Archaeological Investigations at Solitude

The historic property of Solitude received National Register of Historic Places status in May 1989, without the need for archaeological investigations. However, a log outbuilding to the northeast of the main house had been scheduled for restorations to begin that same year.* The work was intended to be carefully executed so as to preserve or restore all the original properties of the building. One of the problems confronted in achieving this goal was that little was known about the building. Its original function was unknown. Some suggested it might have been a slave dwelling, washhouse, or kitchen. Learning the age of the structure, as well as the original function would be important for an accurate restoration. Gibson Worsham's Historic Structure Report on the outbuilding, completed in 1988, revealed some important information. Worsham was able to establish an approximate date of 1870 for its construction, based on archival research and architectural clues. Archaeological evidence supports this date, based on the analysis of almost 10,000 artifacts collected from test units around the outbuilding.

The excavations at the Solitude outbuilding, carried out by Dr. Cliff Boyd and his students from Radford University, consisted of thirty-one 1m² test units, strategically located around the building, using a grid system. Units were placed abruptly adjacent to the side of the structure in hopes of revealing a builder's trench. Artifacts from builder's trenches can confirm the time period of construction because they are deposited and buried during the construction process, often by the builders themselves. Shallow trenches were encountered, and did indeed produce important artifacts. Test units were also concentrated around the "back" door of the building. A common method of household garbage disposal in the 19th century was simply to sweep or throw it out the back door. This is referred to by archaeologists as the Brunswick Disposal Pattern. As expected, this location was especially rich with artifacts. Those that were found strongly supported the notion that the building originally served as a kitchen. It was common long ago to prepare food outside of the main house in the summer, to avoid heating the house. Plate fragments, drinking glass and bottle glass fragments, and animal bones were recovered in large quantities from behind the building. Ceramic types included porcelain, redwares, whitewares, stonewares, yellow wares, and pearlware. Analysis of the ceramics provided a mean date of mid 1860s (Figure A). Architectural artifacts such as foundation stones, brick fragments and nails also made up a large percentage of the artifact assemblage.

Also recovered from the test units were 341 prehistoric artifacts, of which 325 were lithic or stone flakes (byproducts of stone tool manufacture). Five pottery sherds and eleven projectile points were also found. Collectively, these prehistoric artifacts indicate that the site was occupied by more than one group of Native Americans between 2000 BC and 1000 AD, as a rough estimate. Unfortunately, due most likely to landscaping, there was no intact stratigraphy around the building. This has somewhat limited the chronological data that could be obtained through the excavations thus far. Therefore, more investigations in other areas around Solitude are imperative in order to recover any remaining data about the property as a whole, before it is lost forever. Lack of funding has been a continuous setback in this endeavor.

Artifact Analysis

Careful analysis of the artifact assemblage recovered from the excavations was necessary in order to help provide answers to specific questions about the structure. The first step of analysis consisted of sorting and classifying each artifact according to its function. For example, nails and window glass were assigned to the "architectural group," while glassware and ceramic vessel fragments were assigned to the "kitchen group." Stanley South, a highly regarded American archaeologist, developed this classification system. His system was well suited for this project. It defines nine basic functional groups ( Figure A). Within each of these groups the artifacts can be further sorted and classified. For instance, of over 100 existing types of historic ceramics, 8 different types were recovered here. Within each type of ceramic, we can often classify further, based on characteristics such as decoration, type of glaze, or properties of the paste. Ceramics are useful for establishing a timeline for the history of the building, since the period of manufacture of specific type can be determined. We know, for example, that "shell edge" whiteware, found at Solitude, was only manufactured from 1815-1900.

After all 658 ceramic fragments found at the site were grouped accordingly, a mean or average date for all the ceramics was calculated. A mean date was also calculated for the ceramics per each level excavated (Figure A). This process made it possible to determine that there was little "statigraphic integrity" in the areas that were excavated. In other words, the fact that newer types of ceramics had been encountered in the ground often mixed with or even below older types, probably means that the archaeologists did not find them in the same order as they were deposited. This may be due to disturbances such as those incurred during landscaping.

Next, the total number of ceramic fragments per test unit was tabulated and plotted on a site map, in order to allow the distribution to be more easily studied (Figure B). The map depicts the excavated units in relation to the outbuilding, and the proximity of doors and windows to each of the units. It is interesting to note that 126 ceramic fragments were recovered from a single unit, just outside the back door of the building. This makes up 19.1% of the total ceramics found. In fact, 54.3% of the fragments were found in close proximity to the back door, in units 7, 8, 9, 10, and 16.

The distributions of each of the other eight functional groups were plotted in like fashion, on eight separate site maps.

Ceramic Distribution by Type

Type of
Number of
Fragments Recovered
Percentage of
Total Recovered
243 37%
153 23%
Redware 96 15%
Stoneware 73 11%
Porcelain 68 10%
Yellowware 13 2.0%
Earthenware 4 0.5%
Pearlware 3 0.5%