I really enjoyed Kevin Dicus’s recent article in AJA 126.4: “Refuse and the Roman City: Determining the Formation Processes of Refuse Assemblages Using Statistical Measures of Heterogeneity.” This article pushed me to read Guido Furlan’s 2017 article in EJA 20.2: “When Absence Means Things Are Going Well: Waste Disposal in Roman Towns and its Impact on the Record as Observed in Aquileia.” Both articles dealt with the tricky issue of trash disposal in Roman cities and, more importantly, the character of secondary deposits in an urban setting.
Dicus’s article uses two statistical measurements to provide a baseline for peri-urban dump sites: Shannon Diversity and Pielou J Evenness. The former measures the range of types of objects that would appear in an assemblage or sample and the latter measures the distribution of material across the various classes present in the assemblage or sample. Excavated dump sites have proven to have high levels of Diversity and Evenness. For Dicus, this reflects the character of peri-urban dump sites as the end point of city wide refuse collection. This formation process brings together a wide range of discard practices in such a way to produce diverse and evenly distributed assemblages in peri-urban dumps.
Dicus compares the character of peri-urban dumps to material found in various fill context inside Pompeii. For example, domestic refuse tended, in Dicus’s study, to have lower diversity than that from dumps outside the city. This is presumably because it reflects a smaller range of activities (and perhaps developed over a shorter period of time). Curiously, other fills from around Pompeii show levels of diversity and evenness that are not all that different from the municipal dump sites. This suggests that despite their association with domestic spaces and their location within city walls, these fills nevertheless required material from assemblages produced by the more expansive formation process similar to those that formed municipal peri-urban dumps. Dicus’s willingness to unpack and explain the quantitative character of these fills and dumps is really commendable. He not only recognizes the limits of this kind of approach, but also explains it in a way that is both clear and replicable at other sites that have well-excavated dumps and fill contexts that can be compared.
Furlan’s article is less sophisticated quantitatively and focuses on aoristic analysis of the chronology of fills and dumps. He argues that fill levels in urban sites are often far too variable and dependent on the formation process that created the fills to serve as proxies for the economic situation in a particular community. Parallel to Dicus’s later work, Furlan suggests that municipal dump sites which capture a wider range of formation processes from, presumably, a spatially and functionally more expansive catchment would represent more fully the economic life of a city.
This work is relevant to my recent work at Polis. The area of EF2 in the Princeton grid at the site included a massive fill deposits that presumably served to facilitate the leveling and drainage of sloping ravine. The presence of abundant cobbles in the fill suggests that it combined building material. Its location at the edge of the city also suggest a proximity to a per-urban dump. In fact, a large mound of slag to the north of the site suggests a dump from nearby industrial activities. There is reason to expect, then, that the fill incorporates aspects of industrial discard in its assemblage and the percentage of amphora sherds, for example, is suggestive. The significant quantity of fine and tables wares as well as cooking pots that we should probably associate with domestic use, however, indicates that the fill also draws upon a wider range of processes perhaps including those associated with both urban discard and disposal.
Unfortunately, the material that we have available for the study of the fill level at EF2 is a sample of the material excavated in the 1990s. We expect that the excavators discarded a good amount of the course and utility wares that appeared as undiagnostic and this means these kinds of vessels are underrepresented in the assemblage. Or, conversely, table and fine wares are proportionally overrepresented in the assemblage that we studied. In other words, we can’t do the kind of thoughtful analysis performed by Furlan or Dicus. Despite these limitations, their work does give me something to think with when exploring the legacy data produced by the excavations along the northern side of the city of Arsinoë in the village of Polis.