Librarianship is the art of intermediation. Sometimes more, sometimes less. A key law library service is document delivery, which involves a negotiation between the recipient and the library staff. Those negotiations became more common and more complicated during the pandemic. It highlighted some opportunities to improve our document delivery decision-making by looking at where it ran into copyright limitations.

I’m not going to dwell on copyright law too much. This is a post about practical issues. I mean, it exists and so we are constrained by it. My current law library was sued by a legal publisher in the early part of the century so we are particularly sensitive to the issue.

And it’s a sensitive situation for most law libraries because the area carved out for us – fair dealing or fair use – isn’t always really clear. You deal with the ambiguity as best you can. Groups continue to try to shakedown libraries on copyright issues, including the publishers. A defensible approach to copyright is a fundamental need for any library, law libraries included, that do document delivery.

This isn’t different in a law firm law library, either. I can’t find documentation for it anywhere but I remember a lot of concern in the 1990s while working at a law firm in connection to the BNA Daily Labor Law report. Perhaps urban myth, but a law firm had bought a single copy and then photocopied it for all firm offices. Not cool. It was soon after the Kinko’s case so maybe it created an environment where publishers were confident in suing customers.

One thing that is clear in a law library is that lawyers do not understand copyright, as a rule. One can see where a publisher might get concerned that a lawyer and a photocopier might have a negative impact on business. Although, a lawyer willing to photocopy a book was probably not going to be a customer, will probably not maintain the text, and may find himself a subject of a complaint for failure to know the law.

Can’t Rely on Common Sense

It will be the rare courthouse law library that hasn’t dealt with challenging researchers. The pandemic has meant, in our law library, that document delivery has exploded. Where before a lawyer might have walked into the library, found the appropriate texts, and copied them, each interaction has staff involved.

A chart of document delivery interactions (not pages or titles) before the pandemic (left, blue) and after we closed in March 2020 (right, red). The chart scale is daily with a peak of 500.

This has meant explaining – and sometimes re-explaining – why we have a policy on copyright and how it works. It also means firmly but clearly explaining that:

  • you cannot have the entirety of a text;
  • you cannot receive the maximum amount of pages from one text and then have another lawyer or staff person at your firm call to receive the same amount again of the next contiguous part of a title;
  • we do not have an obligation to provide full digital texts even if we have access to them. This is true even when someone asks for an accommodation. Our role isn’t to distribute texts, as I confirmed with the title’s publisher. I spent an incredible amount of time on a book a lawyer was unwilling to spend $120 for. But it was good to learn that some publishers, if they sell the print book, will deliver a digital copy as an accommodation.

It makes you wonder what they do when you’re not directly involved and merely ask them to abide by your copyright policy. And, of course, each interaction uses your limited resources: staff knowledge, collection items, and perhaps most important, time. This process can be two to three times as long done virtually, where each request may require a discussion, or a shared table of contents, and so on.

At the same time, there is a lot of room for discretion and that can slow down the process. Our policy has focused on 5% of a text, which seems well within the fair dealing aspect of a book. But even that isn’t a firm rule so much as a ceiling.

It was the explosion in document delivery, though, that probably made the decision making around copyright and document delivery a lot more acute. We were, literally, in everyone’s business because no lawyer could access the print or digital collections. And when one service expands, it reduces the time and resources available to other services. So every extra decision point became a drag on the entire team.

Rules and Rulers

Some of our document delivery decisions are not based on copyright. Some are just based on fairness and the time required for staff to fulfill. A lawyer can easily dominate staff time to the detriment of other people on things as trivial as pulling cases from online databases.

That’s a resource constraint issue. Government or public domain documents aren’t subject to copyright but that doesn’t mean we are able to generate a document in response to every request. Copyright is a legal obligation that has a real cost, both potentially in money and corporate reputational risk.

The trick then in a law library is to find that process that ensures copyright compliance to the best of your ability without further consuming resources (staff, time, equipment). Large academic libraries may have a copyright office. The most extensive document delivery service I’ve seen in a courthouse law library was at Boston’s Social Law Library.

So this summer, I did a review of copyright policies at other law libraries to see what was common. That can be risky, particularly when your law library is subject to different copyright laws than another law library. But it was interesting to see how much variety there was and unsurprising to find we were on the conservative end.

At this point, there’s a strategic decision to make. The goal is to ensure that document delivery is streamlined while respecting copyright. There many potential tumblers to this lock. There are nearly always multiple ways to get to where you need to get. So I was not facing a choice of looking at our current policy (no more than 5%) and changing that number to get the result (some libraries use 10%).

The first thing we did was to clarify what the staff’s understanding was. This meant that, in addition to the nominal 5%, we set other specific caps. Things that were easier to measure and had their own purpose. A document delivery request will not be automatically fulfilled if it exceeds the lesser of 100 pages, 2 chapters, or 5% of the text.

Why 100? Why not 50? Why 2 chapters and not 1 or 3? How many journal articles. This is all pretty arbitrary stuff. We didn’t think a single measure was adequate (100 pages from 150 page text?) so we had a variety of cut offs. But they are all still somewhat arbitrary, tied as much to effort as copyright law.

Law books are tricky, though. What is 5% of a looseleaf text? Online versions aren’t always paginated. Print format versions may not have sequential pagination.

In the end, on multi-volume titles, we decided to measure the 5% based on the linear width of the text. If you have a great cataloger, you can use the 300 field (Physical Description) which can contain a width. Otherwise you can pull out a ruler and, assuming you’ve got access to your physical collection, measure.

If you do have not done looseleaf filing, you may not be able to picture the pagination challenge. It is not uncommon to pull out a page that has 1 and 2 on either side and replace it with two pages, one of which has 1A and 1B, the second of which has 2 and a blank reverse.

What about when an online ebook or resource, which is easier to use in order to fulfill the document request, has a different pagination (or nonexistent) from the print? Digital ebooks and database content will vary widely depending on the platform and how you output the document selection. Since the digital version was inconsistent, we went with the print format pagination if we had it.

The process of clarifying the guidelines and ensuring each person knows what they are was an immediate improvement. We added a time maximum as well – how long it takes to compile the documents requested – and this also helped to clarify everyone’s expectations.

The ultimate benefit was that, so long as the request didn’t hit any maximums (amount of content, amount of time), it could be processed by the person who received it. This cut one person out of every transaction. We still have a second check if a request exceeds 5% but that’s rare.

We’ll see how this shakes out but I do not believe we will see a return to the pre-pandemic levels of document delivery. It may drop but I think it will be a much larger part of our service delivery than it once was. It’s a good time to review our approach and retool.