Matthew P. Purtill

Gray & Pape, Inc.

Ohio Archaeological Council © 2000


The Madisonville Village and Cemetery Site (33Ha36) located in southwestern Ohio has long been recognized as one of the premier prehistoric/protohistoric archaeological locales in the state and beyond. A large agricultural village occupied primarily during the late Fort Ancient period (ca. AD 1400 - 1643), the site has yielded some 1,450 burials, nearly 1,300 features, and tens of thousands of artifacts including 460 European trade items (Drooker 1997). Recently, several village-wide renovation and commemorative projects were initiated by the Village of Mariemont (Mariemont), which owns part of the site property, and the Madisonville Preservation Foundation (MPF). Chief among these was a proposed commemorative pavilion to be built on location at the Madisonville Site. Unfortunately, no formal archaeological assessment/work was undertaken prior to the initiation of construction. Construction entailed mechanical excavation of a foundation block resulting in unplanned exposure of archaeological deposits. At that time, construction efforts were halted and Gray & Pape was hired to provide a damage assessment and recommendations. The results of this construction project raised concerns throughout the archaeological community. The focus of this paper is to provide a detailed account of the events that led up to this unfortunate occurrence, present the results of the archaeological assessment and recommendations regarding the pavilion area, and to update interested parties about the current status of the project(s).

Project Planning and Implementation

In 1999, as part of the 75th Anniversary Commemorative and Renewal Project, Mariemont and the MPF began a cooperative effort to undertake several renovation and construction projects throughout the village (Lemon 2000). Foremost among the proposals was the construction of the John P. Nolen Pavilion to be located in Dogwood Park, which is situated in the southwestern portion of the Madisonville Site (Figure 1). The purpose of the pavilion was to commemorate the significance of the archaeological site. Funding for the project was provided through grants, private donations (no less than 38 companies/groups) and a large matching grant provided by The Thomas J. Emery Memorial (Mariemont Preservation Foundation 2000). A landscape architect firm and a construction company were hired to design and execute the project. Because of private funding, archaeological assessment under Section 106 of the NHPA was not required and no formal survey was pursued prior to construction.

Figure 1. Map illustrating pavilion location relative to previous investigations.
Figure 1. Map illustrating pavilion location relative to previous investigations.

A location near the bluffs edge overlooking the scenic Little Miami River Valley was chosen to construct the pavilion. Tree clearing, seeding of a lawn, and tree planting (on Arbor Day after all!) was undertaken as early as 1998 (Allen 1999). In early 2000, a 10.5 x 7.5 meter pavilion foundation block was excavated to approximately 60 centimeters exposing numerous prehistoric artifacts. Upon this "discovery," the project was halted by Mayor C. Michael Lemon of Mariemont who feared further damage to the site. After construction was suspended, word of the project spread and local artifact collectors became problematic as several instances of backdirt collecting are rumored. At one point, one or more individuals, seeing bone protruding from the pavilion floor, hand dug the area revealing what was reported later as a "headless infant burial." In order to better control the situation, Mayor Lemon made the site off limits and had police patrol the area regularly, confiscating artifacts from collectors (including one civil servant).

In hopes of seeking guidance, Mariemont voluntarily contracted with Gray & Pape, a CRM firm in Cincinnati, on March 1, 2000 to preform an archaeological assessment of the pavilion block. This work was to identify if any intact archaeological deposits were present within the floor of the pavilion and recommend if additional archaeological work should be undertaken. No formal excavations were planned for this stage. Fieldwork was accomplished between March 2 and 5, 2000 under the direction of Matthew Purtill with the assistance of Patrick Bennett (Gray & Pape). During these investigations several individuals volunteered their time including Ted Sunderhaus, Frank Cowan, Carly Meyer, and Don Frodge.

Results of Archaeological Investigations

At the time of Gray & Pape's involvement, a rectangular (10.5 x 7.5 meters) foundation trench had already been excavated (60 cm bgs). The floor was shovel/trowel scraped revealing an extensive remnant sheet midden, 25 storage/trash pits (F-00-1-4, 6-7, 9(9A)-11,13-27), nine possible post molds (PPM-00-1 - 9), evidence of historic disturbances (looting), and one dog burial (F-00-12) originally thought to be the headless infant (Figure 2). Two soil stains originally assigned feature numbers (F-00-5 and F-00-8) were later determined to be remnant sheet midden. An Oakfield soil probe was used to investigate dark stains to distinguish between true pit features (resulting in deep probes) and remnant sheet midden (resulting in shallow probes). Although artifacts were abundant throughout the foundation trench and the spoils pile, the nature of our investigations did not call for wholesale collection or interpretation. Accordingly, this article focuses primarily on feature discussion.

Figure 2. Plan view of excavated pavilion foundation floor.
Figure 2. Plan view of excavated pavilion foundation floor.

The most sensitive issue regarding these investigations was the determination of the existence of the reported headless infant burial. Christina Beatty of MPF directed us to the location of the remains which had been reburied. It was decided to minimally expose the remains to verify them as human or animal. Excavation revealed a highly disturbed domesticated (?) dog burial situated within a feature (F-00-12). The remains are oriented due south with the dorsal side up and with front/hind legs flexed underneath its body. The cranium was absent but several associated fragments, with fresh breaks, indicate that the head was recently crushed probably during hand digging. The location of the fractured cranium is unknown and appears to have been removed off-site. Interestingly, at this time it is not entirely clear if the burial represents a prehistoric or historic dog. The remains were entirely defleshed yet still greasy. Moreover, local informants indicated that this portion of the park was a favorite "pet cemetery" for neighbors for many years and that Mariemont disposed of road-kill animals nearby.

With only limited testing, scant information is available for the storage/trash pits. In general, soil probing identified pit features as being at least three feet in depth below the excavated foundation floor. Previous excavations have demonstrated that such features often reach a depth of over six feet (Hooten 1920). A number of these pit features (F-00-4, 12, 19, 20, 21, 23, 24, 25, and 27) appear to have been historically disturbed or looted in the recent past as evidenced by extremely loose matrix. In addition, two suspected looters trenches were identified. One trench (containing F-00-23) yielded a pop/beer twist-off cap illustrating downtown Cincinnati with old Riverfront Stadium suggesting a post 1969 excavation date. Features F-00-1 and F-00-2 were mottled with subsoil and contained irregular outlines also indicating historical disturbance, however, compact feature matrix suggests antiquated activities. Although speculative, F-00-1 and F-00-2 (potentially others) may represent the excavations of Charles Metz during his 1879-1880 investigations (Drooker 1997; see also Figure 1). Two features (F-00-9/9A and 10) appear to be intact based on soil probing and the occurrence of well-defined, regular borders. Soil probing of F-00-10 suggests a stratified feature similar to several "cache pits" excavated previously (Hooten 1920:31, Figure 1). Finally, a number of possible post molds (PPM-00-1-9) were identified haphazardly across the pavilion floor. These features do not form any recognizable patterns, although limited exposure (including abundant remnant midden) hampered any definitive determinations.

Where do we go from here?

Based on a March 3 field meeting, Mr. Purtill was informed that the majority of the pavilion foundation block was already excavated to the desired depth. Only portions of the foundation would have to be excavated deeper during the construction of eight piers (2 feet in diameter and 15 feet in depth) and associated grade beams (2 x 2 feet). From this, a letter report with recommendations was submitted to Mariemont on March 10 (Purtill 2000). The focus of these recommendations were on archaeological excavation of features located in areas to be additionally impacted. This included removal of the dog burial (in case it was prehistoric in origin) because it was believed that its shallow depth and fragile condition would cause the remains to be "crushed" during concrete pouring.

On March 18, a City Council meeting was held to discuss the Gray & Pape report and recommendations, ask questions, and to determine future actions. Mariemont City Council, W. Kevin Pape and Mr. Purtill of Gray & Pape, the architect and contractors, and several members of the MPF including project director Christina Beatty were in attendance. At this meeting, Dennis Malone of Malone & Wilcox (project contractor), presented an alternative pavilion construction plan that would eliminate the need for additional pier and grade beam excavation/construction. The new plan proposed a floating matte foundation using the already excavated foundation without additional excavations. The matte foundation plan proposed laying a 26-inch thick poured-concrete foundation to which the pavilion structure would be attached. Malone reported that this foundation would equally distribute weight across the entire floor at an estimated 380 pounds per square inch.

It was Gray & Pape's position that the use of the matte foundation would effectively eliminate additional damage to archaeological deposits in the foundation block. In addition, the concrete foundation was thought to be an effective preservation technique by sealing the now mapped deposits from unwanted disturbances. The only original recommendation still forwarded by Gray & Pape was the removal of the dog burial (F-00-12) for fear that its shallow depth (1 inch below pavilion floor) would cause the remains to be crushed under the concrete weight.

Upon hearing arguments from all parties, City Council voted for continuation of the pavilion project (with new matte foundation construction) by a count of four in favor, one against, and one abstaining. City Council and the MPF rejected recommendations for the removal of the dog burial, instead deciding to pursue in situ preservation. A plan was developed to place a thick foam board over the reburied dog location. It was the belief of Dennis Malone that this procedure would effectively redistribute the weight of concrete foundation away from the dog remains. He stated that such design techniques are commonly (and effectively) used in similar situations (e.g., protection of buried pipes). Construction of the pavilion proceeded soon after this hearing. Concrete was poured in early spring and the pavilion was completed soon after Memorial Day 2000.


The planning and implementation of the John P. Nolen Pavilion at the Madisonville Site unfortunately has to be viewed as a paradoxical set of circumstances. The vision of the pavilion was as a commemorative symbol of the significance of the Madisonville Site and our prehistoric past. The principals of the project, however, did a poor job at considering the potential impacts on the very resource they sought to memorialize. Although the lack of early planning (and its results) should concern all archaeologists/preservationalists, I believe that several favorable results eventually transpired that will better guide future projects. Perhaps most importantly is the fact that although Mariemont and the MPF failed to realize the detrimental impact of their campaign at the projects start, they now have a better understanding of the site and how to manage it as a cultural resource. Of note is the fact that Mariemont's Mayor Lemon took it upon himself to stop additional excavations (adjusting construction schedules), quarantine the site from collectors (complete with regular police checks), hire a consulting firm to provide recommendations (when not required to do so), and bring recommendations before City Council to make an informed decision.

From an archaeological perspective, the limited assessment also provided some valuable information regarding the poorly documented southwestern site area. This area was among the earliest to be excavated by Charles Metz between 1879 and 1880 (Hooten 1920; Drooker 1997; see Figure 1). Unfortunately, only limited notes were maintained by Metz's crew, most of them appearing to have been created out of the field (Drooker 1997). The highly inaccurate site map (Hooten 1920), created some 30 years after Metz's excavations, contains limited spatial information for these initial excavations simply stating that this area yielded "328 burials and 160 cache pits" (Hooten 1920). The scaled map generated during these investigations provides some of the most detailed spatial information currently available for this area of the site. It also suggests some revisions to Drooker's (1997) map showing site boundaries and the location of previous excavations. Taken collectively, the following conclusions are forwarded:

1) The fact that the pavilion block contains numerous pit features, possible post molds, artifact midden, a possible prehistoric dog burial, and artifacts indicates that this area is not outside of the site boundary as suggested by Drooker's (1997) site map. Instead, it is likely that the entire ridge top contains archaeological deposits and should be considered part of the site. Proposed revisions to the site boundaries are illustrated in Figure 1;

2) Several features (most notably F-00-1 and 00-2) appear to have been excavated in the distant past, perhaps representing Metz's 1879-1880 excavations. Although this point is speculative at this time, it may be that Metz excavated along the entire ridge top instead of being restricted to the central and western sections as suggested by Drooker (1997; Figure 1). If true, then Drooker's boundaries of Metz's 1879-1880 excavations need revision;

3) Signs of recent (20-60 years?) looting of the area was evident as 45% of the features identified contained loose matrix (appearing excavated) and two possible looters trenches were identified. At least one artifact (beer/pop cap) indicates disturbance dating to post 1969; and,

4) It appears that at least some of the features (most notably F-00-9/9a and 11) are undisturbed (never excavated). This fact is important because it indicates that even in areas dominated by previously disturbed features (either through earlier excavations or looting), there still remains a potential for intact features which could provide invaluable information.

As of this writing, Mariemont and MPF have several additional renovation/construction projects planned for the Madisonville Site. Importantly, Mariemont has included Gray & Pape in the planning (and re-planning) of these activities to minimize potential impacts. I hope to keep all informed of these activities with future updates.

References Cited

Allen, Howard
 1999  Mariemont Honors Future Museum Site: Ceremony Marks Arbor Day, Local Heritage. Article listed at Enquirer Local News Coverage, Enquirer.com, April 30, 1999.

Drooker, Penelope B.
 1997  The View from Madisonville: Protohistoric Western Fort Ancient Interaction Patterns. Memoirs of The Museum of Anthropology, University of Michigan, Number 31. Ann Arbor.

Hooten, Earnest A.
 1920  Indian Village Site and Cemetery Near Madisonville, Ohio. Papers of the Peabody Museum of American Archaeology and Ethnology, Harvard University 8(1). Cambridge, Mass.

Lemon, C. Michael
 2000  Mayors Annual Report for 1999. Report listed at www.mariemont.oh.us/mayorreport99.htm, June 1, 2000.

Madisonville Preservation Foundation
 2000  The 75th Anniversary Project Continues. Report listed at www.mariemont.oh.us/MPF/mpf_thanks.ht, May 11, 2000

Purtill, Matthew P.
 2000  Recent Excavations at the Madisonville Village and Cemetery Site, Mariemont, Ohio. Unpublished report prepared for the Village of Mariemont by Gray & Pape, Inc.