“Boiling the Ocean”? Cause of Action Institute v. Department of Justice
Is a memorandum or communication encompassing several distinct subjects one “record” or multiple “records”? Thus, in response to a Freedom of Information Act (“FOIA”) request, can an agency redact as non-responsive the portions of the memorandum/communication involving subjects outside the scope of a FOIA request?
I. The Concept of “Scoping”
Imagine, a State Department cable to a large U.S. Embassy. For efficiency, the State Department might bundle inquiries or instructions regarding a variety of distinct subjects directed toward a number of different Embassy offices or employees into one cable, rather than sending numerous separate cables. This scenario might provide the archetypal case for treating one communication as multiple distinct documents, and thus producing only the portion of the omnibus cable responsive to a FOIA request. FIOA Update: OIP Guidance: Determining The Scope Of A FOIA Request, THE UNITED STATES DEPARTMENT OF JUSTICE (Jan. 1, 1995) (“1995 OIP Scoping Guidance”) (the State Department cable scenario was one of two examples the Office of Information Policy provided to illustrate the appropriate use of scoping).
Responding to FOIA requests in this manner, disaggregating one memo or communication into several distinct documents and redacting out the non-responsive “documents” is known as “scoping.” Does the diplomatic cable scenario provide a proper analogy for the treatment of follow-up questions by members of congressional committees transmitted by the committee chairperson as one document and responded to by the agency as one document? A D.C. Circuit recently panel recently held that it did not. Cause of Action Institute v. Department of Justice, Dkt. No. 20-5182 (D.C. June 1, 2021) (accessible on westlaw at 2021 WL 2197710). But in a concurrence that could come to have great significance, Judge Neomi Rao offered broad observations that may have implications for the standards used in judging the adequacy of FOIA requests.
Nothing about scoping or treating one memorandum or communication as two or more distinct “records” is mentioned in FOIA itself. The practice was accorded recognition by DOJ’s Office of Information Policy (“OIP”) in 1995. 1995 OIP Scoping Guidance, supra. OIP provides authoritative guidance to all agencies regarding FOIA’s requirements. The D.C. Circuit discussed the practice in a 2016 case, American Immigration Lawyers Association Ass’n v. Executive Office for Immigration Review (“AILA”), 830 F.3d 667, 676–79 (D.C. Cir. 2016). In AILA, the Court acknowledged that scoping could be appropriate in some circumstances. And, consistent with OIP’s policy guidance, the Court expressed skepticism that deletions of particular sentences in a memo or communication could ever be recognized as a proper application of the practice of scoping. OIP updated its guidance on scoping in light of AILA in January 2017. Office of Information Policy, Defining a “Record” Under the FOIA (January 12, 2017).
II. Cause of Action v. DOJ
A. The Facts and the District Court Decision
Cause of Action made a FOIA request for all communications between high-level Department of Justice (“DOJ”) appointees, members of Congress and White House staff “relating to” two subjects: (1) grants made by three DOJ components, namely the Office of Justice Programs, the Office on Violence Against Women, and Community Oriented Policing Services and (2) Executive Order 13457, which directed agencies to refuse to expend funds based on earmarks not set out explicitly in appropriations statutes (such as earmarks identified in committee reports accompanying appropriations legislation). Cause of Action, slip op. at 6.
The dispute at issue in Cause of Action focused on DOJ’s treatment of four Questions for the Record (“QFR”) documents. Slip op. at 7-9. Each QFR document contained questions posed by members of Congress and, for three of the documents, DOJ’s corresponding answers. As the D.C. Circuit explained, “Each document is self-contained, with a single, overarching heading identifying the contents of the document. The questions and answers in each document are consecutively numbered, and all but one of the documents has consecutively numbered pages.” Id. at 2. DOJ treated the QFR documents analogously to the type of omnibus diplomatic cables described above. In other words, rather than transmitting answers separately to each question coming from the various committee members separately, DOJ claimed that it had consolidated the answers into one combined response encompassing all the questions. DOJ responded to the FOIA request by providing each QFR in its entirety, but redacting all question-and-answer pairings that it considered unrelated to the grants made by the three DOJ components and Executive Order 13457. See, slip op. at 2-3.
Cause of Action challenged the redactions as inconsistent with the D.C. Circuit’s law regarding the permissible scoping of documents. Slip op. at 3. It also asserted a policy-and-practice challenge to OIP’s January 2017 guidance document discussing scoping, the currently governing guidance on the subject. Id. The District Court largely upheld OIP’s approach regarding the segmentation of the QFRs. It then held that Cause of Action lacked standing to pursue a policy-and-practice challenge. Slip op. at 3; Cause of Action Inst. v. DOJ, 453 F. Supp. 3d. 368 (D.D.C. 2020).
B. The D.C. Circuit’s Decision – Judge Edward’s Opinion for the Court
The D.C. Circuit rejected the OIP’s approach to the QFR documents, but refused to adjudicate Cause of Action’s policy-and-practice claim, finding it insufficiently ripe for adjudication. Judge Harry Edwards authored the panel opinion.
With respect to the QFR documents the Court reiterated AILA’s basic approach. If a memorandum/communication is responsive to a FOIA request, it can be redacted only if information contained within the memorandum/communication falls within a FOIA exception. Slip op. at 5. In other words, an agency cannot redact portions of a responsive record solely because it views those portions as beyond the FOIA requests’ scope. Id.
The panel acknowledged that D.C. Circuit precedent, particularly AILA, provides a “range of possible ways in which an agency might conceive of a ‘record,’” Slip op. at 5 (quoting AILA) However, it considered OIP’s approach to QFR documents “an untenable application” of AILA that fell“outside the range of reasonableness.” Slip op. at 11. It noted that “OIP itself [had] treated the self-contained QFR documents as unitary ‘records,’” and had “released the documents, albeit with portions removed, as responsive to Appellant’s FOIA request.” Id.
Judge Edwards explained that DOJ had admitted that the questions and answers “were compiled into one large file for ‘efficiency.’” Slip op. at 11. Moreover, OIP itself had “released full documents, containing overarching titles and consecutive numbering, and merely redacted some questions and answers from within those documents.” Id. OIP had even, in one instance, released an entire page containing only a redacted question, an approach completely inconsistent with its position that each question-and-answer pairing constituted a distinct record. Id. Indeed, OIP’s release of a redacted page that had a response only directed at one question “indicate[d] that DOJ viewed the entire QFR document in which that question appeared as a unitary ‘record.’” Id. Given OIP’s own treatment of the QFR in the context of its FOIA response, “the question-and-answer pairings were not individual ‘records,’ but rather were items of information within a larger record.” Slip op. 11-12.
The panel then addressed Cause of Action’s policy-and-practice claim, rejecting the District Court’s conclusion that Cause of Action lacked standing as well as the argument that invalidating the redactions in the QFR documents mooted plaintiff’s broader policy-and-practice claim. Slip op. 12-14. However, it concluded that plaintiff’s challenge to the OIP Guidance was not ripe for review, employing the standard, long-established Abbott Laboratories v. Gardner, 387 U.S. 136 (1967) two-pronged test. Slip op. 14-15. Under that test courts must assess the fitness of the issue raised for judicial review and any hardship on plaintiff attendant deferring consideration of the issue.
First, the court concluded that appellant’s challenge to the OIP Guidance is unfit for judicial review. As the Court characterized the pleadings, Cause of Action was seeking a declaration that “DOJ’s alleged policy of segmenting one record into multiple records [is] unlawful under any circumstances. Slip op. at 15. (Such a claim is inconsistent with AILA, which recognized, albeit in dicta, the legitimacy of scoping in some circumstances.) In rejecting that plea, the panel observed: “We do not have sufficient confidence in our powers of imagination to affirm such a negative.” Id. Moreover, it explained, “[t]he operation of [FOIA] is better grasped when viewed in light of a particular application.” Id.
Elaborating on its conclusion that the issue was insufficiently ripe, the panel noted that “[n]o bright line rules in the OIP Guidance [had] compelled DOJ’s disputed action,” nor had OIP claimed that its Guidance held the force of law or had to be considered binding. Slip op. 15-16. Appellant’s plea that the Circuit “conclusive[ly]” “provide a workable interpretation of [FOIA’s] statutory term ‘record’ to ensure consistent application across the Executive Branch,” slip op. 16, fell on deaf ears: “We decline the invitation to unnecessarily opine on an issue with such ‘far-reaching’ implications.” Id..
Turning briefly to the second Abbott Labs factor, Judge Edwards noted that no legally cognizable “hardship” would result from the Court’s failure to address the policy-and-practice claim on the merits.” In particular, “the OIP Guidance did not require Cause of Action ‘. . . to engage in, or to refrain from, any conduct.’” Id. “The only hardship [Appellant] will endure,” having to file another suit, does not constitute hardship for ripeness purposes. Id.
C. The D.C. Circuit’s Decision – Judge Rao’s Concurrence
Judge Rao concurred separately, perhaps portending either the D.C. Circuit’s reevaluation of its law on scoping or the Supreme Court’s decision to overturn settled D.C. Circuit law. The Supreme Court has never addressed the practice of scoping. Judge Rao’s analysis perhaps suggests a tightening of the requirements for the specificity of FOIA requests, requiring requests to focus more on identifying particular documents rather than seeking all documents relating to certain subjects, i.e., containing certain information. This, of course, threatens to gut FOIA because requesters cannot realistically be expected to know the precise documents to request by title or some other identifier (other than subject matter), without being familiar with the manner in which the agency creates, names, and maintains documents. (One approach might be to specify all documents containing particular words, but this may lead to even more voluminous document production and even more wide-ranging agency responses.)
Judge Rao began her concurrence with the observation that “[a]gencies, as well as courts, have struggled with how to define a ‘record’ for the purposes of a Freedom of Information Act (‘FOIA’) request.” Slip op. at 1 (Rao, J., concurring) She attributed this difficulty to a “practical asymmetry:” while FOIA requires agencies to release records, courts permit requesters to ask for general categories of information. Id. In her view, that “practical asymmetry,” confers upon agencies “wide latitude to define records with reference to the information requested.” Id. (citing AILA).
Moreover, Judge Rao asserted, such a troublesome asymmetry does not arise from FOIA’s text or structure. Id. Indeed, in her view, FOIA’s text and structure “create a parallel procedure for the request and release of records.” Id. In particular, “[t]o receive information, a person must make a ‘request for records’—not a request for information—and the request must “reasonably describe such records.” Id. (citing 5 U.S.C. §552(a)(3)(A))(emphasis in Rao opinion). The agency in return must produce records, not information.
Judge Rao explained that this symmetrical structure works quite well with respect to some requests, namely requests that precisely identify particular documents. Slip op. at 2 (Rao, J., concurring); accord, 2017 OIP Scoping Guidance. “When a requester seeks a specific document, such as the Smith Memo, she gets that document (or at least its non-exempt portions). With other requests, however, it may be unclear what constitutes the relevant record for the purpose of release.” Id. Such an “identification problem” arises “most acutely when the requester describes information on a particular subject matter with no reference to specific documents that might include the information.” Id.
Judge Rao pounced on Cause of Action’s perhaps injudicious argument that a request seeking “a word and its definition [would make] the entire ‘dictionary … the relevant “record” for disclosure.’” Id. In her view, such a “boil-the-ocean approach to FOIA would inundate requesters with irrelevant material and burden agencies with excessive disclosures.” Id.; see 2017 OIP Guidance.
To continue her oceanic theme, Judge Rao observed that “FOIA does not allow a requester to go fishing for a file and reel in the file cabinet.” Id. Rather, “FOIA permits an agency to identify records in part based on the information requested.” Id. She considered her approach entirely consistent with D.C. Circuit precedent.
A. Judge Rao’s Concern Regarding Boiling the Ocean
Judge Rao seeks to mount a challenge to the current approach to assessing the adequacy of FOIA requests as well as the practice of scoping, viewing the two as related. Such an approach is not compelled by FOIA’s text and structure. But is she correct that the current approach to the adequacy of requests and the limited discretion regarding scoping, in conjunction, require agencies to adopt a “boil the ocean” approach that promises to “inundate requesters with irrelevant material and burden agencies with excessive disclosures”? See, slip. op. at 2 (Roa, J,, concurring),
Receiving too many records is hardly a common complaint among FOIA requesters. Rather, the more typical complaint is receiving too few. The cases challenging the adequacy of agencies’ searches and agencies failure to locate responsive documents are legion. It is a commonplace that civil litigants bury their opponents with documents in responding to discovery requests to make identification of particularly damaging documents difficult. However, FOIA provides informal means for narrowing requests to focus on the documents the requester really wants. 5 U.S.C. §552(a)(6)(B)(ii). And it is in the agency’s interest to do so because it will reduce search and duplication costs and reduce the need to redact documents. See 2017 OIP Scoping Guidance, supra. Moreover, requesters may be able to cut through a mass of documents to find the documents they seek by employing techniques that enable searches of a large set of documents for occurrences of particular names, places, or words.
Perhaps the biggest concern regarding excessive production of documents are the attendant delays and costs, If the requester is paying full duplication costs, and 75% of 10,000 pages of records are irrelevant, that will be a steep price for the requester to pay, and a waste of both the requester’s money and the agency’s time. And such an event will not only delay the response to the requester’s own request, but to requests of others in the queue. And unlike receiving too many documents, delay is a frequent complaint of FOIA requesters. But redactions of irrelevant portions of memos and communications is hardly necessary with regard to the QFR’s that OIP sought to treat as separate documents in Cause of Action.
In short, receiving too many documents could be a detriment of severely limiting agency scoping of responsive records, but it is likely less serious than the problems resulting in too few documents being released.
But what about the harm to agencies from excessive disclosure of its records? At one level the concept of excessive disclosure is inconsistent with the whole emphasis of FOIA, which is that the release of records is to be encouraged unless their release will cause certain types of affirmative harms, reflected in FOIA’s exemptions. This approach has been confirmed by codification of the “reasonable harm” standard for documents falling within FOIA’s exemptions, requiring release of even material protected by an exemption if its release will not foreseeably cause the type of harm the exemption was designed to protect against. 5 U.S.C. §552(a)(8)(A)(i); see, Sen. Rpt 114-4, 114th Cong. 1st.Sess., 7-10 (Feb. 23, 2015)(explaining intended application of the “foreseeable harm” standard). So, if information is subject to release under FOIA, why should an agency consider itself harmed by release of information that fits within the scope of a proper request for records (and falls outside FOIA’s exemptions)? FOIA is not intended to shield agencies from embarrassing disclosures. Nor is it designed to enable agencies to hold back records that may reveal otherwise unknown matters ad prompt further FOIA requests.
The answer to this challenge could be in the burden of reviewing and redacting any document that the agency produces. Before releasing a document, even a non-responsive document or portion of a document, the agency must assure itself that it has redacted all information covered by an exemption (and performed the “foreseeable harm” analysis). The cost of such efforts is borne by the agency. It may be worth requiring an agency to undertake such obligations when the requester has sought, or at least desires, the information. However, requiring the agency to undertake such effort is of far more dubious value when the requester lacks such interest.
Nevertheless, neither agencies nor OIP have made much of a case that the provision of documents meeting requesters search parameters but also containing entirely independent discussion of matters beyond the scope of the request impose excessive burdens on government agencies. It is not clear how often agencies engage in scoping. The very limited scoping caselaw does not suggest that agencies use the approach with great frequency, or at least not in circumstances in which requesters object. The limitation on scoping would have to prove much more burdensome on agencies, in terms of additional analysis of irrelevant portions of responsive material to identify material that merits redaction based on FOIA exemptions, before allowing liberal scoping. Even if such a policy argument could be made, the tenuous statutory basis for scoping would remain an impediment.
Addressing Judge Rao’s concerns by increasing the standards for specificity of requests also seems unwise. As suggested above, it will be difficult for members of the public outside the agency to specify by title or some other identification specific documents they seek when they seek to determine what the government is up to with regard to a particular activity or subject. The current standard seems to provide requesters the flexibility they need, while permitting the government to refuse to process requests that are so vague tthat he agency cannot properly develop a search plan to identify responsive documents. 5 U.S.C. § 552(a)(3) (“reasonably describe”).
And agencies receive further accommodation in terms of the forgiving standard for the adequacy of searches, which is not based on the search results, but the agency’s search methodology. “[T]he adequacy of a FOIA search is generally determined not by the fruits of the search, but by the appropriateness of the methods used to carry out the search.” Steinberg v. Dep’t of Justice, 23 F.3d 548, 551 (D.C.Cir.1994); accord Interralde v. Comptroller of Commerce, 315 F.3d 311, 315 (D.C. Cir. 2003). Thus searches may be ruled adequate even when they fail to identify documents known to have existed. Steinberg, supra, 315 F.3d at 315.
B. The Diplomatic Cable Analogy
I began this post with the example of an omnibus diplomatic cable, combining a number of distinct message destined for distinct offices and personnel in an Embassy into one communication. The Court seemed to reject OIP’s proffered analogy between the QFR’s and diplomatic cables out of hand. But on the surface, a similarity exists. Like omnibus diplomatic cables, the QFRs contains messages responding to requests from separate officials that have been bundled together, first by the congressional committee chairs, and then by DOJ. It hardly seems an appropriate response to the analogy to observe that QFRs are “compiled into one large file for ‘efficiency.’” Slip op. at 11; so are diplomatic cables. Nevertheless, the analogy seems a bit off. I’ll offer three bases for that intuition.
First, the organizational structures of embassies and congressional committees differ significantly. Like administrative bureaucracies, embassies feature a combination of hierarchical control and specialization. See, e.g., Victor A. Thompson, Hierarchy, Specialization, and Organizational Conflict, ADMINISTRATIVE SCIENCE QUARTERLY, Vol. 5, No. 4 485, 486, 499 (Mar., 1961). In embassies, one person sits at the top of the organizational structure and there is a division of labor within the organization. Congressional committees do not have hierarchical control in the same sense, but are essentially multi-member headed bodies. And neither do congressional committees (or in some cases subcommittees) mimic the division of labor typical of agencies. Every member of the committee has responsibility for every matter that comes before the committee. Thus an answer to one committee member’s’ question falls within the responsibility of every other member. A message from the State Department to an embassy official dealing with agricultural aid lies outside the responsibility of embassy officials addressing military concerns, or processing applications for visas.
Second, we might not wish to have diplomatic cables released indiscriminately, even if involving matters left unshielded by FOIA’s exemptions. Confidentiality in the conduct of diplomatic affairs is much more critical than confidentiality in relations between executive branch agencies and Congress. In the latter, the default should be on making information public if it falls within the scope of a FOIA request, even if it is unclear whether the requester really desires the information. In the former, perhaps there should be such much weaker default rule in favor of disclosure.
Third, the chairperson of the congressional committees bundled the questions at issue in Cause of Action, and so perhaps the chairperson’s decision on how the request should be structured should drive the determination of whether the responsive document should be considered one document or multiple ones. For instance, had committee members transmitted their questions independently and at different times and DOJ had chosen to consolidate the requests for efficiency in transmitting the answers to the various committee members, there might be an argument that each separate question-and-answer pair constituted a separate document that had been artificially consolidated with others for ease of transmission, But, suffice it to say, that is not what happened with these QFRs.
C. A Potential Alternative Approach to Assessing Scoping
Ryan P. Mulvey & James Valvo have written an earlier post on the D.C. Circuit’s decision in Cause of Action: Towards a Definition of a FOIA “Record”: The D.C. Circuit’s Decision in Cause of Action Institute v. Department of Justice, NOTICE & COMMENT (June 3, 2021). I agree with many of the observations they make in their very informative posting.
However, they read Cause of Action to hold that “the maintenance of a ‘record’ and the agency’s treatment of it prior to submission of a FOIA request is the key consideration for whether records can be segmented into discrete units (or separate “records”) for the sole purpose of bypassing FOIA’s disclosure obligations.” I see no evidence in the passage they cite nor in the remainder of the opinion for that proposition. Indeed, the Court appears to be discussing nothing other than DOJ’s treatment of the QFRs in its FOIA response, more particularly the inconsistency of the DOJ’s FOIA response with the premise that each question-and-answer pairing constitutes a distinct document. (Though the Court also seems to be referencing the structure of the disputed memo/communication itself, e.g., its pagination, consecutive numbering of question-and-answer pairings, and titling of the document.)
Thus, the Court’s decision may provide a reason for concern. It may well encourage the government to be less forthcoming regarding the matter it is deleting from a memo or communication by way of scoping. The more forthcoming the agency is regarding its precise deletions, the more it undermines the agencies claims that one memo or communication actually contains multiple documents. Should DOJ have been encouraged to delete the page concerning solely an unresponsive question-and-answer pairing without any indication of the deletion?
But regardless of whether or not the Court embraced the approach Malvey and Valvo suggest that it did, is focusing on the Government’s treatment of a memo/communication before receiving the FOIA request, a sensible approach? It may well be a promising one that would help cabin agency discretion in scoping documents.
But pursuing that approach may prove challenging in practice. One complication is that many memos and communications will probably be kept in different forms. One might imagine that an omnibus diplomatic cable would be kept as a whole by the Embassy (in maintaining a record of all cables) but that only parts of the communication would be forwarded to or kept by particular offices or officials who are intended as its ultimate recipients. Is one office’s treatment of the communication to take precedence over the other? Or will the answer depend upon the particular records custodian to whom the FOIA request is directed by the requester.
This is by no means a complete analysis of the approach suggested by Mulvay & Valvo (or perhaps more accurately, the approach they applaud the Cause of Action Court for embracing), and perhaps the problem raised above can be resolved.
Scoping should probably be permissible in some limited circumstances, such as that presented by the omnibus diplomatic cable. However, the practice can very easily be abused, and the cost of the judicial policing of agencies employment of the technique might well prove high, creating yet more FOIA litigation. While the D.C. Circuit is likely to continue to limit scoping, a counter-attack may be forming along the lines sketched out by Judge Rao in her concurrence. The Cause of Action panel seemed wise to avoid eliminating scoping in all circumstances, and in refusing to take on the task of specifying the circumstances in which scoping is appropriate. In any event, Cause of Action clearly does not resolve the issues involving scoping and does not dramatically alter the approach laid out in AILA.
 OIP explained:
“It is not uncommon for both agency files . . . to deal with more than a single subject, possibly even a range of different subjects. In many instances, the multiple subjects . . . will be related in some substantive way, which can bring them all within a requester’s evident scope of interest for a given FOIA request. In other instances, however, there might be no connection between the subjects other than that the agency chose as a matter of administrative convenience to combine them together in a single document, possibly a lengthy one.” (Emphasis added.)
 See note 5 infra.
 For a recent discussion of earmarks, see Congressional Research Service, House Committees Release Guidance for Transportation Earmarks (April 8, 2021) (discussing the end of the congressional prohibition of earmarks and describing new formalized procedures for proposing committee earmarks in transportation bills).
 Production of the accompanying cover letters was also in dispute.
 The argument has particular purchase because OIP’s guidance governs not only its own FOIA practices, but the FOIA practices of agencies across the federal government. OIP was created in 1978 to coordinate agency efforts government-wide to comply with FOIA. Department of Justice, Final Rule: Establishment of the Office of Information Law and Policy, 43 Fed. Reg. 45991 (Oct. 5, 1978) (OIP created to “to advise this Department and other departments and agencies on all questions of policy, interpretation, and application of the Freedom of Information Act”; see Department of Justice, Management Division, ORGANIZATION, MISSION AND FUNCTIONS MANUAL: OFFICE OF INFORMATION POLICY (last accessed June 12. 2021) (“The mission of OIP is to provide legal and policy advice to all agencies on administration of the FOIA. OIP is responsible for encouraging agency compliance with the law and for overseeing agency implementation of it.”). Even before OIP;s creation, the Department of Justice played a major role in providing guidance to agencies on the meaning of the Freedom of Information Act. JAMES T. O’REILLY, 1 FEDERAL INFORMATION DISCLOSURE § 3:4 (accessed via westlaw).
 This occurred recently in the context of exemption 4, with a D.C. Circuit dissent ultimately presaging the Supreme Court’s reversal of settled D.C. Circuit law, see, Food Marketing Institute v. Argus Leader, 139 S. Ct. 2356 (2019).
 Indeed, recently the Third Circuit held it “counterintuitive in the extreme;” those outside an agency should be expected “to have sufficient knowledge of an agency’s organizational units to be able to” be able to specify a particular component of an agency that held a desired record. Biear v. Attorney General, 905 F.3d 151 (3rd Cir. 2018).
 Biear v. Attorney General, 905 F.3d 151 (3rd Cir. 2018).
 While true, the difficulties the courts have encountered in applying the statutory term “records” are largely unrelated to the practice of scoping, Rather they relate to the agencies obligations regarding documents in possession of consultants or contractors, see, Forsham v. Harris, 445 U.S. 169 (1980), documents createdand maintained by government officials that may be considered personal records, Kissinger v. Reporter’s Committee for Freedom of the Press, 445 U.S. 136 (1980), and records agencies hold at the behest of Congress or the White House, see, e.g., United We Stand Am., Inc. v. IRS, 359 F.3d 595 (D.C.Cir. 2004)(congressional records) Judicial Watch v. U.S. Secret Service, 726 F.3d 208 (D.C. Cir, 2013)(White House visitors’ logs) .
 The critical aspect to specifying the production or records rather than information was Congress’ desire to limit government agencies obligations to releasing materials it already possessed, i.e., records, and not require agencies to craft new documents to answer questions posed by the public, i.e., providing information, Forsham v. Harris, 445 U.S. 169, 186 (1980); NLRB v. Sears, Roebuck & Co., 421 U.S. 132, 161–62, (1975); see, Rubman v. U.S. Citizenship & Immigration Services, 800 F.3d 381, 390-91 (7th Cir. 2015)(agency cannot avoid producing records by reviewing the document and creating a summary table to answer a FOIA request).
 If the reader has not heard this particular phrase, I had not either. For more on the phrase see F. John Reh, What the Phrase “Boil the Ocean” Means in Business (Updated November 15, 2019); David Benjamin and David Komlos, It’s Time to Retire the Phrase: “Don’t Boil the Ocean,” FORBES (Nov. 11, 2019) .
 Of course, as Judge Rao observes, when a requester seeks information related to a particular person, event, government action, or government program, the agency will have to make relevancy determinations. But it is not clear that such conclusions are entitled to deference from the court. See, 5 U.S.C. §552(a)(4)(B)(establishing de novo review standard). Moreover, nothing in FOIA suggests that any relevancy inquiry is to be made in terms of releasing a portion of a memo or communication rather than the entire memo or communication.
The Act’s focus on “records” constrains requesters as well as government agencies. The focus limits agencies to deciding whether documents are relevant, rather than focusing on the relevance of portions of documents. But FOIA’s focus on “records” also limits requesters — requesters can obtain information about “what their government is up to” only if the relevant government agency develops such information and then memorializes it in tangible form. Thus, requesters cannot use FOIA to obtain even the most important information regarding agency activities unless the agency has reduced the information to tangible form, i.e., some sort of record, and has continued to maintain it.
 The provision states:
“the agency shall notify the person making the request if the request cannot be processed within the time limit specified in that clause and shall provide the person an opportunity to limit the scope of the request so that it may be processed within that time limit or an opportunity to arrange with the agency an alternative time frame for processing the request or a modified request.”
5 U.S.C. §552(a)(6)(B)(ii) (emphasis added). Agencies must also make available a FOIA Public Liaison “[t]o aid the requester,” in resolving disputes between the requester and the agency, and notify the requester of the right to seek dispute resolution services offered by the Office of Government Information Services. 5 U.S.C. §552(a)(6)(B)(ii).
 A request need merely describe the records sought with sufficient precision to enable agency professional “familiar with the subject area of the request to locate the record with a reasonable amount of effort.” Marks v. U.S., 578 F.2d 261, 263 (9th Cir. 1978) (quoting H.R. Rep. No. 93-876, at 6 (1974)); Kenney v. U.S. Dep’t of Justice, 603 F. Supp. 2d 184, 188 (D.D.C. 2009). See, MuckRock v. CIA, 300 F. Supp. 3d 108 (D.D.C. 2018).
 In Steinberg, the Court provides several possible challenges to the adequacy of a search, such as the failure to search certain offices or files, or the failure to contact an employee who might have been helpful in finding responsive documents.
 This is not to downplay the considerable power of committee and subcommittee chairs in terms of setting the committee agenda or presiding over committee proceedings. But structurally, at least, the chair is the first among equals, rather than a superior to which the other committee members are accountable and to whose direction they are subject. Moreover, some congressional committees delegate most of their authority to subcommittees (in which case there is specialization by subcommittee).
 Even the requirement that the Government maintain a consistent position with respect to its characterization of documents in FOIA litigation can be a useful, albeit limited, tool. In a recent case a District Judge prohibited the FDA from changing its mind regarding whether a “document” constituted one or multiple documents “midway through litigation.” The Judge expressed concern that allowing an agency to do so posed too great a risk of manipulation. Judge Rothenberg Educational Center v. FDA, 376 F.Supp.3d 47, 61 (D.D.C. 2019)..
 After all, Mulvay & Valvo may well prefer that scoping not be considered legitimate at all.
 Judge Rothenberg Educational Center v. FDA provides a recent example. 376 F.Supp.3d 47 (D.D.C. 2019).