Traditional approaches to the description of historical materials focus on indexing named entities associated with the documents and objects. Users often look in archival records for materials by or about individuals and organizations, and place names are also widely used as indexing terms, as many users may use them in conjunction with names of individuals and organizations to narrow searches for information.
Another type of named entity that can be useful is the event. Users researching the history of an incident or occurrence of a phenomenon or cluster of related phenomena will use the name of that event as a search time. Events are special forms of named entities, as they serve as a nexus point that marks a relationship between agents, places, and particular points in time (Gracy 2015). Thus, they act as gathering mechanisms for records of actions and are crucial aspects of archival information systems. According to Hyvönen, Lindquist, Törnroos, and Mäkelä (2012) events “link actors, places, times, objects, and other events into larger narrative structures, providing a rich basis for semantic searching, recommending, analysis and visualization of CH [cultural heritage] data.”
Events can be defined on a micro or macro level. At the micro level, events consist of individual occurrences or activities of people and systems triggered by functional requirements of an organization or governing body, from which systematic recordkeeping systems emerge. Macro-level events, on the other hand—which may consist of multiple smaller-scale discrete events—consolidate those smaller events into a pattern of historical or cultural significance. At both the micro- and the macro-level, one can find descriptions of events in archival records and finding aids for those records; it is assumed that users may be interested in both types of events when searching for information.
Macro-level events with broad scope, involving many actors, places, and objects and stretching out over years, decades, or centuries, present significant challenges for archival information systems. These systems must be capable of aggregating heterogeneous information from various sources and collections. Few standardized vocabulary lists of event entities exist to serve as connectors among these data sources. Archival finding aids often do not even provide sufficient controlled access points to events. This deficiency may be traced to reliance on vocabularies that lack specificity needed for description of such topics in archival collections, such the Library of Congress Subject Headings.
This research project aims to use an archival collection from the Kent State University Libraries as a primary data source: the May 4 Collection, which is a collection of historical materials about the events surrounding May 4, 1970, when thirteen unarmed students were shot by members of the Ohio National Guard at a student demonstration against the Cambodian Campaign (part of military operations during the Vietnam War). The May 4 Events hold great historical significance in the history of Kent State University and the United States and have been studied extensively by historians to ascertain the sequence of actions on May 4 itself and trace precursor and subsequent events that relate to the tragedy on that day. The May 4 Collection at the KSU library is rich in event-related information and represents the largest accumulation of archival evidence on the subject.
Our initial research goals are:
- Using the Kent State May 4 Collection and other historical collections (to be identified) as data sources, test the usefulness of archival finding aids and certain archival materials to create historical event thesauri; these thesauri would provide indexing and semantic definition (via Linked Data) of such material at a greater depth than is currently possible with existing thesauri, vocabularies, and indexing systems.
- Using the Kent State May 4 Collection, create and test an event model encompassing spatio-temporal dimensions and agents associated with the event that could be used to aid in linking historical documents and archival descriptions to other relevant published and archival sources using Linked Data approaches.
Upon completion of these initial activities, the research team will expand our testing of the event model with other historical collections and relevant data sources, and eventually develop a software tool to assist historians and cultural heritage scholars in building and testing hypothetical narratives based on the linking of micro-events. Such a tool would provide direct link to related historical material and help historians compare alternate sequences of events where accounts conflict with one another.
Karen F. Gracy is an associate professor at the School of Information of Kent State University. She possesses an MLIS and PhD in Library and Information Science from the University of California, Los Angeles and an MA in critical studies of Film and Television from UCLA. Recent publications have appeared in Library and Information History, JASIST, Archival Science, American Archivist, Journal of Library Metadata, and Information and Culture. Dr. Gracy’s scholarly interests are found within the domain of cultural heritage stewardship, which encompasses a broad range of activities such as preservation and conservation processes and practices, digital curation activities that consider the roles of heritage professionals and users in the lifecycle of objects and records, as well as knowledge representation activities such as definitions of knowledge domains, development of standards for description, and application of new technologies to improve access to cultural heritage objects.