It would be difficult for most people to think of broken glass as anything more than garbage. While it might be easy to understand why archaeologists would recover and analyze glass artifacts such as medicine bottles or beads, what can small shards of common, undecorated window glass tell us about the past?
During excavations at Riverside, 740 shards of window glass were collected during fieldwork. Once back in the lab, the thickness of each piece was measured and recorded. Although it may not sound like the most exciting of jobs, it turns out that this seemingly menial task provided important insight into the dates of occupation at our site!
During the early 19th century, the cylinder glass manufacturing technique became the primary method of producing window glass in America. This is important because it allows archaeologists to assume a standardized production technique across sites (although there will undoubtedly be variations!), making analyses at one site comparable to the other. Throughout the 19th century, window glass became increasingly thicker due to the increasing popularity of larger windows, which required thicker glass for support. Historical archaeologists have been able to use this gradual increase to correlate glass thickness with a historic date range. A good overview of why this technique works, as well as the different types of analyses can be found in Jonathan Weiland’s article, “A Comparison and Review of Window Glass Analysis Approaches in Historical Archaeology.”
Weiland’s article also points out that there are many different methods for analyzing window glass. At GRA, we utilized Moir’s method for a number of reasons. Not only did it provided a good way to sort out window glass from the total amount of glass shards found at Riverside, but it also specifically covered both the time period and regional area of the site. More importantly, the analysis can produce reliable results with only a small amount of material, saving us both time and money, without compromising our data quality!
This graph illustrates the results which are consistent with the approximate date of dumping episodes that occurred at the Riverside property. Dumping episodes are common occurrences in the archaeology of New York City. They are a result of urban growth and expansion in the 18th and especially 19th centuries, when the landscape of the island was considerably altered for building purposes. These alterations included dumping household garbage including window glass into marshes and ponds to create additional building space.
The formation of the Riverside site makes it difficult to assign exact dates to the dumping episodes found during archaeological excavation. In this case, our best bet is to define a range of time and use a variety of evidence to refine this range as much as possible. Here the glass analysis provided additional evidence supporting our hypothesized time range, which suggests that we had not overlooked a key time period in the site’s formation.
Have you worked on glass analysis from other historic sites? Do you have questions about this technique? Let us know in the comments!