectile point should exhibit slight differences in manufacture as replication of the ideal form progresses from generation to generation as traditional knowledge. Even though the first specimen produced in a lithic tradition may differ greatly from the last, if a large enough sample of specimens were observed, the thread of continuity would become clearly visible. If a tradition of manufacture can be identified and substantiated by stratigraphic discoveries in a number of distinct locales, it is then possible to recognize that particular tradition of manufacture through time and across space. Intrusive technological traditions may also be recognized. Recognition of attributes common to a particular tradition allows the archaeologist to go beyond pigeon-holing and make more meaningful interpretations from these ancient pieces of stone.
A technological tradition must exhibit two key components. Evolution, the changing of form through time and space, and continuum where certain elements, traits or attributes persist through time within a relatively restricted geographical area. If all projectile points ever produced by a specific tradition could be aligned in sequence from the first to the last, an unbroken linear development should be observed. This represents an ideal situation which archaeologists do not encounter since we deal with fragments of the archaeological record.
With the aid of large numbers of specimens, from substantiated time/space contexts, archaeologists can connect these fragments and recognize a technological tradition. Such is the case for the North Carolina projectile point sequence [View NC Projectile Point Chronology Chart] where we see various traditions represented; the most dominant being the Piedmont tradition. But we also see a tradition of triangular point manufacture and several intrusive traditions or representatives of different technologies within the Archaic.
Projectile points found at numerous stratified sites in North Carolina reflect an evolutionary development of certain types along a time/space continuum. This continuum is expressed between those types which are felt to have ancestral roots within the regional area as opposed to those which represent intrusive elements with respect to Coe's organization of this concept and his early work in the North Carolina Piedmont. This continuum of lithic manufacture has been identified as a Piedmont tradition.
The Piedmont tradition may have first developed as a regionalized technological modification in projectile point style by the Paleo-Indians. Many of the oldest points recovered from excavated contexts in North Carolina display facial fluting and in cases when the side-notching or basal portions were missing, they could be mistaken for fluted points of the Paleo-Indian period. The Hardaway Blade, Hardaway Dalton, and Hardaway side-notched are the earliest points discovered in situ in North Carolina. These types have been consistently found in the lowest depths of the Hardaway site associated with distinctive unifacial side scrapers morphologically identical to those found at many classic Paleo-Indian sites.
Coe felt that the earliest of Hardaway types dated close to 10,000 BC and that all three types occurred as variations during a relatively long period of time prior to the beginning of the Archaic. This possibility is supported by the occurrence of small hafted snub-nosed end scrapers identical to those of many late Paleo-Indian complexes with the succeeding and superimposed Palmer complex and a continued trend of decreasing point size through time.
Given a large enough sample of projectile points, it does not require a great deal of imagination to visualize the progression of the Hardaway side-notched into the Palmer corner-notched type. Average sizes are generally quite close, however, the U-shaped corner notches of the latter bases continued to be ground, but no longer show degrees of concavity. Straight, thoroughly ground bases are one of the most characteristic traits of the Palmer corner-notch type. It seems probable that the emphasis placed upon basal grinding may have led to the development of the straight base from the concave based Hardaway side-notched type.
Blades continued to be small and triangular often with deep serration. The use of small hafted snub-nosed end scrapers was the only other change in the cultural inventory to distinguish the Palmer complex from the preceding Hardaway complex.
Evolutionary development of the Piedmont tradition continued as the Palmer corner-notch evolved into the Kirk corner-notched type. A noticeable increase in average size occurred. This increase may have resulted from an emphasis placed upon percussion flaking in the Kirk as opposed to pressure flaking in the Palmer points. The large triangular blades were made with broad, shallow percussion flakes. Lateral edges were shaped by pressure flaking and deep serrations were made as a final step.
Bases ceased to be ground and corner notches became broader and less pronounced as the Kirk corner-notched evolved into the Kirk stemmed type. Blades became narrower and deep serrations continued to be emphasized as the broad notches of the Kirk stemmed evolved into the straight, nearly square stem of the Kirk serrated type.
Coe's estimate of relative age for the Kirk types at 5,000-6,000 BC was conservative. Work conducted later by Chapman, Broyles, Gardner and others identified radiocarbon dates between 6,700 BC and 7,500 BC. These dates indicated a greater antiquity for not only the Kirk types, but also those of the preceding Palmer and Hardaway periods. Suggestions have also been made that the Big Sandy and Lecroy types may have developed from the Kirk horizon. As populations increased during the Early Archaic, new territories were exploited and regional variations of projectile point typology likely occurred as a consequence of divergence across space and through time. Thus, higher frequencies of occurrence for Big Sandy and LeCroy types might be found in areas geographically removed from the central Piedmont of North Carolina.
The Kirk serrated evolved into the Stanly
stemmed type. As the
straight, square stem of the Kirk serrated became smaller and developed a
shallowing curvature of the base blades became broader and more triangular
in shape often resembling a typical christmas tree shape. Bases were thin
and lacked basal grinding. Average size declined, but techniques of
manufacture remained similar. The Stanly stemmed type has been found
associated with polished stone, semi-lunar atlatl weights. Larger points
of this type blend with the smaller of the succeeding Savannah River type.
The development of the Savannah River stemmed represents a shift in the general tendency of the Piedmont tradition types to decrease in average size through time. The Savannah River stemmed type is larger, broader and heavier than any that came before or after it. This dramatic shift in size may once again signal a technological modification or adaptive change. Bases were occasionally straight, but usually display a slight basal curvature reminiscent of the Stanly stemmed type. Percussion flaking predominated. Retouching and pressure flaking occurred only to smooth out irregularities. The Savannah River type has been associated with winged atlatl weights, grooved axes and soapstone vessels in several North Carolina sites. In North Carolina, all such discoveries have been associated with pre-ceramic, Late Archaic contexts, circa 1,000-3,000 BC.
The earlier portion of this period was typified by classic Savannah River stemmed points, rock-filled pit hearths, full-grooved ground stone axes and perhaps an absence of soapstone vessels. Later portions of the Savannah River period have been characterized by generally smaller variations of a projectile point type called Small Savannah River Stemmed. This type was first identified by Stanley South in 1959 from the Roanoke River in a number of stratified contexts and in other areas has been identified by various names by different researchers. One such example is the Otarre Stemmed type defined by Bennie Keel from the western mountainous region of North Carolina.
The Small Savannah River is a medium sized broad triangular stem and a straight or slightly excurvate base. Specimens do not appear to have been reworked from the larger class of Savannah River form. Instead they reflect a consistent attempt to manufacture a proportionately smaller form.
The terminal expression of stem point manufacture in the Piedmont is the Gypsy stemmed type defined by Oliver through analysis of specimens from the Doerschuk, Gaston, Thelma, and Warren Wilson sites. This small triangular bladed point with a square or rectangular stem and straight, slightly incurvate or excurvate base, represents a lineal descendant of the Savannah River and Small Savannah River types and a continued development of the Piedmont tradition into the early ceramic period where it terminates. The Gypsy stemmed type co-occurs with large triangular points and early cord and or fabric-marked ceramics. It dates from near 1,000 BC until several centuries before the Birth of Christ. Gypsy stemmed points may represent attempts by local populations to adapt traditional stemmed point manufacturing techniques for use with the bow and arrow.
The introduction of the bow and arrow coincides with the development of, or adoption of, a triangular tradition of point manufacture. The spread of this introduction appears to have been serpentine in that areas of the Piedmont have an earlier termination of stemmed point manufacture than those found in the mountains to the west. In both regions once triangular points are found in the archaeological record, stemmed point manufacture ceases to have importance and is replaced by triangular types. (We can also see evidence of triangular point technology stretching from the Alaskan sub-arctic at the Denbigh site which Griffin addressed, across the Canadian sub-arctic and down into the northeastern United States as evidenced by Tuck's identification of triangular points with the Squibnocket complex and then down into North Carolina.) This gives us some theoretical ground where we can propose hypotheses that could be tested. If this serpentine fashion is migrating to the south, we should see earlier evidence in the northern regions and later introductions of the triangular technology in the areas further to the south, just as we did in North Carolina with later introductions occurring in the mountainous regions.
Triangular projectile points signal the introduction of the bow and arrow to North Carolina. This technology did not develop independently in the Piedmont. There was no precedent. Instead, it appears to have been introduced from the north. Gradually through time, larger triangular forms gave rise to smaller triangular points, as they became better made and more aerodynamic. In the Piedmont, the earliest large triangular type is Badin Triangular, which evolves through time into smaller, and successively later, point types called Yadkin, Caraway and Clarksville.
Stemmed points were not as significant an occurrence again until between 1725 and the 1800s. During this period at least one group of Indians in the central Piedmont made narrow, crudely stemmed points, some of which were reworked from older flakes and Archaic points. Coe felt this type, called Randolph stemmed, resulted from people going back to bow and arrow usage after supplies of guns and ammunition were no longer made available to the Indians.
The construction of chronology is paramount to archaeological interpretation. The use of projectile points as indicators of cultural and chronological change must be firmly based on superimposition and stratigraphic separation of the type specimens at a significant number of sites within a restricted geographical area. To have meaning, a type must have recognizable form that can be easily identified and associated with a time/space context. The larger the data base and the more numerous the observations with the same stratigraphic context, the more successful the concept is apt to be. That is our purpose here today, to address the concept and to communicate a little bit better concerning the cold hard stones we find in the past. But to extract meaning from the cold hard stones of reality, we have to have larger samples, better data and more meaningful interpretations.