Searching over 5,500,000 cases.

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.

Long v. United States Dep't of Justice

March 25, 2010


The opinion of the court was delivered by: Hon. Norman A. Mordue, Chief Judge



Plaintiffs Susan B. Long and David Burnham bring this action pursuant to the Freedom of Information Act ("FOIA"), 5 U.S.C. § 552, to challenge the response by defendant, the United States Department of Justice ("DOJ"), to their FOIA requests for records from or relating to the DOJ Civil Division's ("the Division") case management system database ("CASES"). Plaintiffs allege that: the search defendant conducted for information responsive to their FOIA request was inadequate; the FOIA exemptions pursuant to which defendant withheld certain records are inapplicable; defendant failed to conduct a segregability analysis and release all reasonably segregable information; and defendant failed to mark redactions with regard to sealed cases. Plaintiffs seek injunctive and "other appropriate relief" to obtain the records and descriptive information relating to the CASES database they claim defendant wrongfully withheld. Presently before the Court are defendant's motion for summary judgment and plaintiffs' cross motion for summary judgment.



In 1989 plaintiffs founded Transactional Records Access Clearinghouse ("TRAC"), a data gathering, data-research and data-distribution organization at Syracuse University. According to plaintiffs, TRAC was established to provide the public with comprehensive information about the federal government's staffing, spending, enforcement and regulatory activities. Plaintiffs further state that TRAC's goal is to improve the functioning of representative democracy in the United States by providing the public with authoritative information about the working of the federal government. Plaintiffs aver that FOIA is the primary tool they use to collect information about the government. Plaintiffs state that once a FOIA request is processed and a government agency releases responsive information, TRAC analyzes the data to describe the extent the agency's activities correspond with the agency's stated goals, and ascertain the trends and geographic distribution of important government activities. According to plaintiffs, TRAC's analyses and underlying data are available to Congressional committees, news organizations, public interest groups, the public at large, as well as to government institutions, such as the Government Accountability Office, Supreme Court Library, and agency inspector generals' offices. Plaintiffs state that academics also use TRAC's analysis in their own research and publications.


Dorothy Bahr, Director of the Civil Division's Office of Management Information ("OMI"), stated in her declaration that CASES is the "current computerized case information system used for tracking basic data on filed civil cases as well as unfiled matters handled in the various litigating components (Branches and Sections) of the Civil Division." Bahr explained that "the data in CASES is stored in a number of different tables in a relational database structure that operates on a Microsoft Structured Query Language ("SQL") server back end, with a third party software, Lawpack, as the user front end." Bahr stated that "OMI has also developed a number of specific database modules for the exclusive use of individual components for specific case tracking and work processing needs."

C. Plaintiffs' FOIA Request

In a letter dated June 7, 2004 to James Kovakas, the Attorney-In-Charge of the Freedom of Information and Privacy Acts Office of the Division, plaintiffs requested:

(A) an electronic copy of those records pertaining to court cases filed or pending in court since October 1, 1999 (FY2000 to date) contained in the CASES database, and

(B) the following descriptive information about the CASES database:

(i) table schema and definitions of all codes used,

(ii) records describing the scope of coverage of cases included (and excluded) from CASES,

(iii) changes in CASES that have occurred during this period, including changes in case coverage, or in tables, fields, and codes that have occurred and when these changes took place,

(iv) current data input and users' manuals, including any directives supplementing (or used in place of) these

(v) descriptions of all regularly prepared reports currently using CASES

(vi) records describing any validation, error checking or other procedures currently used to ensure data quality

Whenever these records exist in electronic form, we request that they be provided on computer media and that you discuss with us the choice of suitable media and recording formats to be used.

Kovakas stated that he "initially denied Part A of plaintiffs' request pursuant to various FOIA exemptions to disclosure." In her declaration, plaintiff Long stated that defendant produced "a handful of documents including a table listing names and a brief description of 42 fields in CASES" in response to Part B.

Plaintiffs appealed defendant's denial of their request to OIP. On September 5, 2005, OIP remanded plaintiffs' request to Kovakas's office for further processing of the responsive records and disclosure of any non-exempt portions of these records.

According to Kovakas, the Division's FOIA office worked with OMI's technical staff to determine whether it was possible to provide an electronic copy of the CASES database while excluding fields containing exempt information. Kovakas stated that to do this, OMI staff wrote a query to select records from the database within the designated time period and limit the selected records to those pertaining to filed cases.

Kovakas stated that he used a "Data Dictionary", which contained descriptive information about individual databases and fields, to "determine[] what database fields were exempt from release". According to Kovakas, when OMI staff informed him that it was not technically feasible to indicate redactions in the extracted data, he "asked them to design the query to exclude the fields of data that [he] designated as exempt", including all sealed case records. Specifically, Kovakas stated, he asked OMI staff to "exclude all qui tam cases and all cases with the word 'sealed' in the caption."

On March 31, 2006, defendant sent plaintiffs a disk with an electronic copy of the CASES database in Microsoft Access. Defendant withheld from this disclosure certain categories of data pursuant to various FOIA exemptions. According to Kovakas, the disk also contained a relational diagram indicating how the table data could be linked together using internal file numbers that were included in the release.

According to Long, plaintiffs received: (1) "a limited amount of CASES data"; (2) "the first-version 'Data Dictionary'"; (3) a Case Classification Manual dated June 2001; and (4) "some electronic files related to the released CASES data, including a purported 'schema.'"

In a letter dated May 16, 2006, plaintiffs' attorney notified defendant that the March 2006 release only contained data for closed cases and did not appear to include data for all filed cases during the requested time period. Plaintiffs also complained, inter alia, that there was no indication where the redactions of exempt data occurred in the released tables or how much data was withheld.

Kovakas stated that upon inquiry he learned that "there must have been an inadvertent error in composing the search query, which resulted in the omission of pending cases from the release." Kovakas asked OMI to produce a second version of the database correcting the error.

On September 8, 2006, plaintiffs filed the instant action. On November 29, 2006, defendant sent plaintiffs a corrected disk containing CASES data for pending and closed cases, which had been "extracted using the new script" and document entitled "List of OMI Reports".

Defendant subsequently provided a "Quality Assurance Procedures Manual" to plaintiffs. Long stated in her declaration that although defendant's provision of information purported to include more case entries and data fields than the March 2006 production, it contained fewer cases with fewer data fields.

In January 2007, the Division's FOIA staff again consulted with OMI technical staff. A new member of the OMI staff, with FOIA response experience, indicated that he knew how to show redactions in the extracted data. Plaintiffs provided defendant with computer code and documentation to aid in redacting marking. Defendant then reprocessed its response to Part A of plaintiffs' request for a third time.

In anticipation of the new response, the parties' attorneys engaged in a series of conference calls regarding various technical aspects of the response. During these calls, plaintiffs clarified their request for "database 'schema'" and indicated that they wanted a record showing the databases structure that was extracted from the database itself as well as particular technical characteristics of the database tables and fields included in the new response. Plaintiffs provided the Division with computer code and documentation describing the necessary steps to extract the schema from the CASES database.

Based on plaintiffs' clarification, the Division extracted a list of tables in the database.

Along with a letter dated February 12, 2007, defendant sent plaintiffs a disk containing database information drawn from a portion of the CASES database and charts and manuals relating to the database. This release also included alert updates to the Case Classification Manual, a Database Integrity Manual, and a time entry guide. Plaintiffs complained to defendant that the manuals included references to appendices that were not produced.

According to Kovakas, the process of extracting table structures from the database revealed that the Data Dictionary document the Division had relied on when making exemptions determinations for previous releases was "flawed". Kovakas explained that "[b]ecause there was no descriptive documentation about these newly-discovered fields and tables" OMI staff created a new Data Dictionary. To do this, technical staff manually reviewed the fields and inserted descriptions in the new Data Dictionary.

Defendant produced its new response to Part A of plaintiffs' request in five stages between February 23, 2007 and August 15, 2007. Additionally, on April 11, 2007, defendant sent plaintiffs materials responsive to Part B of their request, including pages and the appendices that plaintiffs complained had been missing from previous manuals. During the production of the staged release of tables, Kovakas determined that the "ZGRANDJURYPARTY" table, while responsive, was exempt in its entirety.

Regarding Part B of plaintiffs' request, defendant released all responsive material it located, except "JCON IDs". Kovakas explained that JCON IDs are "internal employee identification used" which are also used by employees as "login identification codes that allow access to the Division's network." Kovakas stated that after conferring with Division network security staff, he learned that "Division policy prohibited the disclosure of these codes because they are one half of what an individual would need (along with a password) to access the system." Kovakas therefore determined they were exempt from disclosure.

Defendant deemed the following categories of information exempt under FOIA: JCON IDs; grand jury proceedings; sealed cases; internal and client recommendations on litigation activities; attorney assessments of case valuation and other financial aspects (for open cases only); daily time reporting records of attorneys and paralegals (for open cases only); amounts spent on investigations (for open cases only); attorney notes or comments; names of consulting doctors in vaccine litigation cases; types of vaccines administered and dates of administration, in vaccine litigation cases; addresses and other contact information for pro se litigants, witnesses in vaccine litigation cases, parties to and beneficiaries of structured settlements, and opposing and non-Division counsel; dollar amount recovered from an identified private party; information relating to disbursement of structured settlements; names of parties associated with alternative dispute resolution proceedings or with particular dispositions in a case, captions for unfiled matters, which likely contain names, and names of vaccine litigation witnesses; government-assigned unique identification numbers; information that would identify government employees seeking government representation in their individual capacities; and details of planned absences of attorneys in the Office of Immigration Litigation.*fn1

In support of defendant's categorical withholding, Kovakas explained in his supplemental declaration that: the instant Request for data from the CASES database presents a situation quite different from that involved with the typical record, where it is possible to review the actual contents of the record when making exemption determinations. Here, plaintiffs requested the contents of a database for a time period such that the response involved over 200,000 cases. Each case may include data in any of the 91 tables and approximately 1500 fields that were ultimately identified as part of CASES. I was not able to review each of the up to 300 million data entries individually in order to make an entry-by entry determination of what was exempt. Even if the sheer amount of data did not make that endeavor impossible, individual pieces of data would have no meaning to me without an understanding of the categories (defined by tables and fields) in which the data entries occur. In other words, the database itself constitutes a "categorical" structure, and this structure provides the only workable framework for making exemption determinations.

I therefore used a categorical approach to exemption determinations based on the descriptions of tables and fields provided to me by OMI staff, supplemented by information my staff acquired through consultation with OMI staff and individuals at other litigating components within the Civil Division. When I determined that a field was exempt, I generally withheld the data contained in that field for all case records. This decision was based on my conclusion that entries in any particular database field are not reasonably segregable unless there is a code or another database field that can function as a "flag," such that if the value in the latter is X, then the former field may be released. Without such a flag, even a manual review of the data would not necessarily allow information to be segregated. For example, in order to "segregate" dollar amounts that are estimates from dollar amounts that appear on a plaintiff's complaint, it would be necessary not only to manually review each data entry in this field in CASES, but also to compare the data entry in each field with information .in the complaint and/or any amended complaint filed in that case, in order to compare any requests for monetary relief that appear in these complaints with the amounts that appear in the relevant database field. This task would be immensely time-consuming.



"FOIA was enacted to promote honest and open government and to assure the existence of an informed citizenry to hold the governors accountable to the governed." Grand Cent. Partnership, Inc. v. Cuomo, 166 F.3d 473, 478 (2d Cir. 1999) (internal quotation marks omitted).

"FOIA strongly favors a policy of disclosure and requires the government to disclose its records unless its documents fall within one of the specific, enumerated exemptions set forth in the Act." National Council of La Raza v. Department of Justice, 411 F.3d 350, 356 (2d Cir. 2005) (citing 5 U.S.C. § 552(a)(3), (b)(1)-(9)). Courts construe these exemptions narrowly, resolving all doubts "in favor of disclosure". Local 3, Int'l Bhd. of Elec. Workers v. NLRB, 845 F.2d 1177, 1180 (2d Cir.1988). The government bears the burden of showing "that any claimed exemption applies."

National Council of La Raza, 411 F.3d at 356. Courts review the government's decision to withhold or redact information de novo. 5 U.S.C. § 552(a)(4)(B).

B. Summary Judgment Standard

Summary judgment is appropriate where there is no genuine issue of material fact and the moving party is entitled to judgment as a matter of law. See Fed. R. Civ. P. 56(c). Substantive law determines which facts are material; that is, which facts might affect the outcome of the suit under the governing law. See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 258 (1986). Irrelevant or unnecessary facts do not preclude summary judgment, even when they are in dispute. See id. The moving party bears the initial burden of establishing that there is no genuine issue of material fact to be decided. See Celotex Corp v. Catrett, 477 U.S. 317, 323 (1986). With respect to any issue on which the moving party does not bear the burden of proof, it may meet its burden on summary judgment by showing that there is an absence of evidence to support the nonmoving party's case. See id. at 325. Once the movant meets this initial burden, the nonmoving party must demonstrate that there is a genuine unresolved issue for trial. See Fed. R. Civ. P. 56(e). A trial ...

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.