An Archaeology of Word-Things in Later Medieval Mediterranean Europe

One of the principles of a natural scientific approach is that things carry traces of information about something other than what they are about. Chemical isotopes and genomes carry trace information about the organisms that made them: what they ate, where they lived, where they traveled, and so on. The common understanding here is that you can always squeeze more information out of a given object of analysis, especially when quantitative methods can be brought to bear. The object of analysis itself can be anything, and this includes documents. Historians and philologists have long been aware that documents can be read not only for their intended meanings but also for their incidental or unintended trace information. As computational history has shown, for example, patterns of word choice and syntax carry distinctive signatures not unlike the signatures left in isotopes and DNA.

This paper explores some of the information that can be wrung from the textual sources extant from late medieval Mediterranean Europe. I assume that archaeological  and computational methods can be brought to bear on word-things, that is to say the objects of material culture found in archival strata. My sources include household inventories and inventories of debt recovery, a source base containing several tens of thousands of objects, far beyond anyone’s capacity to read and understand. I will demonstrate how an “index of descriptness” can be applied to a range of discrete objects, such as items of clothing, furniture, linens, ceramics, and metal wares, via a statistical approach to the words used to modify nouns. The purpose of the exercise is to identify a quantitative measure of ascribed symbolic value, and to use that measure to demonstrate how the symbolic value of material objects varied over time and space.