Access to information about museum collections is especially important in university museums because of the academic nature of museum's missions. This paper proposes the use of a ubiquitous tool to provide an additional access point to information about museum collections: the Library of Congress Subject Heading (LCSH). By integrating the museum object with the library's monograph collections, users gain access to great amounts of potentially relevant information. This paper examines a case study conducted at the Spurlock Museum at the University of Illinois Urbana-Champaign in which LCSH were attached to object records in the museum's database. It was found that LCSH was not an efficient tool for the project, therefore I examine the promise of a new tool to do the same work: the LC Class Web.
The Spurlock Museum at the University of Illinois has recently opened a new facility having been temporarily housed in attics and basements, broom-closets and barns for nearly a century. Its collection of 50,000 objects is diverse, holding ethnographic and archaeological collections from across time and around the world.
Since opening in September 2002, one of the most resounding queries facing front-of-house staff is one requesting more information on specific artifacts. However, because the museum has only recently acquired a professional staff and has, in the past, had a very catholic collecting strategy, information about items is often not readily available or organized consistently, if it is known at all.
I began looking into methods in which visitor queries could be answered by sending interested parties to the library to find the information on their own. By linking Library of Congress Subject Headings (LCSH) this project is intended to facilitate library use by museum audiences, enabling interested parties to incorporate material culture items into traditional research and hopefully include the museum in the process.
This paper discusses a case study in which the artifacts in a gallery at the Spurlock Museum at the University of Illinois are given LCSH access points and therefore a point of interaction with the University Library system and a point for researchers to further search for information about the objects in the Museum's collection.
Libraries have been very adept at adopting standards for the classification and cataloguing of books. These standards, often adopted internationally, allow for the great deal of variation present in libraries and their holdings. The Library of Congress has developed a cataloguing system that supports flexibility in collecting in vastly different institutions. LCSH allows researching at the patron's level through authority-based subject searching.
In this project, I do not propose that museum objects be catalogued using library standards. Although it has been shown that library organizational schemes can work for museum cataloguing, registrars are often reluctant to adopt them. (Beirbaum, 1990) I am proposing that LCSHs be linked to objects through the object database in order for museum objects to be used as a tool for access into library catalogues.
The library is one of the most used tools on campus. There can be little excuse for students who do not know how to use the library and do not do so regularly. By using traditional library tools for the integration of material culture into traditional academic research we can hopefully encourage more researchers to use objects in their research. As students enter university having used computers for the entirety of their education, they often demand that information be immediately available to them. This is no different for information about the objects in museums.
By offering a point of entry into the library, we can perhaps encourage researchers to look beyond the very little knowledge and bibliographic information about collections like the Spurlock Museum's collection. This condition, caused by years of understaffing and neglect of the collections catalogue can be addressed if we can provide ready access to information and thus facilitate active research by people outside of the museum.
The primary impetus for this project is the Spurlock Museum does not have an active research program. There is neither staff dedicated to researching collections or encouraged accessibility for external research. If the Museum is to distinguish itself as a University Museum (as opposed to a museum at/in a university) a research program must be created. If this project can facilitate integration of the Museum's collections into undergraduate research, we can get quality research about our collections to integrate into our knowledge base.
In this instance, I feel it is key to focus our energies on facilitating undergraduate materials-based research. If we can convince the next generation of scholars, policy makers, and general public that university museums provide access to objects that are necessary for basic research, then the community may be sufficiently engaged to protect the university museum from obscurity.
Linking the Spurlock Museum's objects to the monographs in the Library directly benefits the campus community in many ways. Current University budget restrictions encourage innovative cross-campus collaborations like this one. Although in the past museums and libraries have often worked quite closely at the UIUC this collaboration has been traditionally hard-pressed. There is also a benefit to the Library as this may encourage library usage for those students who may be otherwise remiss to use the library. It also provides a link to museums for libraries increasingly engaging in interpretive programs that have traditionally been in the realm of museums.
Further to the Spurlock Museum's advantage, we can provide research resources without maintaining a separate library within the Museum. By relying on the University Library we don't have to hire a librarian to maintain a library on site. The library's 10 million collections are ample enough to cover our required knowledge base.
This study had two distinct objectives: to determine the feasibility of adding LCSH to 50,000 object records, and to determine the usefulness of the outcomes. It seemed obvious from the outset that it would take time to add this data to the object records, but I wanted to explore what the true cost of such a project would be and whether the benefits would make it worth doing (and whether, perhaps, the project might attract external funding).
The first step in the study was to familiarize myself with the current cataloguing and classification literature. Once a solid understanding was gained of how cataloguing works, we had to evaluate how objects are looked at to determine how we look at items.
This study originated with the idea that we would just add some LC Subject headings to our existing object records, enabling users to identify relevant topics in the library catalogue. We encountered three main problems. 1. Users can look at objects from many perspectives. 2. How broad of an entry into the library do we want? 3. Are LC Subject headings the wary to go, do they offer enough breadth?
We soon realized that researchers would be looking at objects in multiple ways. For example, this artillery shell [see figure 1] could be looked at as an example of artillery shells. Or as representative of modern armament and warfare. It could also be examined in provenance research or in regard to its donor. Materials and manufacturing could also be the focus of the research of this item. Our solution for this problem was to determine 10 ways/genres in which people could examine and research an item and to offer terms for subject searching in these regards.
Concepts to be covered by LCSH in database
After determining the data structure, it was necessary to modify the museum object database and to begin assembling the Subject Headings.
LC Subject headings are hierarchically arranged. The more terms you add, the more specific the heading. For example: our Artillery Shell. If one uses the LCSH "Artillery" we receive 271 entries from the University of Illinois Library and 21 from the local Urbana Free Library. If we use the term "Artillery - Usage - World War - 1914-1918" we get one entry from the University library and none from Urbana Free. Educators and Librarians often prefer a search with a large number of results. This broad source matrix offers students a way to structure their arguments and a larger pool from which to draw their research.
The major problem with LCSH is that one term, while it may be related to other terms, does not necessarily link to it. For example, someone searching for the term "artillery" may also be interested in the terms "armament" "warfare" "cannon", "projectile", and many other terms. The presents a problem with the current Library catalogue in which only one SH be searched for at a time. Therefore we can only add one SH per category per object. This is wholly inadequate in the way that SH is structured. Often more than one term would be appropriate for the same concept. Furthermore, LCSH is not compatible with the Dewey Decimal Classification system (another widely used library tool, particularly in primary school libraries).
I found that it took a lot of knowledge of both LCSH and the collections - a combination that no one currently has (in the Spurlock museum). As such, it took nearly 45 minutes per item to add the Subject Headings to the object files. This resulted in a total time of 45 labor-hours for the 60 items surveyed. While this is excessive, it should be noted that because of similarity in collections some of the subject headings used for these items can be expanded to roughly 15,000 items. For example, our artillery shell shares its headings with the other 12 shells from that same accession group. It shares the time period headings with the 227 items in the collection from WWI. Likewise our Roman fibula, shares its culture and time period subject headings with 1643 items in our collection.
Although these incidental connections would shorten the overall time for adding LCSH to the entire collection (with at least a few SH for each item), the problems with the system remain. These incidental connections may prove to be just as useful if using another system.
Despite the benefits that this linkage of objects to the library could potentially provide, the time/cost requirements of enacting this linkage and the inherent failures of LCSH mean that this project cannot progress out of concept stages. It will simply cost too much to add LCSHs to the collection of the Spurlock Museum. However, in the progress of this project, another, perhaps more appropriate system was discovered.
One theme throughout this conference has been a push for university museums to gain relevance to the university community as a whole. I think that this could be used as a tool to gaining relevance by allying the museum with a proven university mainstay, the university library. Also, by adding this component to museum objects, we are promoting the training of students to use material culture in their studies and promote hands on training. This inclusion of students in research and staffing plans promotes the training of our future colleagues and those who will carry our profession through this century.
I have begun looking at a tool that the Library of Congress is currently assembling: Class Web. This tool will enable users to search for a subject heading and find within which classification schedule that subject heading is most often found in. The system will also provide up to ten closely related subject headings that are also found in that classification schedule (with similar classification numbers). This will allow users to find and search for terms related to the term they have in mind through the classification number. This also enables easy use of libraries equipped with the DDC system because LCC can be related to DDC.
The Class Web's syndectic data structure can potentially provide horizontal, multi-tier access to information. However, the system is still in testing phases and still requires a point of entry. I am currently working with Professor Emerita Pauline Cochrane of the University of Illinois and her PhD classification class to try to find ways in which that first step can be made into the class web system, and then if the system is useful.
This requires consistent metadata in the museum object files. Location, culture, name, and other data should come from established thesauri and authorities. If consistent data is used in the classification of museum objects, it may then be possible to mechanically map that data into the class web system providing useable results. Otherwise, the process remains a manual one and the time/cost component may again become prohibitive.
Special thanks to Beth Watkins and Professor Pauline Cochrane for their insights into the relationships between libraries and museums. Special thanks to the Spurlock Museum for access to its collections for the purpose of this project.
Bierbaum, Esther Green, "MARC in Museums: Applicability of the Revised Visual Materials Format" in Information Technology and Libraries (ALA, Chicago; December 1990) pp. 291-299.