It’s Complicated: Restrictions in University Archives

By Tracy Jackson

I process University Archives (UA) collections. Here at Duke University, UA collections are popular and frequently used by students, administrators, and community members alike. Processing these collections is satisfying work, making complex materials neat, organized, and accessible. Or at least accessible. But there is one common hitch in providing access to UA materials, one descriptive element that makes simple minimal processing more complicated: Conditions Governing Access, DACS element 4.1, required. The element that “provides information about access restrictions due to the nature of the information in the materials being described, such as those imposed by the donor, by the repository, or by statutory/regulatory requirements.”[1]

Restricted access to materials seems to come up quite a lot for UA records. In recent years, especially, researchers have wanted to explore the university’s complicated history, including incidents of racism and sexism, protests and demonstrations, labor conflicts, and community relationships. The records that document these topics are often the records that have the most access restrictions, making smooth access difficult.

University Archives boxes showing screening dates and stamps used to apply information related to restrictions. Image courtesy Tracy Jackson.

Part of the problem is the number of different restrictions that apply specifically to UA records. The big one is FERPA, the Family Educational Rights and Privacy Act, a law that governs the availability of student information during and after the student’s lifetime. Duke University also has two explicit restrictions governing access to its own records: Board of Trustees (BoT) records are closed for 50 years from date of creation, and Administrative records are closed for 25 years from date of creation.

In addition to these mandates, UA has noted four additional types of restricted materials for which we ideally want to screen: personnel records, donor information, medical information, and legal documents. While we are not a HIPAA-covered entity (the Duke University Medical Center is a separate entity and their records are held by a separate Medical Center Archives), UA, as part of the Rubenstein Rare Book & Manuscript Library, has decided to screen and protect medical information, as well as the other three types of information listed above, as an ethical practice.

Seven different types of restricted material that may commonly occur in UA records seems like a lot, and feels intimidating. While these seven are specific to Duke, I imagine many other college and university archives, particularly at private institutions, have a similar list.

With so many types of restrictions, restricted materials seem to pop up everywhere. As much as we wish they wouldn’t even come through our doors, they exist throughout collections we’ve already aquired, and can be difficult if not impossible to prevent coming in even with a robust records management program. Our (ideal) policy when screening for these types of records is to keep and restrict Administrative and BoT records, since they are documentation of the University’s functions and history, and shred the materials that fall under the other types of restrictions as we find them.

Only it isn’t even that easy. It would be one thing (one still very time-consuming thing) to simply shred or delete everything that had personnel, student, or donor information in it. But so often this information creeps in to exactly the types of records that we want to keep: the minutes and offer letters that describe how curricula is shaped, the reports and correspondence that discuss student activities and campus life, and the memoranda that discuss the direction of the university’s development through gifts from alumni and friends, among others.

When is it essential to screen for this mixed restricted content? We want to provide as much access to our collections as possible, which means balancing getting as many collections processed at the minimum acceptable level as possible while also ensuring we are restricting at the minimum level that we can be reasonably certain protects the records for which we are responsible. It also means figuring out the best time to screen for restricted materials, whether it is during accessioning, just before a box is delivered to a researcher in the reading room, or somewhere in between.

When we find materials that include restricted content but also include information of historical relevance, we can be paralyzed by it: do I shred this? Is the important information elsewhere? Do I restrict it? Do I restrict this whole folder, this whole box? Do I separate this material and if so how do I maintain original order? Do I redact parts of this? If so, do I keep the original elsewhere or shred it? Which parts, and how much? And so processing or screening slows down, until for some collections it feels like we may as well be processing at the item level, each document a minefield of risk, while the mantra of MPLP sounds in our ears, leading to, for me at least, an underlying feeling of guilt and anxiety I can’t shake.

Folder from the Vice-President for Student Affairs Records, showing a restricted date. Image courtesy Tracy Jackson.

Then, separately but tangled in the issues of time and process, once we have made those decisions, how do we clearly communicate and enforce these restrictions to the user before they ever come into the building, to the staff members who retrieve and serve these materials to users, to future versions of ourselves when the types and terms of restrictions inevitably change? How do we adhere to the law, to university policy, and to ethical guidelines while ensuring we consistently document the people, activities, and history of the university, given our limited time and resources? How do we make our own lives easier?

With all of these factors to consider, there are times it seems prudent to ask the question: how much do we really need to worry about it?

Answers to all of these questions no doubt vary from institution to institution. Considering your own institutional restrictions and policies, how much your collections are requested and used, what the record-keeping practices are and have been for both your creators and your collectors, your staffing capacity, and ultimately, your own and your institution’s level of comfort with risk, your answers and conclusions may be very different.

Here at Duke, we’ve been struggling with these questions for some time, and we’ve made progress. One focus is on how to clearly communicate the status (restricted, open, in need of review) of UA materials to our researchers and to other special collections staff who encounter them. Another focus is on clarifying our policies around the kinds of records we shred or restrict as well as how we go about identifying and treating them.

In many ways, the clarification of policies is the easier task. UA staff have hammered out policies outlining the types of documents that fall under each restriction type and how we prefer to treat them. With these policies in place, some of the burden and anxiety of decision-making when looking at records is eased, as we can see how we’ve already decided how to handle a certain type of record and have something to point to later on if any questions arise.

For new or previously unprocessed collections in the queue, we can identify the likelihood of restricted materials showing up in the collection as well as the anticipated demand for the materials and process accordingly, applying minimal processing to many collections while giving a few collections the more in-depth screening treatment. But of course, not every collection is new, and we are all familiar with the fact that processing practices have changed over time. For previously processed collections, we may need to screen boxes as they are requested.

Generally, I rely on folder titles to indicate whether I should look in a folder for restricted materials. Names and certain topics indicate the likelihood of some types of restricted materials. When I do go through a folder, I try to make a quick decision about whether I should restrict the whole folder according to one of the types above, moving it into a separate restricted box, or whether there is a small enough amount of material I can shred, or less frequently, redact.

Clarifying and communicating the status of materials is more complicated. Analog materials must be physically labeled, while digital materials may need to be stored on separate servers. Restriction information must appear in the collection description as well as circulation systems. At Duke that means three different systems containing different types of information (collection level, descriptive level, container level), as well as a large staff who are not all deeply familiar with the UA restrictions. We have evolved methods combining restriction notes and circulation limits in our systems with stamps and notes on boxes to indicate the accessibility of materials. For newly processed collections, we can achieve some consistency and clarity, while for legacy collections we juggle lists of screening needs and work completed. We try to indicate the date when screening for restricted materials was completed, which I hope will help future versions of ourselves when restriction guidelines change again, to know which standards were in use at the time of screening.

It is impossible to catch every instance of restricted material, as it is impossible to read every single page of every single record that comes into the archives. Researchers must sign a registration agreement that includes language about third party privacy rights, so some onus is put on the researcher for responsible use of information found. We also believe that our written policies and practices as shared above indicate our good faith efforts, which we can point to in the event that information is shared inappropriately in spite of these efforts.

This is the point at which I look at our colleagues in public universities with something akin to envy. But every archival institution has some version of these problems, and different methods of dealing with them. It feels like time for a wider discussion, so how are you handling yours?


Tracy Jackson is the Head of Center Manuscript Processing at the David M. Rubenstein Rare Book & Manuscript Library at Duke University and processes University Archives collections. She holds an MSLS from UNC-Chapel Hill and has been at Duke for four years. She is currently a member of the College & University Archives Section Steering Committee and previously served on the Description Section Steering Committee and the Theodore Calvin Pease Award Subcommittee.