NOTES FROM THE FIELD, 1999: MORE HOPEWELL "HOUSES" AT THE STUBBS EARTHWORKS SITE

Frank L. Cowan, Ted S. Sunderhaus, and Robert A. Genheimer

Cincinnati Museum Center

The Cincinnati Museum Center renewed its campaign of extensive excavations at the Stubbs Earthworks site (33Wa1) throughout the summer of 1999. The 1999 field work was unanticipated and initiated on short notice, spurred by last-minute schedule changes that pushed forward the building of an additional access road onto the grounds of the new Little Miami High School. The access road crosses one of three of the Archaeological Conservancy's conservation easements, collectively named The Fleischmann Ancient Indian Culture Reserve, that preserve portions of the large Ohio Hopewell geometric earthworks complex mapped in 1839 by Charles Whittlesey (1852). The easement agreement allows strictly limited encroachments onto the easement property but provides for up to 120 days for archaeological investigations prior to the onset of construction.

The Cincinnati Museum Center initiated excavation of the 400-square-meter construction impact area on June 10, 1999 with the cooperation of the Little Miami School District, the Archaeological Conservancy, Inc., and the Dugan & Meyers Construction Company. The Fleischmann Foundation provided the generous financial assistance without which these field investigations could not have taken place. The deadline for fieldwork completion is early October; hence, this update is being written while we are still in the final days of the field recovery process.

The 1999 fieldwork is only the latest phase of intensive excavations that began in May 1998 in advance of the construction of the new Little Miami High School (Cowan and Clay 1998; Cowan, et al. 1998). During the six-month-long 1998 field season, Cincinnati Museum Center archaeologists conducted nearly three hectares of magnetometer survey and 3,400 square meters of subsurface excavation across the northern portion of the 86-acre school property (see Figure 1). The unprecedented scale of those investigations was made possible by the assistance of hundreds of volunteers, including professional, student, and amateur archaeologists, and by generous funding support from Gray & Pape, Inc. and 3D/International, Inc.

Figure 1. Map of Stubbs Earthworks site excavations for 1998 and 1999.

Highlights of the 1998 field season included documentation of the post mold remnants of thirteen wooden Hopewell structures, located both within and outside the conjoined circle-and-rectangle geometric earthwork enclosure. These included large-scale civic-ceremonial structures as well as structures that may have been temporary accommodations for participants who gathered at this site for periodic celebrations and events.

Inside the earthwork enclosure, two large submound wooden structures were identified to the north and south of the preserved mound easement (Transects 11 and 25; see Figure 1 for locations). These structures are almost undoubtedly parts of a larger, multi-chambered "big house" that was eventually dismantled and covered by mound fill. The nature and location of those structures substantiate earlier interpretations that the presently visible mound is but a small remnant of the once much larger, irregularly shaped mound mapped by Whittlesey. The excavated structures were entirely devoid of artifacts, mortuary facilities, and human remains, although some cultural debris was include in the soils used to refill the post molds prior to mound construction.

Located outside of the earthwork enclosure at a location where Whittlesey had mapped a large circular earthwork was Structure 8, an immense, 73-meter diameter circle of 172 huge postholes, exposed by Transects 15 - 20. The postholes could have supported posts 20-35 cm in diameter and several meters in height. The circular enclosure was eventually dismantled, the postholes were deliberately refilled, and the former circle of posts may subsequently have been capped by the ring-shaped earthen berm illustrated in the 1839 map. The post molds were often filled with sediments containing cultural debris, but since no artifacts were observed on the plow zone surface in that portion of the site during earlier surveys described by Genheimer (1996,1997), the artifact-bearing fill must have been transported from other portions of the site. A very small "C-shaped" structure, Structure 9, was found inside the circular post enclosure.

Outside of the earthwork enclosure, excavators documented nine "house-like" post structures: one in Transect 2, five in Transect 10, and three in Transect 26 (see Cowan, et al. 1998:Figures 3 and 4 for illustrations of the structure in Transect 2 and of four of the five structures within Transect 10). One of the more interesting aspects of the nine house-like structures is their architectural diversity. Three structures conform to conventional expectations for Ohio Hopewell house structures, i.e., they are rectangular and have rounded corners. On the other hand, one structure is square with squared corners, three are circular, one is "C-shaped," and one is a circular structure with evenly paired post molds forming inner and outer wall rings. To our present knowledge, this latter structural form has no analogs in eastern North American prehistory, much less in Ohio Hopewell. All evidence indicates that these architecturally disparate structures are Hopewell in origin.

The focus of the 1999 excavations is Transect 27, a 9.0 by 48.5-meter excavation covering the area where the proposed access road will cross the Archaeological Conservancy=s easement Tract One (Figure 1). The 5.7-acre easement tract encompasses the eastern half of the Stubbs Mill Blade site, 33Wa256, the largest and most artifact-rich of the 28 Hopewellian artifact concentrations that together make up the Stubbs Cluster in the Little Miami River valley (Genheimer 1996, 1997). The southernmost end of the Tract One easement, the location of Transect 27, exhibits very dense concentrations of fire-cracked rocks within the plow zone, but other artifact classes appear to be relatively sparse in that portion of the site.

Excavations were initiated by stripping the transect of plow-disturbed topsoils with a self-loading pan and a smooth-edged backhoe bucket. Mechanized plow zone and topsoil removal exposed 31 post molds and pit features. Subsequently, three-and-a-half months of repeated shovel-shaving and trowel-scraping of the 400-square-meter area have brought the total number of features within the transect to more than 380. Many post molds were not detectable until the transect surface had been stripped of as much as 15-20 cm of sub-plow zone subsoils.

Eight "house-sized" and "house-like" Hopewell structures are exposed by Transect 27. Figure 2 is a schematic sketch-map illustrating the sizes, shapes, and locations of the house-like structures and of some of the larger and more distinct pit features within the transect. All of the structures are rectangular with rounded corners. Three structures, Structures 14, 15, and 16, are sufficiently exposed to permit measurement of complete "house" dimensions. These vary from 5.5 by 5.5 meters to 7.0 by 6.0 meters. Structures 17 and 21 may be a bit larger, although the complete structure outlines are not exposed. The transect exposed only the southernmost walls and corners of Structures 18 and 19, and Structure 20 is represented only by its northwestern corner.

It is evident that this portion of the Stubbs Earthwork complex was a very busy place some 1,600 to 2,000 years ago. But it is also clear that these building remnants are the accumulation of repeated building and use episodes that took place over significant periods of time, as indicated by structure overlap (Structures 15 and 21) and by the direct superimposition of Structures 18 and 19. In addition, the extremely close proximity of some other "houses" (e.g., of Structure 15 with both Structures 17 and 19) suggests that those structures probably were not strictly contemporary. Nevertheless, the scant artifactual evidence indicates that all are products of Hopewellian period construction and use. Also intriguing are the apparently common orientations of many of these structures. Structures 15, 17, 18, 19, and 20 are identically oriented, and Structures 14 and 21 share the same northeasterly orientation. Structure 16 exhibits an orientation intermediate between the two more common alignments. It is conceivable that the principles that may have governed structure orientation in this portion of the site changed through time.

One of the surprising details of these structures is the method of their construction. Each of the eight structures within Transect 27 was built using a wall-trench construction technique, a building method conventionally thought to have become common in eastern North America only in Mississippian times. The wall-trenches are very subtle, even subtler than the post molds, and the evidence for their existence did not become apparent to us until quite late in the field season. Nonetheless, we have been able to trace out remnants of the broad, shallow wall-trenches for parts of each of the structures. The wall-trench features are characterized by very compact, somewhat mottled silty soils with trace amounts of fine gravels redeposited into linear depressions in the clayey silt B-horizon subsoils. Removal of wall-trench fills frequently exposed remnants of post molds previously "missing" from wall lines; the wall-trench sediments appear to have filled in the holes left by pulled posts, obscuring them from all efforts at detection at higher elevations.

Figure 3 is a sketch-map illustrating Structures 16 and 20 as examples of wall-trench construction in the easternmost end of Transect 27. The eastern wall of Structure 16 provided our first clue for the use of wall-trench construction, but we did not at first recognize the true significance of that wall-trench feature, and we did not generalize from it until we'd stumbled across additional "anomalous" wall-trench segments in other structures. Subsequent reexamination has extended the wall-trench to include the northeastern corner of the structure, and we have found profile evidence of the wall-trenches in the northern wall of the transect. We expect to be able to trace out at least some of the wall-trench along the southern and western sides of Structure 16 before the field season is concluded. Structure 20 is located in a corner of the transect that we didn't shovel-shave until very late in the excavation season, by which time we were looking for wall-trenches. Even in that instance, we found most of the post molds before we were able to detect the very subtle traces of the wall-trench outlines. Rodent burrow disturbances presently obscure a portion of the surviving Structure 20 wall-trench where an additional two or three post molds should be located.

Figure 3. Structures 16 and 20, with presently documented post molds and wall trenches.

Figure 4 is a northern wall profile of Transect 27 at the point where the western wall of Structure 16 extends beyond the excavated transect. The profile shows the broad, shallow wall-trench along with the mold of a wall post that was pulled when the structure was dismantled. Although the plow zone is very clearly differentiated from the underlying soils, the distinctions between undisturbed subsoil, wall-trench fill, and post mold fills are very subtle, even in cross-sectional profile.

Figure 4. Profile of Structure 16's wall-trench (F. 1237) and post mold (F. 1169), along the north wall of Transect 27.

The extreme subtlety of the distinctions between the wall-trench fills and their surrounding subsoils (often only minor differences in the ped structures of the two soils) suggests that only same-depth subsoils were usually being used to refill the wall-trenches once the post holes were dug and posts put into place. It seems unlikely that this could have occurred if the wall-trenches were excavated from the original ground surface, since very dark, readily visible topsoils would have been mixed into the wall-trench fills. Our tentative conclusion is that the prairie sods and topsoils must have been removed from the entirety of the house footprint to expose a B-horizon subsoil "floor" prior to excavation of the wall-trenches and the emplacement of the wall-posts. This conclusion must be considered, for the present, to be an untested hypothesis as we have, as yet, no direct evidence for even the most shallow semi-subterranean house-basins preserved beneath the plow zone. Yet, this hypothesis seems to be the most economical way to explain the difficulties we=ve had in detecting wall-trenches and the sudden appearance of subtle post mold "tops" deep within the sub-plow zone subsoils. In retrospect, we strongly suspect that at least all of the rectangular house-like structures we excavated in 1998 were also of wall-trench (and possibly house-basin) construction - we simply missed (or misinterpreted) the signs of those construction methods (we weren=t looking for them, after all). Whether or not any of the circular "houses" were built using the wall-trench method, we just don=t know and probably won=t be able to reconstruct from our recorded evidence. We are currently still seeking further corroboration for the house-basin hypothesis in Transect 27.

Despite increasingly compelling evidence that these "house"-like structures were well built and designed to last, there is equally compelling evidence that these were not routinely occupied residential (domestic) structures. Despite the presence of eight structures, the artifact content of Transect 27 is extraordinarily sparse. With the exception of fire-cracked rocks in the plow zone, we estimate that there are literally more features than artifacts in this transect. Even assuming compulsive fastidiousness in "house-cleaning" and site maintenance by the builders and users of this portion of the site, we would have expected that at least very small flakes, sherds, and bits of charcoal and burnt soil would have accumulated sufficiently that such debris would commonly have become trapped in vacant pits, post molds, and other depressions, or ground into subsurface soils. That is not the case. Most features are entirely devoid of artifacts, and features containing even traces of charcoal or burnt soil are remarkably rare. It would appear that these well-built structures were seldom occupied and were used only for relatively brief periods of time. There is nothing to suggest that the full range of daily tasks and activities were being carried out within or adjacent to these structures.

The Stubbs Earthworks site continues to astonish us with a wealth of new clues to the probable appearance of large Ohio Hopewell earthwork complexes during their actual periods of use. While we have tended to visualize these sites in terms of the historically surviving earthen architecture, it may have been primarily wooden architecture, including wooden architecture of truly monumental proportions, which dominated the local landscape during much of the active use-lives of these sites. Earthen structures may have gradually replaced wooden ones only in the later stages of site use. Temporary housing accommodations, also of wood, abounded all around the earthworks. Nearly every excavation unit not situated on low-lying ground and where there was even a modicum of surficial debris outside the area of the earthworks has yielded evidence for "house-like" structures and, usually, evidence for densely packed multiple structures. The seventeen "house-like" structures thus far exposed exhibit remarkable architectural diversity and unexpected methods of construction. These empirical observations have changed our mental images of what these large earthwork complexes must have looked like during their periods of use.

References Cited

Cowan, Frank L., and R. Berle Clay
 1998  Ground-Truthing Magnetometry Data at an Ohio Hopewell Site. Paper presented at the 43rd Midwest Archaeological Conference, Muncie, Indiana.

Cowan, Frank L., Ted S. Sunderhaus, and Robert A. Genheimer
 1998  Notes from the Field: An Update from the Stubbs Earthworks Site. The Ohio Archaeological Council Newsletter 10(2):6-13.

Genheimer, Robert A.
 1996  Bladelets are Tools Too: The Predominance of Bladelets Among Formal Tools at Ohio Hopewell Sites. In A View from the Core: A Synthesis of Ohio Hopewell Archaeology, edited by P. J. Pacheco, pp. 92-107. Ohio Archaeological Council, Inc., Columbus.

Genheimer, Robert A.
 1997  Stubbs Cluster: Hopewellian Site Dynamics at a Forgotten Little Miami River Valley Settlement. In Ohio Hopewell Community Organization, edited by W. S. Dancey and P. J. Pacheco, pp. 283-309. Kent State University Press, Kent, Ohio.

Whittlesey, Charles
 1852  Descriptions of Ancient Works in Ohio. Smithsonian Contributions to Knowledge, Smithsonian Institution, Washington, D. C.

Only registered users can write comments.
Please login or register.

Powered by AkoComment 2.0!