When I went to law school, one of my more disagreeable professors gave us a memorable definition of the task of courts in statutory interpretation cases: “A judge’s job is to figure out who Congress was trying to screw, and then screw them.” Even if you accept that formulation (and it seems to rest on a pretty simplistic view of judging), what it leaves unstated is an even more compelling imperative for the courts: to make sure that people Congress wasn’t trying to screw don’t end up getting screwed anyway through the inadvertence, inattention, or perversity of the courts.
There was a good deal of talk on the cyberprofs listserv this past weekend concerning the Fifth Circuit’s new DMCA anticircumvention opinion, MGE UPS Systems Inc. v. GE Consumer and Industrial Inc.. At one level, MGE is all about making sure that people Congress didn’t mean to snare within the statute’s prohibition on circumventing DRM technologies don’t find themselves on the losing end of a DMCA claim. The problem, as many others have observed and as I have argued (at probably undue length) elsewhere, is that the statute is simply a mess: its liability provisions invoke concepts that have no settled meaning in the law, its definitional provisions confound rather than clarify, and the policies underlying the DMCA (regulating access to works) simply do not fit with hundreds of years of copyright (regulating uses of works). The statute’s basic incoherence and incompatibility with prior copyright law makes it pretty difficult for a court to know whether it is applying the statute in a way that Congress intended or not.
There are basically two possible responses to the incompatibility between the text of the DMCA as written and the prior corpus of federal copyright law. One is to cast the rest of copyright aside and make a clean break: to declare that the DMCA ushered in a new statutory era in which prior copyright concepts are irrelevant. This is basically the approach the courts took in the earliest wave of DMCA cases, such as in Universal City Studios v. Reimerdes, 111 F. Supp. 2d 294 (S.D.N.Y. 2000), aff’d, 273 F.3d 429 (2d Cir. 2001). Presented with the argument that the DMCA appeared to penalize conduct (such as fair use) that was clearly lawful under copyright, the court said, in essence, copyright doesn’t matter. The DMCA is a different statute, the Reimerdes court reasoned, and “[i]f Congress had meant the fair use defense to apply to such actions, it would have said so.” (p. 322). To construe the DMCA, in other words, we don’t need to look outside the DMCA itself. As I tried to argue in my article, however, the text of the DMCA simply doesn’t permit a court to discard the rest of copyright law, because the DMCA relies at multiple junctures on concepts that have no legal meaning outside of copyright. Trying to wall off copyright from the DMCA ultimately makes it very difficult to interpret the DMCA, because it was not written to be self-contained.
The other, more difficult, alternative for the courts is to try to harmonize the DMCA with the rest of copyright law and make them coexist. This turns out to be a surprisingly difficult endeavor because although the DMCA uses copyright language and concepts, it does so in the service of a legislative purpose (regulating access) that was unknown to prior copyright law. Complicating the interpretive process still further is what seems to be a deep reluctance on the part of the courts to recognize forthrightly that the text of the statute is as messy and contradictory as it is fact is. That is to say, even when the courts in DMCA cases are in fact engaged in a delicate balancing of competing policies, making tradeoffs that Congress should have spelled out clearly but did not, they nevertheless insist that their interpretations are commanded by the statutory text alone. That reluctance to highlight the DMCA’s shortcomings head-on has occasionally led the courts to say some things that are surely incorrect, such as that copyright law presumptively allows the copying of unencrypted content (as my friendly local appeals court suggested in Lexmark International v. Static Control Components, 387 F.3d 522, 546 (6th Cir. 2004)). Nevertheless, despite their shortcomings, the more recent generation of DMCA cases does work much harder than the first generation of cases did to ensure that the DMCA doesn’t trample over the rest of copyright law.
Which brings us to the Fifth Circuit’s new MGE decision, in which a panel (consisting of all Republican appointees, although I doubt that fact holds much significance) rejected a plaintiff’s DMCA claim against a company that bypassed a little encryption dongle that the plaintiff had distributed along with its copyrighted software. The software was designed to query the dongle to make certain it was allowed to run, but the defendant in the case (relying on information gleaned from the Internet) modified the software to bypass that security check.
At one level, MGE falls into the same category as cases like Lexmark and Chamberlain v. Skylink, in which the DMCA seemed to have been invoked merely to increase competitors’ costs. At another level, however, MGE seems even more hospitable to defendants in DMCA cases than Chamberlain, because it essentially reasons that circumventing a DRM mechanism is not actionable at all unless copyright infringement results:
Merely bypassing a technological protection that restricts a user from viewing or using a work is insufficient to trigger the DMCA’s anti-circumvention provision. The DMCA prohibits only forms of access that would violate or impinge on the protections that the Copyright Act otherwise affords copyright owners. … Without showing a link between “access” and “protection” of the copyrighted work, the DMCA’s anti-circumvention provision does not apply. The owner’s technological measure must protect the copyrighted material against an infringement of a right that the Copyright Act protects, not from mere use or viewing.
(p. 6). If that’s so, of course, then the Reimerdes court’s refusal even to inquire into the proffered fair use defense in that case was clearly erroneous.
Although I happen to like the result in MGE, the opinion has some flaws, most of which were pretty well ventilated in this past weekend’s cyberprof discussion. The court’s treatment of the DMCA issue is pretty cursory relative to its importance. And by essentially making infringement of a Section 106 right a precondition to circumvention liability under Section 1201(a)(1), the court seems to leave Section 1201(b) without much work to do. More fundamentally, the same criticism that others have leveled against cases like Chamberlain v. Skylink seems to apply equally to MGE, namely, that nothing in the DMCA actually requires proof of an antecedent act of copyright infringement before the circumvention provisions become applicable. The MGE court’s treatment of the statutory meaning of “access” to the copyrighted work also does what the courts seem to do depressingly frequently: quote the definition once, then essentially ignore it in the entire discussion that follows in favor of policy arguments about what sorts of “accesses” ought to be allowed.
Still, MGE is a welcome addition to the DMCA canon, and I certainly expect that it will wind up as an assigned reading in my spring Copyright course. The court’s remarks about the possibility of a fair use defense to an anticircumvention claim are about as strong a statement as any court has ever offered in support of such a rule, although here too (as in all the other recent cases) it’s merely dicta.