CONFERENCE ON THE ARCHAIC PREHISTORY OF THE NORTH AMERICAN SOUTHWEST

OCTOBER 26, 1996

RESULTS OF THE PROJECTILE POINT WORKSHOP SURVEY

Bruce B. Huckell
Maxwell Museum
February 4, 1997

My apologies for not having gotten this information compiled and available to you sooner. Now that 1997 is upon us, I wanted to pass along the results of the informal projectile point identification survey that we conducted on the morning of the workshop. A total of 18 respondents took part in the identification exercise, and I think that you will find the results both interesting and worrisome. With regard to the people who participated, it is probably fair to say that they represent a cross-section of Southwestern archaeologists, although of course responses were anonymous. However, in glancing over my shoulder to see who was taking part in the exercise, I believe that the range extended from people with long experience and interest in the Archaic and its suite of projectile points to those who have little experience with points and for whom the Archaic is not a principal focus of research.

The accompanying table presents the results of the exercise and the figure reproduces the outlines of the five points that people were asked to identify. I should state at the outset that there are, at present, no "correct" identifications--the reason that the workshop was held is that we currently lack a coherent approach to classifying points, despite our reliance on them as cultural-temporal markers. Therefore, the table simply catalogues the responses. Also, in order not to bias the exercise, no provenience information was provided as to where the points had been collected. For the record, Points 1-4, all of obsidian, were recovered by William Roosa in the early 1950s from the original group of sites close to Grants, New Mexico, that were used by Kirk Bryan and Joseph Toulouse (1943) to define the San Jose and Lobo complexes. These specimens were part of a lot of about 90 Archaic points and bifaces from these Grants sites that came to the Maxwell Museum of Anthropology a short time ago. Point 5 (please forgive me--I was just trying to round out the sample and it was the closest thing at hand!) is one that I manufactured of silicified limestone as a replica of a San Pedro point from southeastern Arizona. In the following paragraphs, I 'd like to offer a few observations about the results and what they tell us about the state of the art of Southwestern Archaic projectile point typology.

First, it is instructive to note that a total of 18 different type names were used by people participating in the exercise, along with a "no identification" category. The latter includes such responses as "?", or a blank space, or a "No ID" next to the point number. The 18 type names range from specific, traditional types to more generic labels such as "Late Archaic" or "Plains Corner-Notched." Most of the type names should be familiar; they represent types defined over the last 60 years that are closely tied to the Cochise, Oshara, and Great Basin models of the Archaic. Also, some of the problems of splitting types in the Oshara Tradition typology are reflected by compound identifications (San Jose/Armijo, Bajada/San Jose, etc.).

Second, and perhaps most striking, is that the maximum agreement of typological assignment on any one point--not counting "no identification"--was achieved on Point 5: an even 50 percent of the respondents identified it as San Pedro. Ironically, this was the San Pedro point replica. The minimum agreement, 27.8 percent, was on Points 1 and 3. However, for Point 1 this low percentage was achieved by a splitting of identifications between two different types, Chiricahua and Elko Earred (each with 27.8 percent). I suspect that this is a good reflection of where one was trained or has had work experience, for this looks like a pretty even northern Southwest-southern Southwest split on the identification of one morphological form with two different type names. For Point 3 only a single type, San Jose, achieved this same low percentage of agreement. I suspect that had provenience information for the points been provided, these low percentages of agreement would have risen. However, that information really should not be allowed to sway identifications that must be based first and foremost on morphological criteria.

Third, and closely related to the lack of agreement discussed above, note that none of the five points was assigned to fewer than five different types--Points 1 and 2 were assigned to five different types, Point 3 to six different types, and Points 4 and 5 to seven different types. Some of this is understandable, inasmuch as point number 4 has a morphology that has long tweaked typologists: Pinto, Bajada, San Jose, and Pinto/San Jose have all been used to refer to relatively short, straight-stemmed, concave-based forms with blades either serrated or not. The other point with seven different type names applied to it is Point 5. Interestingly, despite the fact that maximum agreement was achieved on this form, its simple morphology also permitted its assignment to Colorado Plateau and Trans-Pecos Texas types, along with one identification as Cody.

A fourth observation is that use of the "no identification" category ranges from a low of about 11 percent (Point 4), to a high of 61 percent (for Point 2). Although one might think it a good sign that only 11 percent of the respondents could not identify a point to type, notice that seven different type names (not counting "no identification") were used by the 18 respondents in placing Point 4. As for Point 2 with its 61 percent "no identification" rating, this should tell us that there are certain morphological forms that occur in the record that everyone has difficulty in assigning to a type. Only three (16.7 percent) of the respondents were in agreement in placing this point in a type (Armijo).

In summary, this exercise demonstrates that we clearly have a problem with the typological assignment of Archaic points and, moreoever, that the depth and breadth of the problem are perhaps more profound than we would like to think. The most salient results to emerge from the exercise are that: 1) many rather common points can easily be assigned to one or more types; 2) use of a particular type is often determined by where you received training or have most of your field experience or the geographic area in which you work; and 3) in the absence of objective morphological criteria and clear definitions of types, we will continue to have this problem. It is important to recognize that many points of particular, distinctive morphologies achieve distributions that blanket the entire Southwest, and that we should strive to recognize such points with a single name.

The course of action that was agreed upon to foster the creation of a more systematic approach to Southwestern Archaic points was to hold a series of regional conferences to bring together interested scholars and important collections. Tentative plans called for four conferences to be held, one each at Tucson, Flagstaff, Albuquerque, and Las Cruces; major collections would be assembled and identified at the host institution. Planning for these conferences is underway, and information about them will be provided as work continues and plans are developed. Thanks to all of you who participated in the projectile point identification survey, and to those of you who attended the Conference.

Back to Archaic Projectile Points Home Page