Cataloging Oral History Collections, Or Why Archival Work Will Never Be Featured in a Mountain Dew Commercial
In this blog post, I write about something very important to me and my sense of professional identity. Many people think archival work consists of digging around in boxes scented with the musk of ages, reverently holding yellowed, centuries old documents in white gloved hands, enveloped in silence only punctuated by the whispers of similarly dutiful colleagues and researchers who feel more at home with the dead than with the living.
If that described my daily life as an archivist, I would run screaming from the university archives.
I’m going to blog today about cataloging and metadata.
There’s a librarian joke that I suspect is seriously unfunny outside of the profession:
Standards are like toothbrushes; everyone agrees we should use one, but no one wants to use anyone else’s.
To continue the toothbrush simile, the reason no one wants to use another librarian’s — or archivist’s — standards isn’t because we’re afraid of germs. It’s because we all want a standard to meet our perceived needs. “I want it to really clean my molars, but it should also be soft on my sensitive gums!” At this point, the GLAM (Galleries, Libraries, Archives and Museums) sector has so many different kinds of toothbrushes/standards, some clever folks were able to devise this nightmare-inducing metadata chart. I consider myself moderately metadata savvy, but there are metadata standards in this chart I’ve never heard of and I don’t feel bad about that.
Last year, we New School archivists began cataloging our legacy (archives jargon for stuff your predecessors started and then left for you to deal with) oral history collections. Not being super familiar with creating records for this type of historical resource, we decided that we all needed to be on the same page. Although different staff members present and future might be assigned to describe different groupings of oral history recordings, researchers should expect to see the same categories of information for each collection and see that information expressed consistently across collections, regardless of the archivist cataloging it. This is basically the point of standards. Additionally, these standards should be applicable in as many information environments as possible. There are plenty of different external standards to choose from as evidenced by the aforementioned chart. Which ones should we pick?
Our resultant oral history cataloging document combines three basic external-to-the-New School Archives standards, none of which cancel the others out. After conducting a considerable amount of background research and learning what other librarians and archivists have been doing, we selected Describing Archives: A Content Standard (DACS), PBCore, and oral history-specific recommendations.
DACS (www.archivists.org/governance/standards/dacs.asp) is a content standard for creating finding aids, those peculiar documents that archivists write to inform potential researchers about collections, and which researchers sometimes read. Here at the New School Archives, we refer to them on our website as “collection guides.” I find DACS a bit nebulous, and still heavily geared toward text-based records and conventional manuscript collections, but there are definite improvements in the latest edition, which you can download for free. There’s a long history of archivists neglecting and short-changing audio-visual records, which we (as in profession and society) are now collectively paying for, but that’s another post for another day (lots of folks have already written extensively on this matter, but my favorite is the AVPreserve blog).
To supplement DACS’s suggestions for what to include in the collection guide and in which part of the guide, I turned to Nancy Mackay’s Curating Oral Histories from Interview to Archive. (Walnut Creek, CA: Left Coast Press, 2007) and Elinor Mazé’s, “Metadata: Best Practices for Oral History Access and Preservation,” (2012. Oral History in the Digital Age. Institute of Museum and Library Services. Retrieved February 5, 2014.) By cross-referencing these texts with DACS, I was able to assign information specific to oral history recordings to the most appropriate field in DACS. For example, in describing oral histories, researchers should know where the interview was conducted. An interview conducted in the comfort of an interviewee’s home may be interpreted differently than one conducted in the interviewer’s office. Additionally, oral histories are sometimes conducted by archivists, which is the case with many of the New School Archives’ recordings. It is important to document how interviewees were selected for inclusion in an oral history program (usually denotes an ongoing endeavor) or project (finite duration).
Transparency is the main principle underlying these descriptive choices. The oral history cataloging document renders transparency a mandate, rather than an individual archivist’s whim. Many folks who engage in oral history production are keenly aware of power and privilege in the archive. It’s often why they’ve taken up microphone and recorder in the first place. Overlaying oral historians’ political concerns on top of DACS takes us further than relying on DACS alone.
The image below is a screenshot of our collection management and collection guide publishing software, Archivists’ Toolkit, displaying a record for an oral history collection.
The New School Archives doesn’t depend on the collection guides alone to provide access to records. We also have a Digital Collections site where researchers can directly access some of the resources described in the collection guides. The metadata standard we use for recording information about the audiovisual materials we upload to our Digital Collections site is called PBCore.
PBCore http://pbcore.org is a metadata standard developed specifically for audiovisual materials. Although the “PB” in the name stands for “Public Broadcasting,” this standard outgrew its initial implementation and has been adopted by archivists outside of that specific community.
I like PBCore because it adheres to this very librarian-y concept called Functional Requirements for Bibliographic Records, or FRBR. It is among the more complex topics librarians (and now archivists belatedly joining the conversation) discuss when they’re alone together. The Library of Congress offers this very handy FRBR document.
I won’t go into it too deeply here, but the oversimplified point is that we can differentiate an idea from the container holding the idea. This is very important for oral histories because we often end up with a number of containers holding the same idea.
For example, in my test collection, I described a set of oral histories that were recorded on audiocassettes (idea container #1) in 1994. A vendor, SafeSound, digitized the audiocassettes and saved the resultant preservation sound files as BWAV (idea container #2) in 2010, then created access files that we could offer on our website (idea container #3). Finally, we sent those access files to Mim our trusty transcriptionist, who created text-based, digital files using Microsoft Word, which we then saved as PDFs (idea container #4). All of these idea containers document the same event — 1994 oral history interview — but each container also has a different date of creation and other information about the format that we need to document for our own preservation purposes. I incorporated as much of this information into the collection guide, matching it with the most relevant DACS field, and recorded it in our collection management software, Archivists’ Toolkit.
Using PBCore led to the discovery of a “hidden” interview on one of the audiocassettes. When I couldn’t get the digital files (idea containers #2-3) to match up with the transcripts (idea container #4), I realized our initial assumptions regarding what was on each audiocassette was wrong. We had an extra idea in those idea containers!
The New School Archives uses another type of software, CollectiveAccess, to manage our Digital Collections site. This is the software that allows researchers to find and listen to the digital files via the Web, and our pre-set “profile” for audio and visual materials is modeled on PBCore. The way we’ve structured our oral history cataloging document, all of the information that needs to go in CollectiveAccess is already produced within Archivists’ Toolkit during the creation of finding aids. No new information needs to be created.
Below is a screenshot of the same oral history collection as it appears in CollectiveAccess:
It is our hope to use this oral history cataloging document as early as possible in the life of an oral history collection, rather than allowing backlogs to build up. The primary drawback of this workflow is that it can be initially time intensive. However, because all of the descriptive tasks are frontloaded, it’s easy to get everything online and findable by researchers once the initial work is completed. As archivists, we just (just!) need to engage in digital preservation activities. Some archivists have argued against item-level description of audiovisual materials, but from my perspective, the effort pays for itself in discovery, research use, and long-term preservation. I guess one downside is that future New School archivists will lose out on obtaining future “Cataloging Hidden Collection” grants…
Actually, come to think of it, I can return to my toothbrush analogy. This cataloging workflow is a little like brushing your teeth; if you make it a part of your routine early in your life, you may be able to prevent large-scale dental disasters in the future. Brushing becomes something one does as a matter of course. Nothing exciting, romantic, or particularly mystic about it. And I’m totally okay with that.