Back in December, WIRED magazine did a piece about Carl Malamud of Public.Resource.org and his mission to make federal court records freely available on line, rather than exclusively through the court system’s own proprietary PACER network.   For anyone who has used PACER, the idea of a free, easy-to-use alternative has great appeal. The PACER system is not word searchable, so the only way to find documents is by entering the docket number or case name, two pieces of information that many researchers do not have when approaching a general legal question.  Also, PACER charges eight cents a page for materials other than court decisions, which actually generates a surplus for the courts – reportedly $50 million in 2006.  And PACER is not available to all curious Web users with the click of a mouse, but requires a government-issued password and login number.  
 
Working with Mr. Malamud, a number of government-reform activists have begun copying large amounts of material from PACER and placing them online without restriction. The New York Times of February 13 reports that one such activist, Aaron Schwartz, downloaded over 19 million pages of documents through a free PACER connection at a public library, an amount of text representing about 20% of the content on PACER.  (The free public library access was suspended on September 29 because of a reported security breach; it is unclear whether the mass downloading by activists such as Mr. Schwartz was at issue.) 

As a copyright matter, works of the U.S. government, such as court orders and opinions, are not subject to copyright. 17 U.S.C. 105.  The mass digitization of such documents, to enhance the public’s access to them, seems to provide a benefit to the citizenry at no cost to any copyright owner.  So far so good.

Many documents on PACER, however, are not U.S. government works and are subject to copyright protection.  Most obviously, papers such as briefs, affidavits and pleadings prepared by the parties in a lawsuit are works of authorship by private parties, not by the government.  They do not lose their copyright protection simply by virtue of being filed with the court, any more than a photograph loses its copyright protection when it is registered with the Copyright Office.   Certainly their authors must understand that such works will be accessible to the public as part of the official court file, and thus there is at the very least an implied license for PACER to distribute the works.  That implied license would not extend to a third party who copies and redistributes the works in a completely unforeseen and unrestricted way, however.  As a practical matter, few parties might care if their legal briefs and affidavits were made freely searchable on the Web, but if they did object there would be no argument short of fair use that would preclude a finding of infringement.  (I have first-hand knowledge of at least two instances in which attorneys have asserted copyright in their legal briefs, both having to do with an attempt to prevent competing attorneys from using the works. Neither matter resulted in litigation.)  Depending on the specifics of the use, of course, a fair use argument might or might not succeed – the mere fact that the plaintiff’s work is a legal document does not end the analysis.

Moreover, in copyright cases, these (often decidedly unmarketable) legal filings frequently have voluminous attachments consisting of more intrinsically valuable copyrighted works such as photographs, novels, blueprints, computer source code, screenplays, and graphic designs.  Few attorneys now think twice about attaching such documents to legal filings, when the case requires it, but the prospect of knowing that any exhibit in a case could be available the next day on the open internet could certainly give one pause.  Mr. Malamud of Public.Resource.org has flagged the related problem of court documents that disclose Social Security numbers and other private information, and this is indeed a serious problem. But the copyright side of that problem could also create legal liability for those doing the digitizing, and this risk should not go overlooked. 

This is not in any way to denigrate the effort to make uncopyrightable PACER documents more usable and accessible, but it is to suggest that the free-PACER forces should carefully consider the copyright implications of their project.  If the PACER reformers wish to minimize legal risk, they should be mindful of two critical facts: (1) not everything on PACER is U.S. government work, and (2) a copyrighted work does not lose its protection simply because it is filed with a court or other government agency.