Weeds can be helpful: indirect evidence and archaeological analysis

Submitted by Sammy Smith (sammy@thesga.org)

Sheffield_dept_bannerArchaeologists often use indirect data to infer past cultural practices. This is because only certain data are preserved in archaeological contexts. Yet, we have questions that extend beyond that preserved data. Other types of data allow archaeologists to identify important information not directly available from the (somewhat limited) archaeological record.

For example, researchers at the University of Sheffield in England, have been interested in crop husbandry practices. This means they’re interested in what species were chosen to husband, or use, for food or other purposes. The information about the chosen species is often incomplete, so the researchers decided to look beyond direct data (e.g., seeds found stored in vessels in houses they excavated) to information they could get from associated weed species. Because the weed species were associated with the preferred species, they constitute indirect data.

These researchers found, not surprisingly, that the weed species at the archaeological sites they studied were most linked to ecological variation, especially productivity and disturbance. They note:

The range of attributes related to productivity indicate that both soil fertility and water availability play a part in this variation and that there is an interaction between productivity levels and the level of disturbance. Seasonality is a secondary factor relating primarily to water availability in arid environments and sowing time in more temperate regions.

Researchers say that as a result of this analysis they were able to infer that irrigation was used at an archaeological site where they had no direct evidence of it. At another site, they were able to identify sowing time and intensive cultivation, using the patterns of weed species, etc.

The paper is called “Crops and weeds: the role of weed functional ecology in the identification of crop husbandry methods,” and is by G. Jones, M. Charles, A. Bogaard, and J. Hodgson, all at the University of Sheffield, Department of Archaeology and Prehistory.

The paper was published on pages 70–77 in the Journal of Archaeological Science (2009). At present, the paper can be downloaded for free. Get to it by clicking here; it’s paper number 9.