Court Hands Jason Leopold A Partial Victory In Case Involving Sealed Dockets And Electronic Surveillance
from the openness-with-a-small-'o' dept
A half-decade’s worth of litigation by “FOIA terrorist” Jason Leopold is finally bearing fruit. The petition, filed in 2013 to peel back a few layers of opacity from the Feds’ favorite court (DC District Court), has been partially granted by Chief Judge Beryl Howell. (h/t Mike Scarcella)
Nearly two years ago, substantial progress was made when Judge Howell ordered the US Attorney’s Office (USAO) to examine sealed dockets (of which there are many — the DC circuit is home to hundreds of dockets rendered invisible by government requests for secrecy) and to start unsealing anything that wasn’t related to ongoing investigations.
The government fought back, but as the lengthy opinion [PDF] shows, there was much more cooperation between the USAO and Leopold than one would expect, given the government’s antipathy towards him goes so far the Pentagon once offered Leopold a stack of documents in exchange for him promising to never file another FOIA request.
Several pages are given over to the facts of the case. They’re definitely worth reading. It details the push-and-pull of the two parties, the logistical hurdles standing in the way of compliance, and the unexpected compromises made by both parties. Leopold was requesting access to Pen Register/Trap and Trace (PR/TT) requests and Stored Communications Act (SCA) warrants, many of which are apparently obtained by DOJ components with little input from from the USAO. Leopold agreed to limit his request to USAO documents, only to find out the USAO had no reliable system in place to track its own involvement in federal prosecutions.
In particular, determining whether the USAO or a different DOJ component had filed a PR/TT or SCA application would be challenging, as the USAO maintained no lists of docket numbers for PR/TT and SCA matters initiated by USAO prosecutors, and because the USAO’s internal tracking system for criminal investigations did not correspond to the Miscellaneous (“MC”) docket numbers assigned by the Clerk’s office. Moreover, even the USAO lacked access to the sealed MC dockets and, thus, could not determine which PR/TT and SCA applications were filed by the USAO or a different DOJ component or the status, as open or closed, of the investigations in connection with which those applications were filed.
After working out the kinks in the system, the USAO began identifying cases it was involved with. The culling still left the office with plenty of paperwork to dig through.
The Court provided the parties with the total number of matters responsive to the abovedescribed searches, with the significant caveats that such numbers could be under-inclusive by not capturing all SCA matters initiated by the USAO and other DOJ components, and may also reflect double-counting of § 2703(d) matters “since applications for § 2703(d) orders filed in more than one year in the same Miscellaneous matter will result in the same matter being counted in more than one year.” […] Specifically, the number of § 2703(d) matters responsive to the searches and filed in the following years were: 2008 – 80; 2009 – 55; 2010 – 136; 2011 – 90; 2012 – 64; 2013 – 160; 2014 – 334; 2015 – 581; 2016 – 1,136. Section 2703(d) Notice. The number of SCA warrant matters responsive to the parties’ suggested searches and filed in the following years were: 2008 – 0; 2009 – 68; 2010 – 121; 2011 – 152; 2012 – 164; 2013 – 131; 2014 – 107; 2015 – 252; 2016 – 271. SCA Warrant Notice. Thus, the total approximate number, over the relevant nine year period, of § 2703(d) matters was 2,636, and SCA warrant materials was 1,266.
A sampling was taken from the 2012 set by the USAO, which scrubbed personal info and anything else it felt might harm ongoing investigations or national security in general. Then it attempted to make the sampling process the regular process and eliminate much of the transparency the petitioners were seeking.
The redacted PR/TT materials, stripped of identifying information about the individual or underlying criminal activity under investigation, revealed that the USAO’s PR/TT applications largely used the same language to describe (1) the service provider from whom the USAO sought to compel production, (2) the scope of legal authority sought, (3) the need for such authority, (4) the steps the USAO would take in exercising that authority, including technical assistance to be required of the service provider, and (5) a request for sealing. See generally id. The parties expressed disagreement as to the significance of the information that the sample PR/TT matter materials revealed. The USAO described the materials as “substantially similar and reveal[ing] largely boilerplate information,” and argued that any additional information that unsealing all of the remaining 2012 PR/TT matters might yield would have little value to the public relative to the significant “expendi[ture of] judicial and prosecutorial resources” that such broad unsealing would entail. 3 rd Jt. Rpt. ¶¶ 4, 7. Consequently, the USAO proposed instead to undertake the same sampling process used for the ten PR/TT matters in 2012 for a representative sample of PR/TT applications in other years.
Leopold (and fellow petitioners Reporters Committee for Freedom of the Press) disagreed with this assessment, using the same four documents the USAO produced.
The petitioners described the unsealed sample materials as “substantively of interest to the[m], and the press and public more generally,” given that the materials “reveal[ed], among a number of other things, the carriers involved in each matter, as well as the fact that in two of the four matters the government repeatedly sought extensions of the court’s authorization to use a [PR/TT] device.”
The USAO agreed to an extraction process proposed by the court to sniff out cases of public interest — specifically, denied applications and/or substantive judicial opinions. Even with this drastically limiting what the USAO had to turn over to Leopold, the USAO insisted on applying the extraction method to only 10% of each year’s total. This brought the litigation to a standstill. It’s this standstill that’s addressed in the DC court’s opinion. It answers the question central to the petition’s forward momentum: does the public have a First Amendment right to access documents filed under seal?
After weighing all the factors, and the long history of nondisclosure that covers warrants and pen register requests, the court grants Leopold a partial victory. While access to underlying documents won’t be granted beyond the 10% sampling of past cases, the USAO will have to be more open going forward.
In sum, five of six Hubbard factors, including the “single most important” such factor, id. —need for public access to the documents at issue, fact of objection and objector’s identity, strength of any property and privacy interests asserted, possibility of prejudice, and purposes for which the documents were introduced—weigh in favor of prospective disclosure of PR/TT, § 2703(d), and SCA warrant matter docket information, while one factor—the extent of previous public access to the documents—has limited applicability. The Court thus concludes, considering the Hubbard factors together and as applied to the recently adopted administrative changes in the USAO and this Court’s Clerk’s Office, that the common law affords the petitioners a prospective right of access to the PR/TT, § 2703(d), and SCA warrant matter docket information.
This won’t, however, open up access to everything filed previous to this opinion. According to the court’s reasoning, extracting information from the thousands of cases already identified would be too burdensome on the government. (The USAO estimated at least 720 hours of work.) It acknowledges there are problems inherent to random sampling, but comes down on the side of the government.
Through these submissions, the petitioners make amply clear that, in their view, retrospective access to extracted information in PR/TT matters is an all-or-nothing situation: such access to less than one hundred percent of USAO-initiated PR/TT matters would have little to no value, and in fact might be worse than useless by yielding incomplete and misleading information. Yet, the administrative burden on the USAO and Clerk’s Office that would be triggered by compelling the USAO to fulfill the petitioners’ request for unsealing and disclosure of fifteen categories of extracted information in all PR/TT matters filed over nine years, would be unduly significant.
Accordingly, upon consideration, under Hubbard, of the USAO’s and Clerk’s Office’s particularized interest in avoiding undue administrative burden, the common law right of access does not entitle the petitioners to any additional retrospective relief.
The good news is the changes ordered were long-needed and will streamline future requests for information. Tracking cases filed under seal will be much easier for the USAO and sealed dockets will now provide enough info that members of the general public will be able to determine what sort of electronic data requests the government is making.
The Clerk’s Office has also adopted new CM/ECF case types to more readily identify the type of criminal investigative matter being initiated. Thus, rather than assigning general MC numbers to many different types of sealed criminal investigative matters, the Clerk’s Office will assign more specialized docket numbers reflective of the matter at hand. For example, “PR” will be used for PR/TT applications and “SC” for SCA applications, with special designations within the SC case type for SCA warrant and § 2703(d) materials.
Also added into the mix is periodic reporting on sealed cases by the DC district court.
[T]he Clerk’s Office, as set out in the MOU, plans to provide biannual docket reports about various types of sealed criminal investigative matters filed twelve through six months prior to each report’s publication. While some of these matters will remain open, the six-month delay in reporting somewhat reduces the risk to an ongoing criminal investigation. These reports will provide information about the total numbers of such matters and, as reflected in the standardized caption for each matter, the number and type of target accounts (e.g., landline telephone, cellular telephone, and/or email), the providers’ names, and the primary offense statutes under investigation. This information on the sealed applications subject to judicial review will provide additional transparency as to the processing of these sensitive matters, without jeopardizing either privacy or law enforcement interests. Indeed, this information is far more robust than that seemingly sought by the petitioners.
It doesn’t tear the roof off sealed dockets but it does get the public’s foot in the door. In the coming years, this should only improve. The court seems to recognize the government doesn’t do its best work in darkness and that the public should be part of the oversight process. This isn’t everything the petitioners were seeking, but it’s more than the government was willing to grant. It’s a small, but significant, victory and should lead to greater transparency in the future.