The Chameleonic Library of Congress

Back in March, the U.S. District Court for the District of Columbia released a significant (and very bad) opinion, Medical Imaging & Technology Alliance v. Library of Congress, that may have significant implications for the crafting of U.S. copyright policy and oversight. If the court’s reasoning stands on appeal, it effectively provides Congress with a sneaky way of marshaling copyright policymaking through the Library of Congress that renders it effectively immune from judicial and executive oversight.

Section 1201 and the Triennial Review

First, some background: this involves an appeal of the triennial review of exemptions from the anti-circumvention provisions of 17 U.S.C. § 1201,1 conducted by the Copyright Office and Library of Congress. Normally, if you want to make a non-infringing fair use of a copyrighted work, you just do it—no permission required. But if your use happens to involve a copyrighted work encumbered with some kind of digital rights management that you have to circumvent to engage in the use, Section 1201 will still leave you on the hook for the circumvention, unless you either qualify for one of the statute’s narrow built-in exemptions or persuade the Office and Library to grant you a temporary one in the triennial review. Though many advocates have urged reading a fair use exemption into Section 1201 under a “nexus with infringement” theory, that view’s not been widely adopted by the courts enough to safely rely on it. So fair users—people with disabilities, security researchers, people repairing their stuff, teachers, video remixers, and many more—have to descend on the Copyright Office every three years to plead for new and renewed temporary exemptions from Section 1201 (in a proceeding that’s now ongoing).

Normally, complaints about the results of triennial review come from advocates for exemptions when the Copyright Office doesn’t go far enough. But here, the beef comes from opponents of an exemption involving the repair of medical devices who think the exemption shouldn’t have been granted.2 (The notion of the Copyright Office going too far with an exemption is painfully ironic to advocates who repeatedly have seen the Office whittle exemptions down to the nub, as I explained at length in my Senate testimony a couple of years back. But I digress.)

Challenging the Triennial Review

That brings us to the exemption proponents’ interrelated challenges to the triennial review:

  1. Either the Office/Library are subject to the Administrative Procedures Act and the court should throw out the medical device repair exemption as arbitrary and capricious; or
  2. If the Office/Library aren’t subject to the APA, that’s because the Library of Congress is, well, part of Congress, and Congress can’t delegate to itself the ability to pass laws without going through bicameralism (the passage of a bill by both houses of Congress) and presentment (the presentation of the bill to the President, who must sign it into law or have his veto overridden by Congress), as I think I learned in Schoolhouse Rock.3

As a matter of substance and policy, I think the exemption opponents are pretty obviously wrong on the arbitrary and capricious argument. The Office recommends exemptions only very begrudgingly, and the narrow thing for the court to have done here is to have accepted that the APA applied arguendo, evaluated the APA challenge on the merits, and determined that the exemption was properly granted.

Nevertheless, proponents and opponents share some common interest in the position that the triennial review ought to be subject to oversight by the courts. But Judge Howell, who happens to have served as counsel to the Senate Judiciary Committee in 1998 when the DMCA was passed, goes through a remarkably circular line of reasoning to reach the even more remarkable conclusion that the Library of Congress (and the Copyright Office, so long as its work is blessed pro forma by the Library) can adopt binding changes to copyright law without any oversight at all, either from the judiciary or the executive branch.

Huh? Let’s break this down.

The Library of Congress, the Copyright Office, and the APA

As to the first challenge: ordinarily, suits against the government are barred by sovereign immunity, which must be explicitly waived by Congress. The APA, however, waives sovereign immunity for actions by administrative agencies, meaning that people drag agencies into court all the time for not doing rulemakings right.

So why isn’t the triennial review obviously subject to the APA? The Library/Office are certainly acting like an agency, and if that weren’t enough, the Register of Copyrights is explicitly subject to the APA under Title 17!

Not so fast, says the court. The APA specifically excludes Congress as an “agency,” and the Library of Congress is, you know, of Congress. (There’s some old Supreme Court case law about the Library being not an “agency” subject to the Freedom of Information Act when it’s doing library stuff, like holding tapes of Henry Kissinger. 😬)

But what about the fact that the Copyright Office is explicitly subject to the APA? According to the court, the fact that the Copyright Office is specifically covered but not the Library of Congress actually establishes that Congress didn’t want to subject the Library to the APA.

That brings us to an important aside: who actually conducts the triennial review? As a practical matter, the review is conducted by the Copyright Office, whose staff drafts almost all of the relevant rulemaking documents, conducts the hearings, and ultimately puts out the lengthy recommendation (attributed to the Register of Copyrights) that deals with all the messy substance of reconciling the often-contentious record. But as a formal matter, Section 1201 tasks the Librarian of Congress with actually issuing the final rule. That’s the case even though the Librarian almost always approves the Register’s recommendation.4 Also, query what to make of the fact that the Copyright Office is an inseparable part of the Library of Congress. But none of this matters to the court; the Librarian issues the final rules, and that’s what matters for the APA analysis.

One might sensibly react at this juncture by saying something like, “Hey, the Library doing rulemaking looks more like something the executive branch does than what the Library does when it’s doing library stuff with Henry Kissinger’s dirty laundry! Shouldn’t we read the APA as only excluding the Library as part of Congress when it’s actually acting as part of Congress?”

Nope, says the court: the Library of Congress is of Congress. It’s right there in the name! There isn’t any kind of “hybrid” creature in the APA; under the APA, “Congress” is Congress. The court points out that this is all a “choice that Congress could revisit at any time by amending the APA’s definition of ‘agency’.”

The Library of Congress and Separation of Powers

One might then sensibly react by asking:Doesn’t that then raise a separation of powers problem? We have Congress delegating rulemaking authority to Congress. Congress isn’t subject to judicial oversight under the APA because normally it has to pass a bill through both houses of Congress and get the President to sign it—and go through judicial review on the backend to make sure it’s acting within the bounds of the Constitution. Isn’t this what we learned in Schoolhouse Rock?”

Nope, again, says the court: for the purposes of separation of powers, it turns out the Library is a “hybrid” agency after all. Sometimes the Library does legislative stuff and is part of the legislative branch, but other times, like when it rubber-stamps the triennial review, it is actually part of the executive branch ‘cuz the President can hire and fire the Librarian. And so long as the Library is not exercising legislative and executive powers simultaneously, it’s all good.

“But wait a minute!” You (an admirably astute reader) now wonder: “If the Library of Congress is doing a rulemaking as part of the executive branch for the purposes of the Constitution, isn’t it subject to Congress’s explicit checks on the executive branch, like the APA’s waiver of sovereign immunity, for that same rulemaking? It sure seems like the Library needs to be acting in one capacity or another!”

Nope, says the court (my paraphrase): the Library of Congress is a chameleon that can escape judicial scrutiny as a non-hybrid legislative thingy that is inseparable from Congress for the purposes of the APA and simultaneously escape separation-of-powers scrutiny by acting as a hybrid legislative-executive thingy for the purposes of the Constitution, and oh yeah, it’s the executive part of the hybrid thingy that’s doing the rulemaking for Constitutional purposes even though it’s the legislative whole of the non-hybrid thingy that’s doing the rulemaking for APA purposes.


What Now?

The only saving grace of this totally circular reasoning is that at this point, Title 17 doesn’t currently delegate all that much rulemaking authority to the Library/Office—well, aside from gatekeeping fair uses by a wide range of marginalized communities! (As Paul Ohm and I described in Regulating Software When Everything Has Software, we’re headed for a world where nearly all copyrighted works are entangled in software that can be conceptualized as an access control under Section 1201. That means that fair use of works will increasingly implicate Section 1201, and fair users increasingly will have to rely on the Copyright Office to bless their uses ex ante with temporary exemptions.)

Okay, this is actually pretty bad.

But it gets worse. This dynamic presents Congress with the opportunity to shadow-delegate other strains of copyright policymaking to the Copyright Office and avoid any kind of judicial or executive oversight by laundering the results through the Library of Congress. By extension, there’s now a perverse disincentive for Congress to extend the APA to the Library.

On the other hand, pressing the obvious separation-of-powers issues with an imperial Copyright Office charged with making copyright policy might create a remedy that’s worse than the disease when it comes to fair use. If the Library/Office can’t rulemake and thereby can’t issue exemptions at all, wither fair use in a world where so many uses implicate Section 1201?

One possibility is for a court to rule that the triennial review is not severable from Section 1201 and thus that the whole statute is unconstitutional on separation-of-powers grounds. If that fails, then Section 1201 effectively bars a wide range of fair uses. This is arguably a violation of the Supreme Court’s “fair use resolves all the major First Amendment problems with copyright law” handwaving in Eldred and Golan. Though the courts ought to conclude that there’s no way Section 1201 can be consistent with the First Amendment absent any outlets for fair use, the DC Circuit’s recent holding in Green v. DOJ doesn’t inspire confidence that that’ll happen.

Channelling Winston Churchill, the unfortunate reality may be that the triennial review remains the worst option for fair users except for all the others that have been tried. With no love from Congress or the courts, a world where the Library and the Office begrudgingly dole out narrow exemptions for fair use is better than a world with no exemptions at all. But at a minimum, it seems critical for proponents of balanced copyright policy to be on guard for further delegations to the Library and Copyright Office.

1 Though it’s often cited as Section 1201 of the Digital Millennium Copyright Act (DMCA), that’s wrong! Section 1201 of Title 17 is actually added by Section 103(a) of the DMCA, Pub. L. 105-304. And unlike the Communications Act, Title 17 of the U.S. Code is positively codified, which means that it’s actually the law. So, 17 U.S.C. § 1201 is the law, Section 1201 of the DMCA doesn’t exist. </pedantry>

2 The details are complicated, but you can read the exemption itself at 37 C.F.R. 201.40(b)(15) and a lengthy discussion of the issues starting around p. 193 of the Recommendation of the Register of Copyrights.

3 There’s a third argument that the grant of the exemption was ultra vires—an extreme departure from the bounds of Section 1201—that doesn’t really get anywhere because the standard for a UV argument is an even harder to bar for a challenger to meet than the already difficult arbitrary-and-capricious standard.

4 The one exception I’m aware of is in the 2010 review, when the Librarian rejected the Register’s shameful recommendation not to renew the exemption for e-book accessibility for blind and low vision readers.