Methodology: A Documentary Archaeology

In recent years, research and teaching in the humanities has experienced a striking turn toward the material. The desire to study tangible things follows naturally upon an era of scholarship that has been oriented largely around the study of words and categories of thought, but it also tracks the changing preoccupations and concerns of ordinary people, for everyone who has lived through the last half-century has been intensely aware of the dramatic shifts in the profile of consumption in the United States, Europe, and in the developed world more generally. 

DALME’s intellectual underpinnings are profoundly influenced by this interest in things. We believe that humanity cannot be understood apart from the material surroundings in which all people and their communities are embedded. The study of material culture is thus vital to gain insight into otherwise undetectable aspects of human society.

The spatial, temporal, and thematic focus of the project centers on late medieval Europe, a time span extending from the mid-thirteenth to the early-sixteenth century. This period merits our attention because it lies on the front edge of a documentary revolution whose vestiges offer piercing insight into humanity’s deeply structural relationship with our material surroundings.

Up until ca. 1250, archaeological finds provide the best resource for studying the material culture of the region. Starting at around that time, however, the arrival of paper as a writing support in southern Europe, coupled with a rapid expansion in legal and administrative bureaucracy across the continent, created the conditions for the survival of copious records pertaining to the material aspects of households and life-ways of a broad spectrum of the population.

The chosen primary source for DALME are inventories. This focus is not new: archaeologists, art historians, and other students of material culture have long relied on the genre for the study of objects and materials. Inventories are particularly valuable sources for comparative analysis because the systematic biases that generate silences in the textual record are often quite different from the systematic biases that affect the survival of tangible things. Indeed, the comparison of the evidence from each of the two domains will make the systematic biases inherent in each readily apparent and subject to quantitative analysis.

The number of inventories available after ca. 1250 is orders of magnitude greater than that from earlier periods in European history. Furthermore, the texts and object lists that survive from the early medieval world are scattered, unsystematic, and skewed toward social élites. They lend themselves well to anecdotal discussion but less so to serial analysis. With this new and diverse source base, the full array of material culture in Europe emerges from the evidentiary shadows. These records are valuable not, as some have argued, because they shed light on the material conditions that led to Europe’s global economic hegemony in the early modern era and into the twentieth century. They are valuable because they constitute important textual evidence for understanding the deep interdependence of humans and things. 

DALME aims to make inventories and other lists of objects available online as machine-actionable datasets in a manner that makes comparison possible between objects described in widely varying ways in documents, as well as with museum objects and excavated artefacts. 

The problem is that the manner in which a collection is described is not independent of the historical conditions that brought about the collection in the first place. 

Our approach is wholly aligned with the intellectual goals, general methodologies, and objects of study characteristic of the discipline of archaeology. The only difference is that DALME does not rely, for its primary source of evidence, on descriptions of artifacts and objects generated by archaeologists and museum curators. Instead, we rely on descriptions of things generated by contemporaries in the act of reflecting on their own material world.

Indeed, our goal is to lay the theoretical and methodological foundations of a novel approach for the study of past material culture. Our kind of documentary archaeology approaches things, as described in extant documents, as traces of the attributes and relational networks of objects that once existed, and aims to define a conceptual framework that would make it possible to draw legitimate comparisons between these textual things on the one hand, and museum objects and excavated artifacts on the other. 

Our goal is to make it possible for researchers to interrogate substantial datasets of material culture—archaeological artifacts, museum objects, and textual things—all of which are necessarily represented through the medium of language. The potential of such combined datasets is enormous. Their scale alone opens new research possibilities by making sophisticated, large-scale computational analyses feasible.

The methodology pioneered by DALME is based on the realization that, when analyzing or deploying tangible things as evidence, we are not working with the objects themselves, but rather with linguistic surrogates: that is to say, with descriptions, more or less structured, of the actual objects. These descriptions are the result of a process of encoding whereby an individual “identifies” an object and its attributes by classifying them according to a pre-defined system of rules that articulate how classes of things relate to each other and how membership in them is determined. 

In the case of an excavated artifact or a museum object, the system of classes and rules that determines an object’s linguistic surrogate is a complex domain ontology. This ontology, whether explicit or not, results from the complex interaction of a number of factors, such as professional practices, methodological approaches, and theoretical considerations, and is articulated through expert taxonomies and controlled vocabularies. Textual things are no different. Our research has shown that, within reasonably constrained temporal, spatial, and linguistic scopes, people tend to describe material culture in very similar ways. The range of vocabulary used to describe material culture by notaries in late medieval Marseille, for example, is both remarkably consistent and circumscribed, as are the grammatical and linguistic structures used to express or imply their relational configuration. This is the result of a similar system of classes and rules, a folk taxonomy, that determines the outcome of the encoding process. For our purposes then, a thing, whether textual or tangible is always encoded into a description.

The approach we have developed relies on the unique capabilities afforded by the systematic analysis of large datasets to tease out the structure of these folk taxonomies. By articulating the principles of this underlying grammar, a mere description can be decoded from its original taxonomical schema and re-encoded into a different, common ontology, alongside both textual objects from different contexts and tangible things from archaeological and museological collections. The ultimate goal of a documentary archaeology, then, is to provide a framework that translates between the modern domain ontologies used by scholars to characterize museum objects and archaeological artefacts and the historical folk taxonomies used to describe objects by people in the past.