Stratigraphy and Chronology of the Stevenson Area
ChronologyBy Mark D. Mitchell
Four lines of evidence can be used to estimate the period and duration of occupation of the Stevenson area. These include a series of tree-ring dates; an analysis of associated pottery types; site stratigraphy and feature fill sequences; and data on feature construction, remodeling and abandonment. Chronological interpretations derived from these data can be compared to those developed for nearby sites, as well as to general trends in the occupational history of the region.
Tree-ring Dating
A total of 37 tree-ring dates are available for the Stevenson area (Table 1). These samples derive from five structures, including all four of the pitstructures and one of the pit rooms. The non-cutting dates range from A.D. 457 to A.D. 668, although more than 50 percent of these fall into the seventh century. All three cutting dates fall in the latter third of the seventh century, with one in A.D. 669 and two in A.D. 676.
| Provenience | LTRR Number | Species | Date |
|---|---|---|---|
| Pitstructure 1 | MTC-30 | Juniper | 519vv |
| Pitstructure 2 | MTC-6 | Juniper | 457vv |
| Pitstructure 2 | MTC-3 | Juniper | 539vv |
| Pitstructure 2 | MTC-5 | Juniper | 595vv |
| Pitstructure 2 | MTC-4 | Juniper | 605vv |
| Pitstructure 2 | MTC-1 | Juniper | 617vv |
| Pitstructure 3 | MTC-16 | Juniper | 504vv |
| Pitstructure 3 | MTC-12 | Juniper | 531vv |
| Pitstructure 3 | MTC-133 | Juniper | 536vv |
| Pitstructure 3 | MTC-136 | Juniper | 558++vv |
| Pitstructure 3 | MTC-37 | Juniper | 559vv |
| Pitstructure 3 | MTC-132 | Juniper | 571vv |
| Pitstructure 3 | MTC-135 | Juniper | 585vv |
| Pitstructure 3 | MTC-134 | Juniper | 595++B |
| Pitstructure 3 | MTC-13 | Juniper | 598vv |
| Pitstructure 3 | MTC-14 | Juniper | 605vv |
| Pitstructure 3 | MTC-8 | Juniper | 633vv |
| Pitstructure 3 | MTC-33 | Ponderosa | 635vv |
| Pitstructure 3 | MTC-34 | Ponderosa | 644vv |
| Pitstructure 3 | MTC-32 | Ponderosa | 648vv |
| Pitstructure 3 | MTC-35 | Ponderosa | 667vv |
| Pitstructure 3 | MTC-31 | Ponderosa | 676r |
| Pitstructure 4 | MTC-137 | Juniper | 564vv |
| Pitstructure 4 | MTC-24 | Ponderosa | 615vv |
| Pitstructure 4 | MTC-19 | Ponderosa | 619vv |
| Pitstructure 4 | MTC-26 | Ponderosa | 621vv |
| Pitstructure 4 | MTC-28 | Ponderosa | 628vv |
| Pitstructure 4 | MTC-20 | Ponderosa | 656vv |
| Pitstructure 4 | MTC-21 | Ponderosa | 656vv |
| Pitstructure 4 | MTC-22 | Ponderosa | 665vv |
| Pitstructure 4 | MTC-23 | Ponderosa | 668vv |
| Pitstructure 4 | MTC-27 | Ponderosa | 669r |
| Pitstructure 4 | MTC-29 | Ponderosa | 676r |
| Room 1.08 | MTC-140 | Juniper | 585vv |
| Room 1.08 | MTC-15 | Juniper | 598vv |
| Room 1.08 | MTC-139 | Juniper | 601++B |
| Room 1.08 | MTC-143 | Juniper | 622vv |
These data suggest that both Pitstructure 3 and Pitstructure 4 were constructed during or shortly after A.D. 676. While the most recent dates for Pitstructure 1 (A.D. 519) and Pitstructure 2 (A.D. 617) are earlier, there is no reason to believe that these non-cutting dates accurately reflect the years in which these pitstructures were constructed. The non-cutting dates associated with Pitstructure 3 span a period of 163 years, while those associated with Pitstructure 4 span 104 years. The presence of early non-cutting dates in these structures can be attributed to the longevity of juniper trees and perhaps to the use of deadwood for construction. With all non-cutting dates an unknown number of outer rings have been lost and it is not uncommon for tree-ring samples in this region to have lost many outer rings to deterioration in the years since site abandonment. These same factors—rather than an earlier construction date—likely account for the early non-cutting dates in Pitstructures 1 and 2. Accordingly, tree-ring dates cannot be used to determine the construction dates of these structures, or of any of the pit rooms.
Ceramic Cross-dating
Preliminary ceramic analyses conducted by Wheat and his students indicate that Stevenson area gray ware pottery is dominated by Chapin Gray, although small amounts of Twin Trees Gray also are present. The white wares consist entirely of Chapin Black-on-white, which account for approximately five percent of the total assemblage. Very small amounts (less than one percent) of much later, intrusive types also are present. This is not surprising considering the proximity of the Pueblo II and III occupations in the Porter area of the site.
In the central Mesa Verde region dated assemblages similar to that of the Stevenson area span the period between A.D. 575 and A.D. 725. The absence of brown wares, such as Sambrito Utility, indicates that the Stevenson area was probably not occupied prior to that time. Twin Trees Gray was produced as early as A.D. 400 or 450, but persisted in some areas until A.D. 700; it appears to have been a very minor component of assemblages post-dating A.D. 600 (Wilshusen 1999aWilshusen, Richard H.
1999a Basketmaker III (A.D. 500-750). In Colorado Prehistory: A Context for Southern Colorado Drainage Basin, pp. 166-195, edited by William D. Lipe, Mark D. Varien, and Richard H. Wilshusen. Colorado Council of Professional Archaeologists, Denver.; Wilson and Blinman 1991Wilson, C. Dean and Eric Blinman
1991 Ceramic Types of the Mesa Verde Region. Handout prepared for the Ceramic Workshop, Annual Meeting of the Colorado Council of Professional Archaeologists, Boulder, Colorado.). The absence of red wares in the Stevenson assemblage indicates that it pre-dates A.D. 700 or 725 (Wilshusen 1999aWilshusen, Richard H.
1999a Basketmaker III (A.D. 500-750). In Colorado Prehistory: A Context for Southern Colorado Drainage Basin, pp. 166-195, edited by William D. Lipe, Mark D. Varien, and Richard H. Wilshusen. Colorado Council of Professional Archaeologists, Denver.; Wilson and Blinman 1991Wilson, C. Dean and Eric Blinman
1991 Ceramic Types of the Mesa Verde Region. Handout prepared for the Ceramic Workshop, Annual Meeting of the Colorado Council of Professional Archaeologists, Boulder, Colorado.).
These data are in general agreement with the tree-ring dates for Pitstructures 3 and 4, and support the conclusion that the Stevenson area was occupied during the latter third of the seventh century. Moreover, the similarities among the assemblages associated with each of the Stevenson area's four pitstructures suggests that all of them were occupied during this period.
Site Stratigraphy and Feature Fill Sequences
No master stratigraphic profile is available for the Stevenson area, but data on local soil series can be used to estimate the degree of recent sediment deposition or erosion. The Stevenson area is located on soils of the Witt series, which form in silty calcareous loess on slopes ranging from one to 12 percent (Leonhardy and Clay 1985Leonhardy, Frank L. and Vickie L. Clay
1985 Soils. In Dolores Archaeological Program: Studies in Environmental Archaeology, compiled by Kenneth L. Petersen, Vickie L. Clay, Meredith H. Matthews, and Sarah W. Neusius, pp. 139-153. Bureau of Reclamation, Engineering and Research Center, Denver.). The Stevenson area is located on the southeast-facing side of a low ridge that varies in slope from about five to eight percent. On Witt series soils pedogenic carbonates accumulate at depths ranging from 1.0 to 1.5 meters. In the Stevenson area excavation depths varied from as little as 10 centimeters to as much as 1.9 meters, but no mention is made in the field notes of intact caliche (with the possible exception of Room 1.13). If the surface of the ridge on which the Stevenson area is located had been truncated or deflated, either before or after Basketmaker III times, then it is likely that the excavations would have encountered caliche.
Prior to Wheat's excavation all of the Stevenson area had been plowed to a depth of about 20 or 30 centimeters. Although this disturbance made it impossible to define the relationship between the prehistoric and modern ground surfaces, it is evident that most or all of site's major features originated on or near the modern ground surface. Major elements of each structure, such as wing walls, slab-lined cists, and slab walls, were encountered at the base of the plow zone. In some cases, plowing appears to have directly affected these features. This suggests that the modern ground surface approximates the prehistoric surface.
The paucity of occupational debris suggests that the occupation of the site was relatively short. Although thin, sparse cultural deposits are present on the site surface, nowhere do they appear to exceed about 20 centimeters in thickness. The densest, most extensive, deposits are located near the foot of the ridge, south and east of Pitstructure 2; however, this midden area was not investigated during Wheat's excavations.
This conclusion is further supported by the fill of the site's pitstructures. Although structure fill deposits were not described systematically, and a measured sediment profile is only available for Pitstructure 2, the fill of the pitstructures and pit rooms consists primarily of roof fall. Intact deposits of domestic debris appear to be absent; field notes indicate that the fill of most pitstructures contains relatively few artifacts. Moreover, while artifacts were not systematically collected from fill contexts, ceramic density values are comparable to those from well-dated, short occupation duration Basketmaker III settlements (e.g., Brisbin and Varien 1986Brisbin, Joel M. and Mark Varien
1986 Excavations at Tres Bobos Hamlet (Site 5MT4545), a Basketmaker III Habitation. In Dolores Archaeological Program: Anasazi Communities at Dolores: Early Anasazi Sites in the Sagehen Flats Area, compiled by A.E. Kane and G.T. Gross, pp. 117 210. Bureau of Reclamation, Engineering and Research Center, Denver.).
Construction Sequences, Remodeling and Abandonment
Very little superpositioning is evident among the site's features. In a few cases it is possible to identify a sequence of construction for adjacent structures. For example, Room 1.02 was constructed after 1.01 had been destroyed, and after Room 1.03 had been built. Rooms 1.10, 1.11, 1.12 and 1.13 may have been constructed sequentially. Roomblock 2 appears to have been constructed over the course of three or more building episodes. However, the lack of superpositioning suggests that most of the site's structures were in use concurrently, or were constructed successively over a relatively short period.
There is very little evidence for structure remodeling. Room 1.05 appears to have burned and been rebuilt. Minor changes were made to the interior features of Pitstructure 4. This lack of reconstruction suggests that the structures of the Stevenson area were not occupied longer than a single use-life. Ethnographic data indicates that such structures were seldom occupied more than 10 or 12 years (Schlanger 1987Schlanger, Sarah H.
1987 Population Measurement, Size, and Change, AD 600-1175. In Dolores Archaeological Program: Supporting Studies: Settlement and Environment, compiled by Kenneth L. Petersen and Janet D. Orcutt, pp. 567-613. Bureau of Reclamation, Engineering and Research Center, Denver.).
All four pitstructures were abandoned in the same way. Each was burned, probably intentionally. In several cases floor features were filled with clean sand and sealed with plaster, suggesting ritual closure of the structure. This may indicate that all of the structures were abandoned simultaneously and the site was not reoccupied.
Chronological Assessment
These data indicate that the Stevenson area was occupied for a relatively short period between about A.D. 675 and A.D. 700. The lack of extensive occupational debris; the absence of feature superpositioning; and the lack of evidence for remodeling or reconstruction indicate that the occupation of the site was brief. Moreover, the lack of domestic debris in the fill of the pitstructures and the uniformity of the associated ceramic assemblages indicate that the site's principal structures were occupied either simultaneously or in quick succession.
The three available tree-ring cutting dates suggest that Pitstructure 3 and Pitstructure 4 may have been constructed and occupied concurrently. Although it cannot be directly demonstrated, the lack of domestic debris in the fill of Pitstructures 1 and 2 suggests that they also were in use at that time. The presence of more than one contemporaneous pitstructure has been noted at other nearby Basketmaker III hamlets. For example, at Knobby Knee Stockade, pitstructures with and without antechambers may have been occupied at the same time (Morris 1991Morris, James N.
1991 Archaeological Excavations on the Hovenweep Laterals. Four Corners Archaeological Project Report 16. Complete Archaeological Service Associates, Cortez, Colorado.).
This chronological interpretation differs from that developed by Wheat (n.d.a.Wheat, Joe Ben
n.d.a. The Architecture of the Stevenson Site. Ms. on file, University of Colorado Museum, Boulder.). He argues that the Stevenson area was used over a period of about 200 years, beginning during the first decades of the sixth century. In his account, Pitstructure 1 was occupied first in the early sixth century, followed by Pitstructure 2 in the early seventh century. Pitstructures 3 and 4 were occupied successively during the late A.D. 600s and early A.D. 700s.
Recent research has demonstrated that the Basketmaker III occupation of the central Mesa Verde region was confined primarily to the A.D. 600s and early A.D. 700s (Wilshusen 1999a:172, Table 6.1Wilshusen, Richard H.
1999a Basketmaker III (A.D. 500-750). In Colorado Prehistory: A Context for Southern Colorado Drainage Basin, pp. 166-195, edited by William D. Lipe, Mark D. Varien, and Richard H. Wilshusen. Colorado Council of Professional Archaeologists, Denver.). An A.D. 519 date for Pitstructure 1 would make it the earliest Basketmaker III structure in the region. However, there is no chronometric, architectural or assemblage data to back such a claim. Moreover, the available stratigraphic data, coupled with structure use-life estimates, indicate that the Stevenson area was probably occupied for one, or perhaps two, generations. This alternative explanation, although differing from Wheat's original interpretation, has been shaped by the fifty years of research that began with his pioneering excavations at Stevenson.