finding that agency affidavits are presumed to be made in good faithSummary of this case from Minkovski v. U.S. Dep't of Treasury
No. 15-1055-cv August Term, 2015
David E. McCraw, (Jeremy A. Kutner, on the brief), New York, NY, for Sergio Florez. Jessica Jean Hu (Christopher Connolly, on the brief), Assistant United States Attorneys, for Preet Bharara, United States Attorney for the Southern District of New York, New York, NY.
David E. McCraw, (Jeremy A. Kutner, on the brief), New York, NY, for Sergio Florez.
Jessica Jean Hu (Christopher Connolly, on the brief), Assistant United States Attorneys, for Preet Bharara, United States Attorney for the Southern District of New York, New York, NY.
Before: Straub, Livingston and Chin, Circuit Judges.
Judge LIVINGSTON dissents in a separate opinion.
Straub, Circuit Judge:
This appeal arises from a request submitted to Defendant–Appellee Central Intelligence Agency (“CIA”) by Plaintiff–Appellant Sergio Florez (“Mr. Florez”), pursuant to the Freedom of Information Act (“FOIA”). See 5 U.S.C. § 552 et seq . Mr. Florez's request, dated November 3, 2013, sought “the disclosure and release of any and all records between 1958 and 1990 related to and or mentioning [his] father, Armando J. Florez” (“Dr. Florez”). Joint App'x at 62. During the 1960s, Dr. Florez served in several high-level diplomatic roles on behalf of the Republic of Cuba, including as chargé d'affaires in Washington, D.C. He defected to the United States in 1968, became an American citizen in 1979, and died in October 2013.
A chargé d'affaires is “[a]n officer in charge of an embassy who is not an ambassador, (as when, for example, the level of relations between two states has been lowered to below the ambassadorial level) and who is accredited to the minister of foreign affairs, rather than to the chief of state.” Chas. W. Freeman, Jr., The Diplomat's Dictionary 29 (2d ed. 2010).
On November 20, 2013, the CIA answered Mr. Florez's request with a so-called Glomar response, stating that it “can neither confirm nor deny the existence or nonexistence of records responsive to [Mr. Florez's] request.” Id . at 70. It asserted that the existence or nonexistence of such records “is currently and properly classified and is intelligence sources and methods information” that is exempt from disclosure under FOIA. Id . On December 4, 2013, Mr. Florez timely filed an administrative appeal with the CIA's Agency Release Panel.
The term “Glomar response” refers to “a response that neither confirms nor denies the existence of documents responsive to the request.” Ctr. for Constitutional Rights v. CIA , 765 F.3d 161, 164 n. 5 (2d Cir.2014), cert. denied , ––– U.S. ––––, 135 S.Ct. 1530, 191 L.Ed.2d 559 (2015). The term “arises from the CIA's successful defense of its refusal to confirm or deny the existence of records regarding a ship named the Hughes Glomar Explorer in Phillippi v. Cent. Intelligence Agency , 546 F.2d 1009, 1011 (D.C.Cir.1976),” Conti v. Dep't of Homeland Sec. , No. 12–cv–5827, 2014 WL 1274517, at *3 n. 2 (S.D.N.Y. Mar. 24, 2014).
The CIA's Agency Release Panel is composed of various CIA officials and is tasked with, inter alia , rendering “final Agency decisions from appeals of initial adverse decisions under the Freedom of Information Act.” 32 C.F.R. § 1900.41(c).
On February 18, 2014, while Mr. Florez's administrative appeal was pending, Mr. Florez timely filed the underlying action. The CIA and Mr. Florez filed cross-motions for summary judgment. On April 22, 2014, while the motions underwent briefing, the Agency Release Panel denied Mr. Florez's administrative appeal.
“The exhaustion of administrative remedies by plaintiff is not at issue here. Since the CIA did not respond to plaintiff's appeal within 20 days, he is ‘deemed to have exhausted his administrative remedies with respect to such request,’ 5 U.S.C. § 552(a)(6)(C)(i), and may file suit pursuant to 5 U.S.C. § 552(a)(4)(B).” Florez v. CIA , No. 14–cv–1002, 2015 WL 728190, at *2 n. 3 (S.D.N.Y. Feb. 19, 2015) ; see also Oglesby v. Dep't of Army , 920 F.2d 57, 62 (D.C.Cir.1990).
On February 19, 2015, the United States District Court for the Southern District of New York (Sidney H. Stein, Judge ) granted summary judgment in favor of the CIA and denied Mr. Florez's cross-motion for summary judgment, holding that “the CIA's Glomar response was justified and the existence of any records is exempt from disclosure under FOIA Exemption 1 (for classified national defense or foreign policy secrets) and Exemption 3 (for matters specifically exempted from disclosure by statute).” Florez v. CIA , No. 14–cv–1002, 2015 WL 728190, at *1 (S.D.N.Y. Feb. 19, 2015). This timely appeal followed.
During the pendency of this appeal, pursuant to a separate FOIA request, the Federal Bureau of Investigation (“FBI”) released several declassified documents pertaining to Dr. Florez on June 23, 2015, and one additional such document on July 24, 2015 (collectively, “FBI Disclosures”). Mr. Florez requested that the CIA revise its response to his FOIA request in light of the FBI Disclosures. The CIA reviewed the FBI Disclosures, but declined to alter its position that a Glomar response is supportable in these circumstances. See Letter from Preet Bharara, United States Attorney for the Southern District of New York, to Catherine O'Hagan Wolfe, Clerk of Court at 1, Florez v. CIA , No. 15–1055-cv (2d Cir. Dec. 18, 2015), ECF No. 57-1 [hereinafter “CIA Ltr.”].
Mr. Florez challenges the CIA's Glomar response as inadequate under the FOIA, but we do not reach the merits of his challenge at this time. Because we find the FBI Disclosures relevant to the issues presented, we remand for the District Court to pass on the import of those documents in the first instance.
I. Standard of Review
By statute, a district court must review de novo an agency's determination to withhold information requested under the FOIA, see 5 U.S.C. § 552(a)(4)(B) ; Main St. Legal Servs., Inc. v. Nat'l Sec. Council , 811 F.3d 542, 542 (2d Cir.2016), and we subsequently review de novo the district court's ruling, see Ctr. for Constitutional Rights v. CIA , 765 F.3d 161, 166 (2d Cir.2014), cert. denied , ––– U.S. ––––, 135 S.Ct. 1530, 191 L.Ed.2d 559 (2015). “The government bears the burden of demonstrating that an exemption applies to each item of information it seeks to withhold, and all doubts as to the applicability of the exemption must be resolved in favor of disclosure.” Ctr. For Constitutional Rights , 765 F.3d at 166 (internal citations and quotation marks omitted). Such “[e]xceptions to FOIA's general principle of broad disclosure of Government records have consistently been given a narrow compass.” Id . (internal quotation marks and ellipsis omitted).
“An agency may carry its burden by submitting declarations giving reasonably detailed explanations why any withheld documents fall within an exemption, and such declarations are accorded a presumption of good faith.” Id . (internal quotation marks omitted). We find a Glomar response justified only in “unusual circumstances, and only by a particularly persuasive affidavit.” N.Y. Times Co. v. Dep't of Justice , 756 F.3d 100, 122 (2d Cir.2014) (internal quotation marks omitted); see also Halpern v. FBI , 181 F.3d 279, 295 (2d Cir.1999) (“[T]he good faith presumption that attaches to agency affidavits only applies when accompanied by reasonably detailed explanations of why material was withheld. Absent a sufficiently specific explanation from an agency, a court's de novo review is not possible and the adversary process envisioned in FOIA litigation cannot function.”).
II. FOIA Exemptions 1 and 3
On the merits, which we do not reach in this opinion, this appeal presents an ordinary Glomar inquiry: whether the existence or nonexistence of documents, within the CIA's possession and responsive to Mr. Florez's request, is itself a fact exempt from disclosure under one of two FOIA exemptions. Because the relevancy of the FBI Disclosures is determined by the scope of the claimed exemptions, we briefly describe the two exemptions at issue.
The CIA relies upon FOIA Exemptions 1 and 3 to support its Glomar response. FOIA Exemption 1 “exempts from disclosure records that are ‘specifically authorized under criteria established by an Executive order to be kept secret in the interest of national defense or foreign policy,’ and ‘are in fact properly classified pursuant to such Executive order.’ ” Ctr. for Constitutional Rights , 765 F.3d at 164 (quoting 5 U.S.C. § 552(b)(1) ). The agency asserts that the existence or nonexistence of responsive records is information properly classified pursuant to § 1.1(4) of Executive Order 13,526, which permits classification of information that, if disclosed, “reasonably could be expected to result in damage to the national security, which includes defense against transnational terrorism, and the original classification authority is able to identify or describe the damage.” 75 Fed. Reg. 707, 707 (Dec. 29, 2009).
This excerpt states one of the four requirements for classification pursuant to Executive Order 13,526. Mr. Florez does not dispute that the other three requirements for classification are met: “(1) an original classification authority is classifying the information; (2) the information is owned by, produced by or for, or is under the control of the United States Government; (3) the information falls within one or more of the categories of information listed in section 1.4 of this order,” 75 Fed. Reg. 707, 707 (Dec. 29, 2009), one of which is the “intelligence activities (including covert action) [and] intelligence sources or methods” category, id . at 709.
FOIA Exemption 3 “permits the Government to withhold information from public disclosure provided that: (1) the information is ‘specifically exempted from disclosure by statute’; and (2) the exemption statute ‘requires that the matters be withheld from the public in such a manner as to leave no discretion on the issue’ or ‘establishes particular criteria for withholding or refers to particular types of matters to be withheld.’ ” ACLU v. Dep't of Justice , 681 F.3d 61, 72 (2d Cir.2012) (quoting 5 U.S.C. § 552(b)(3) ). Here, the CIA invokes Section 6 of the CIA Act of 1949, 50 U.S.C. § 3507 (exempting the CIA from any law that “require[s] the publication or disclosure of the organization, functions, names, official titles, salaries, or numbers of personnel employed by the Agency”), and Section 102A(i)(l) of the National Security Act of 1947, 50 U.S.C. § 3024(i)(1) (protecting “intelligence sources and methods from unauthorized disclosure”).
III. FBI Disclosures
During the pendency of this appeal, the FBI Disclosures were brought to our attention by Mr. Florez and submitted to us by the CIA. Our threshold inquiry in this appeal is whether the FBI Disclosures should be considered in adjudicating this case. Mr. Florez, on one hand, urges us to consider the documents ourselves and conclude that “the FBI release thoroughly undermine[s] the CIA's position.” Letter from David E. McCraw, Attorney for Sergio Florez, to Catherine O'Hagan Wolfe, Clerk of Court at 1, Florez v. CIA , No 15–1055–cv (2d Cir. Dec. 18, 2015), ECF No. 53. The CIA, on the other hand, urges us to evaluate its response to Mr. Florez's FOIA request “as of the time it was made and not at the time of [our] review,” CIA Ltr. at 1 (internal quotation marks omitted), and therefore ignore the FBI Disclosures in our analysis of the merits. If, however, we “determine that the FBI disclosures are relevant to the issues on appeal, the CIA respectfully requests that this case be remanded to the district court to allow the CIA to submit additional declarations addressing the FBI disclosures.” Id . at 3 n. 2. Accordingly, we first address whether the FBI Disclosures are relevant to the case. Answering that question in the affirmative, we next consider whether we should ignore the disclosures—resolving the merits based on the record as it existed at the time of Mr. Florez's FOIA request—or remand the case to allow the District Court to weigh the significance of the documents in the first instance. Our precedent, judicial efficiency, and common sense all militate towards the latter and we therefore remand the case to the District Court.
Due to issues of timing, the District Court below never had the opportunity to weigh the significance of the FBI Disclosures and, accordingly, on appeal, “we lack the benefit of an evaluation of this issue by the district court.” Official Comm. of Unsecured Creditors of WorldCom, Inc. v. SEC , 467 F.3d 73, 79 (2d Cir.2006) (Sotomayor, J. ). We therefore limit our inquiry at this juncture to whether the documents are relevant to the merits of this appeal, such that we need consider whether to accord the District Court the opportunity to assess them in the first instance. See Eric M. Berman, P.C. v. City of New York , 796 F.3d 171, 175 (2d Cir.2015) (per curiam) (“[I]t is this Court's usual practice to allow the district court to address arguments in the first instance.” (internal quotation marks omitted)); see also, e.g. , United States v. Salameh , 152 F.3d 88, 159 (2d Cir.1998) (per curiam) (concluding that the “best solution” for evaluating “newly discovered evidence” was for such evidence to “be addressed ... by the district court first”), cert. denied , 526 U.S. 1044, 119 S.Ct. 1345, 143 L.Ed.2d 508 (1999).
“Evidence is relevant if: (a) it has any tendency to make a fact more or less probable than it would be without the evidence; and (b) the fact is of consequence in determining the action.” Fed. R. Evid. 401. See also United States v. Certified Envtl. Servs., Inc. , 753 F.3d 72, 90 (2d Cir.2014) (“[T]he definition of relevance under Fed. R. Evid. 401 is very broad.”). More briefly stated, evidence is relevant if it has “appreciable probative value.” Relevant , Black's Law Dictionary (10th ed. 2014); see also United States v. Litvak , 808 F.3d 160, 179–80 (2d Cir.2015). “Relevance is a legal determination.” United States v. Staniforth , 971 F.2d 1355, 1358 (7th Cir.1992) (Posner, J. ) abrogated on other grounds by United States v. Wells , 519 U.S. 482, 117 S.Ct. 921, 137 L.Ed.2d 107 (1997). Conversely, it is “the province of the finder of fact to determine what weight to accord” such relevant evidence. Jim Beam Brands Co. v. Beamish & Crawford Ltd. , 937 F.2d 729, 736 (2d Cir.1991). Accord Barefoot v. Estelle, 463 U.S. 880, 898, 103 S.Ct. 3383, 77 L.Ed.2d 1090 (1983) (“[T]he rules of evidence generally extant at the federal and state levels anticipate that relevant, unprivileged evidence should be admitted and its weight left to the factfinder, who would have the benefit of cross-examination and contrary evidence by the opposing party.”).
We conclude that the FBI Disclosures are relevant to the present Glomar inquiry. At minimum, the FBI Disclosures are germane to the CIA's asserted rationale for asserting a Glomar response, which is that confirming the existence or non-existence of responsive records would confirm either the Agency's interest or disinterest in Dr. Florez as an intelligence asset. Specifically, the documents appear to include the following revelations: (1) the FBI investigated Dr. Florez's background and tracked his career development, official activities, and international relocations, see, e.g. , CIA Ltr., Ex. A at 13, 27, 32, 36, 45–46; id . Ex. B at 1; (2) the FBI cultivated informants in order to obtain information concerning Dr. Florez, including material which pertained to both his professional and personal conduct, see, e.g. , id . Ex. A at 6, 16–17, 19, 21, 23, 29, 35, 45–46; and (3) several other government departments and agencies provided to or received from the FBI information concerning Dr. Florez, see, e.g. , id . Ex. A at 25 (Department of State; Immigration & Naturalization Service (“INS”)), 27–28 (Department of State), 34–35 (Department of State; Navy's Office of Naval Intelligence (“ONI”); Air Force's Office of Special Investigations (“OSI”); Army's Chief of Staff for Intelligence (“ACSI”)), 47 (same); id . Ex. B at 1 (INS).
These public disclosures have appreciable probative value in determining, under “the record as a whole,” “whether the justifications set forth in the [CIA's] declaration are logical and plausible in this case.” Ctr. for Constitutional Rights , 765 F.3d at 168. The CIA's declaration relies heavily on the import of masking the government's intelligence interest (if any) in Dr. Florez and in maintaining complete secrecy as to whether any intelligence activities were focused on him. Though the FBI Disclosures do not reveal the CIA's activities or involvement, they appear to suggest that multiple government departments and agencies were investigating, monitoring, and had an intelligence interest in Dr. Florez, and that the FBI cultivated informants to gather information about him. This now-public information may bear on the CIA's position that the mere acknowledgement that it does or does not have possession of documents that reference Dr. Florez would harm the national security, or otherwise disclose Agency methods, functions, or sources.
Although the release of information from a third party agency may more directly bear upon whether another agency's revelation of the existence or nonexistence of records “reasonably could be expected to result in damage to the national security,” 75 Fed. Reg. 707, 707 (Dec. 29, 2009), the determinative consideration under Exemption 1, there is no obvious reason why such third party disclosures cannot, in certain instances, likewise permit an inference contradicting an asserting agency's Glomar response under Exemption 3. Indeed, in this case, beyond a separate boilerplate recitation of the Exemption 3 framework, see Joint App'x at 54-56, the CIA has proffered a single general rationale with respect to both Exemptions. See Joint App'x at 43-46. We leave it to the District Court, in the first instance, to assess “on the whole record” whether, in light of the recent FBI Disclosures, this rationale “objectively survives the test of reasonableness, good faith, specificity, and plausibility.” Gardels v. CIA , 689 F.2d 1100, 1105 (D.C.Cir.1982).
The Dissent disagrees that the FBI Disclosures even might bear on the sufficiency of the CIA's Glomar response and therefore concludes that we should deem those disclosures irrelevant. See Dissent at 3–11. With respect to our dissenting colleague, we simply diverge on this point. Whereas the Dissent casually dismisses the FBI Disclosures as “disclos[ing] little regarding Dr. Florez,” see Dissent at 192, we believe the documents contain disclosures that bear upon whether the CIA is able to carry its burden in this case, to wit, “whether on the whole record the Agency's judgment objectively survives the test of reasonableness, good faith, specificity, and plausibility.” Gardels v. CIA , 689 F.2d 1100, 1105 (D.C.Cir.1982).
The Dissent attempts to downplay the scope of the FBI Disclosures at every turn, shrugging them off as revealing nothing more than “that the FBI maintained some interest in Dr. Florez for some period of time.” See Dissent at 195. But to be clear, the disclosures in fact reveal a wealth of information, including that the FBI maintained an active interest in Dr. Florez for well over a decade, tracking the development of his career, including his foreign deployments and travel on behalf of the Cuban government in Europe, documenting his familial affiliations and personal affairs, and taking a particular interest in the fact he had grown disillusioned with the communist regime in Cuba and contemplated defection to the United States. These findings were shared with a myriad of other agencies, including, inter alia , the Department of State, the Navy's Office of Naval Intelligence, the Army's Chief of Staff for Intelligence, and the Air Force's Office of Special Investigations, in addition to being circulated to the FBI's Foreign Liaison Desk and FBI legal attachés in Madrid and Paris. We believe this information would be probative to a reasonable finder of fact attempting to discern the “reasonableness, good faith, specificity, and plausibility” of the CIA's current Glomar response.
The FBI Disclosures sometimes refer to these legal attachés as “legats.” See, e.g. , FBI Disclosures at 40.
Rather than grapple with the full scope of the disclosures, the Dissent fixates on the fact that the disclosures emanate from the FBI, rather than the CIA. Although apparently hesitant to plainly say as much, the Dissent essentially argues that, under the official acknowledgment doctrine, the disclosures of other federal agencies—regardless of the extent to which they bear on the validity of another agency's Glomar rationale—are never relevant and must be wholly disregarded. See Dissent at 195–96 & n. 7. This conclusion confuses the act of waiver—which we uniformly recognize as a privilege reserved to the agency asserting a Glomar response—with an agency's independent obligation to “carry its burden by submitting declarations giving reasonably detailed explanations why any withheld documents fall within an exemption,” Ctr. For Constitutional Rights , 765 F.3d at 166 (internal quotation marks omitted), a burden which can only be carried when the agency's declarations “are not called into question by contradictory evidence in the record.” Elec. Privacy Info. Ctr. v. Nat'l Sec. Agency , 678 F.3d 926, 931 (D.C.Cir.2012). In other words, a third party agency's disclosures cannot waive the asserting agency's right to a Glomar response, but such disclosures may well shift the factual groundwork upon which a district court assesses the merits of such a response.
Indeed, the Dissent appears to recognize as much, repeatedly explaining why these particular disclosures supposedly do not bear upon the CIA's Glomar rationale. For instance, the Dissent touts the fact that the FBI Disclosures “do not even mention the CIA,” Dissent at 191, or “even discuss the CIA or its activities,” Dissent at 193. Accordingly, the “documents here ... are simply not helpful in assessing the logic and plausibility” of the Glomar response at issue. Dissent at 192. See also Dissent at 194– 95 (“[T]hese documents ... are simply not relevant to the question whether the CIA's justification for its Glomar response in this case is plausible and makes sense” (emphasis in original)). This tacit acknowledgment that certain disclosures from a third party agency might weigh on the validity of a Glomar response (i.e. , those expressly referring to the asserting agency or its activities) makes all the more puzzling both the Dissent's refusal to allow the District Court to weigh the facts in the first instance (as the Dissent itself has plainly done) and its insistence on propagating a per se rule barring consideration of third party disclosures on the sufficiency of an agency's Glomar response.
The official acknowledgment doctrine prohibits agencies from “provid[ing] a Glomar response when the existence or nonexistence of the particular records covered by the Glomar response has been officially and publicly disclosed.” Wilner , 592 F.3d at 70. This waiver is limited only to official and public disclosures made by the same agency providing the Glomar response, and therefore does not “requir[e] [the agency] to break its silence” as a result of “statements made by another agency.” Frugone v. CIA , 169 F.3d 772, 775 (D.C.Cir.1999) (refusing to “treat the statements of the [Office of Personnel Management] ... as tantamount to an official statement of the CIA”).
But we do not impute the FBI's decision to disclose information about Dr. Florez to the CIA, or suggest that the FBI Disclosures necessarily preclude the CIA's right to assert a Glomar response. See Wilson v. CIA , 586 F.3d 171, 187 (2d Cir.2009) (declining to “infer official disclosure of information classified by the CIA from ... [the] release of information by another agency”). Rather, we simply conclude that the FBI Disclosures are relevant evidence—unavailable to the District Court at the time of its initial decision—bearing upon the sufficiency of the justifications set forth by the CIA in support of its Glomar response. Put simply, the official acknowledgment doctrine has no impact on our opinion in this case. The Dissent's exclusive reliance on the official acknowledgement doctrine, to create out of whole cloth a rule limiting the evidence a district court may consider in a Glomar inquiry, only serves to demonstrate that it is promoting a solution of no applicability to this case.
The Dissent attempts to extrapolate this limitation from the “animating principles” of the official acknowledgment doctrine, which, it contends, bar the “ ‘anomalous result’ that disclosures by one agency could open the door to compelled disclosure by another.” See Dissent at 196 n. 7 (quoting Frugone , 169 F.3d at 775 ). Suffice to say, we do not believe that allusion to the “animating principles” of an inapplicable doctrine serves to vindicate the Dissent's proposed rule. Moreover, as already explained, this categorical limitation is at odds with the Dissent's own acknowledgment that certain agency disclosures might, in some hypothetical instances, lead to disclosure by another agency. See, e.g. , Dissent at 195.
Rather than place an arbitrary limitation on the range of evidence a district court may consider in assessing the sufficiency of an agency affidavit filed in support of a Glomar response, our cases instruct that we accord “substantial weight” to such affidavits, but only “provided [that] the justifications for nondisclosure are not controverted by contrary evidence in the record ....” Wilner , 592 F.3d at 68 (citation omitted). See also Elec. Privacy Info. Ctr. , 678 F.3d at 931 (district courts, in Glomar case, may grant summary judgment on the basis of agency affidavits “if they are not called into question by contradictory evidence in the record”). Indeed, categorically excluding public documents as evidence when reviewing a Glomar response flies in the face of our own instruction that “[i]n evaluating an agency's Glomar response ... [t]he court should attempt to create as complete a public record as is possible.” Wilner , 592 F.3d at 68 (internal quotation marks omitted). It defies reason to instruct a district court to deliberately bury its head in the sand to relevant and contradictory record evidence solely because that evidence does not come from the very same agency seeking to assert a Glomar response in order to avoid the strictures of FOIA. See Gardels , 689 F.2d at 1105 (“The test is not whether the court personally agrees in full with the CIA's evaluation of the danger—rather, the issue is whether on the whole record the Agency's judgment objectively survives the test of reasonableness, good faith, specificity, and plausibility in this field of foreign intelligence in which the CIA is expert and given by Congress a special role.” (emphasis added)).
Having determined that the FBI Disclosures are relevant to the merits of this case, we briefly address why remand is appropriate in this instance.
Although we conclude that the FBI documents are relevant to this case, our precedent nonetheless permits us to set aside the FBI Disclosures by adhering to the so-called “general rule” that “a FOIA decision is evaluated as of the time it was made and not at the time of a court's review,” N.Y. Times , 756 F.3d at 110 n. 8. In our view, however, departure from that practice is warranted in this instance. Indeed, having now found the FBI Disclosures relevant to the sufficiency of the asserted Glomar rationale, the CIA asks us to proceed in this precise manner. See CIA Ltr. at 3 n.2.
The agency unambiguously stated that if we “were to determine that the FBI disclosures are relevant to the issues on appeal,” it would “request[ ] that this case be remanded to the district court to allow the CIA to submit additional declarations addressing the FBI disclosures.” See CIA Ltr. at 3 n. 2. Having so determined, and for the additional reasons detailed herein, we accede to the agency's request.
First, the policy underpinning the general rule does not apply here. As one of our sister circuits has explained, “[t]o require an agency to adjust or modify its FOIA response based on post-response occurrences could create an endless cycle of judicially mandated reprocessing each time some circumstance changes.” Bonner v. Dep't of State , 928 F.2d 1148, 1152 (D.C.Cir.1991). This case features no such judicially mandated reprocessing, as Mr. Florez, not the courts, requested that the CIA modify its FOIA response in light of the FBI Disclosures. Without being directed to do so, the CIA then reviewed those documents and determined not to alter its response to Mr. Florez's FOIA request after due consideration. See CIA Ltr. at 1 (“The CIA has reviewed [the FBI Disclosures] and informed [Mr.] Florez's counsel that they do not alter ... the CIA's response ....”). Thus, the CIA has already voluntarily undertaken and completed any “reprocessing” that could arise from remanding the case to allow the District Court to consider the FBI Disclosures.
Indeed, that is likely why the agency itself asks us to remand the case to the District Court, if we find the FBI Disclosures relevant to the issues on appeal. See CIA Ltr. at 3 n. 2. Although the CIA has indicated that it may wish to submit additional declarations in support of its Glomar response on remand, such litigation is distinct from an agency's often-lengthy processing of FOIA requests. See Adam Cohen, The Media That Need Citizens: The First Amendment and the Fifth Estate , 85 S. Cal. L. Rev. 1, 64 (2011) (describing agencies' “generally lengthy processing delays” of FOIA requests).
Second, consideration of the FBI Disclosures is “the most sensible approach” in this case. N.Y. Times , 756 F.3d at 110 n. 8 (rejecting the government's argument “that we cannot consider any official disclosures made after the District Court's opinion,” and concluding that “[t]aking judicial notice” of “ongoing disclosures by the Government made in the midst of FOIA litigation” is “the most sensible approach”); accord ACLU v. CIA , 710 F.3d 422, 431 (D.C.Cir.2013) (taking notice of statements post-dating district court's grant of summary judgment). If we were to proceed to the merits, and therefore potentially affirm the decision of the District Court, without considering the FBI Disclosures, we would accomplish little more than consign Mr. Florez to filing a fresh FOIA request, beginning the process anew with the FBI Disclosures in hand. Such an outcome makes little sense and would merely set in motion a multi-year chain of events leading inexorably back to a new panel of this Court considering the precise question presented here, perhaps in about two-and-a-half years (Mr. Florez's FOIA request was submitted in November 2013). This delay would not serve the purposes of FOIA or the interests of justice, and it is inefficient to send Mr. Florez back to the end of the line. Because the CIA has already reviewed the FBI documents and reaffirmed its Glomar response, we know how the CIA would respond to Mr. Florez's renewed request, and there is no reason we should not take into account the reality in which this action proceeds.
See Staff of H. Comm. on Oversight and Gov't Reform, 114th Cong., FOIA Is Broken: A Report 36 (2016) (“Delays waste time and taxpayer money.”); Michael E. Tankersley, How the Electronic Freedom of Information Act Amendments of 1996 Update Public Access for the Information Age , 50 Admin. L. Rev. 421, 425 (1998) (“Despite the intention of the FOIA, the public's access to government information is inefficient, ineffective, and costly.” (internal quotation marks omitted)).
Cf. Andrew Christy, The ACLU's Hollow FOIA Victory Over Drone Strikes , 21 Geo. Mason L. Rev. 1, 8 (2013) (noting a “startling disconnect between reality and the law in Glomar cases”).
Third, remanding the case is in keeping with our general policy that the trial court should consider arguments—and weigh relevant evidence—in the first instance. See Quinones on Behalf of Quinones v. Chater , 117 F.3d 29, 36 (2d Cir.1997) (“As the [factfinder] did not address this evidence, we think it best to remand the case so that he can consider in the first instance what weight to accord it.”); Eric. M. Berman, P.C. , 796 F.3d at 175 ; Salameh , 152 F.3d at 159 ; accord Sprint/United Mgmt. Co. v. Mendelsohn , 552 U.S. 379, 387, 128 S.Ct. 1140, 170 L.Ed.2d 1 (2008) (“Rather than assess the relevance of the evidence itself and conduct its own balancing of its probative value and potential prejudicial effect, the Court of Appeals should have allowed the District Court to make these determinations in the first instance, explicitly and on the record.”); Pullman–Standard v. Swint , 456 U.S. 273, 293, 102 S.Ct. 1781, 72 L.Ed.2d 66 (1982) (chastising the Court of Appeals for “fail [ing] to remand for further proceedings” after determining that “the District Court had failed to consider relevant evidence” and instead making “its own determinations”).
In sum, proceeding to decision while willfully ignoring relevant materials would breed judicial inefficiency and produce an outcome contrary to that which might result from consideration of additional materials that—through no fault of Mr. Florez's—were unavailable to him at the time the FOIA request was made.
C. Jacobson Remand
In the interests of judicial economy and orderly resolution of this matter, we find prudent a limited remand to the District Court pursuant to our practice under United States v. Jacobson , 15 F.3d 19, 22 (2d Cir.1994). The remand permits the District Court to reconsider its prior conclusion in light of the FBI Disclosures, and any additional materials it permits the parties to submit, see N.Y. Times , 756 F.3d at 110 n. 8 (explaining that the panel granted “the Government's request for an opportunity to submit new material concerning public disclosures made after the District Court's decision”), and then return its determination to us for consideration without the need for a new notice of appeal, briefing schedule, and reassignment to a new panel unfamiliar with the case. See, e.g. , N.Y. State Citizens' Coal. for Children v. Velez , 629 Fed.Appx. 92, 94 (2d Cir.2015) (summary order) (remanding pursuant to Jacobson “[b]ecause th[e] issue was not raised in the district court,” and “we conclude that it should be addressed in the first instance there”); United States v. Persad , 607 Fed.Appx. 83, 84 (2d Cir.2015) (summary order) (remanding pursuant to Jacobson “to address the disputed issue of Vermont law and practice in the first instance, and to conduct any further fact-finding that may be required”); Weifang Xinli Plastic Prods. Co. v. JBM Trading Inc. , 553 Fed.Appx. 42, 44 (2d Cir.2014) (summary order) (remanding pursuant to Jacobson “for the district court to supplement the record”).
Accordingly, we remand to the District Court with instructions to enter an order stating whether its prior conclusion that the CIA adequately justified its Glomar response must be revised in light of the FBI Disclosures and any post-remand submissions. Within thirty days after such entry, either party may restore jurisdiction to this panel by filing a letter with the Clerk of this Court; this letter, not to exceed twenty double-spaced pages, should set forth the grounds for claiming error in the District Court's decision and attach a copy of the order. Upon the filing of such a letter, the opposing party may file a response of the same maximum length within fourteen days. Oral argument will be scheduled at the panel's discretion. If neither party files an initial letter within thirty days of the order's entry, appellate jurisdiction will be restored automatically, and an order affirming the District Court will issue immediately.
To be clear, we unequivocally pass no judgment on the sufficiency of the agency's rationale at this juncture. Nor do we presume to tell the District Court what weight or significance it must attach to the FBI Disclosures. Further, we note that Congress has recently passed—and the President signed into law—the FOIA Improvement Act of 2016. See FOIA Improvement Act of 2016, Pub. L. No. 114-185, 130 Stat. 538 (2016). We express no view as to the effect, if any, of that Act upon FOIA law generally or the CIA's Glomar response in this case. We leave these issues to the District Court, equipped with the relevant facts, in the first instance.
We conclude that the FBI Disclosures are relevant to the issues raised in this appeal and that those documents should be considered in this case. Accordingly, we REMAND the case to the District Court with instructions to enter an order that states whether its prior conclusion that the CIA adequately justified its Glomar response must be revised in light of the FBI Disclosures and any post-remand submissions. Either party may restore appellate jurisdiction by filing a letter with the Clerk of this Court, as prescribed above. In the interests of judicial economy, any such reinstated appeal will be assigned to this panel. The mandate shall issue forthwith.
Debra Ann Livingston, Circuit Judge, dissenting:
I respectfully dissent because the majority's decision is, in my view, both legally erroneous and likely to mislead other courts in future Freedom of Information Act (“FOIA”) litigation. The majority does not reach the merits of this appeal, which arises from a FOIA request by Sergio Florez, who seeks records from the Central Intelligence Agency (“CIA”) concerning his deceased father, Dr. Armando J. Florez (“Dr. Florez”), a Cold War-era Cuban diplomat who ultimately defected to the United States. Instead, it determines that declassified documents concerning Dr. Florez that were released by the Federal Bureau of Investigation (“FBI”) during the pendency of this appeal are relevant to the CIA's position (reaffirmed after review of the FBI documents) that the existence or nonexistence of responsive records in the CIA's possession constitutes information exempt from disclosure pursuant to FOIA, rendering a Glomar response appropriate. Maj. Op. at 184–85. The majority remands to the district court for that court to consider, in the first instance, whether these FBI disclosures affect its conclusion that the CIA's explanation for invoking FOIA Exemptions (b)(1) and (b)(3) and declining to confirm or deny the existence of responsive records appears “logical and plausible” and is thus sufficient. Id. at 15 (quoting Ctr. for Constitutional Rights v. C.I.A. , 765 F.3d 161, 168 (2d Cir.2014) ); see Wilner , 592 F.3d at 69–70, 73 (noting that in evaluating an agency's Glomar response, agency affidavits are to be afforded “substantial weight” and are sufficient when those affidavits describe the justifications for nondisclosure with “reasonably specific detail” and the basis for invoking an exemption “appears logical or plausible”).
This Circuit joined the D.C., First, Seventh, and Ninth Circuits in recognizing the propriety of a Glomar response in 2009. As this Court then noted, “[t]he Glomar doctrine is well settled as a proper response to a FOIA request because it is the only way in which an agency may assert that a particular FOIA statutory exemption covers the ‘existence or nonexistence of the requested records' in a case in which a plaintiff seeks such records.” Wilner v. N.S.A ., 592 F.3d 60, 68 (2d Cir.2009) (quoting Phillippi v. C.I.A. , 546 F.2d 1009, 1012 (D.C.Cir.1976) ). The Court “join[ed] our sister Circuits in holding that ‘an agency may refuse to confirm or deny the existence of records where to answer the FOIA inquiry would cause harm cognizable under a[ ] FOIA exception.’ ” Id . (second alteration in original) (quoting Gardels v. CIA , 689 F.2d 1100, 1103 (D.C.Cir.1982) ).
Respectfully, it is the legal determination that the FBI disclosures are relevant to the disposition of this matter with which I disagree. I conclude that the Government is correct in its contention that the FBI disclosures “do not bear on this Court's consideration of the issues raised on appeal,” Gov't Letter Br. 3, so that there is no basis in FOIA or the cases construing it for the remand that the majority directs. Accordingly (and unlike the majority) I would reach the merits and would affirm the judgment of the district court.
In remanding to the district court, the majority suggests that it is proceeding “in [the] precise manner” requested by the CIA. Maj. Op. at 187. This is incorrect. The CIA has asserted that the FBI's disclosures “do not affect the disposition of this case,” that the FBI's decision to release information regarding Dr. Florez “does not cast doubt on the propriety of the CIA's claimed exemptions,” and that, accordingly, the FBI documents “do not bear on this Court's consideration of the issues raised on appeal.” Gov't Letter Br. 2-3. The Government requests remand only in a footnote, id . at 3 n.2, and only in the event that this Court rejects, as it has, the CIA's position that the FBI disclosures lack “any bearing on this Court's consideration of the issues raised in this appeal,” id. at 1.
* * *
At the start, the FBI disclosures at issue, made pursuant to a separate FOIA request by Florez to the FBI, do not even mention the CIA, much less the existence or nonexistence of a classified relationship between Dr. Florez and the CIA, or the existence or nonexistence of Agency records regarding him. In such circumstances, it is difficult to discern (to say the least) how these FBI documents could affect, in any way, the adequacy of the CIA 's showing (supported by declaration) that its Glomar response to Florez's FOIA request was appropriate—that information regarding the existence or nonexistence of records in the CIA's possession is exempt from disclosure under two separate FOIA exemptions.
The majority principally asserts that the FBI disclosures may call into question whether revealing the existence or nonexistence of records in the CIA's possession relevant to Dr. Florez reasonably could be expected to result in harm to the national security. Maj. Op. at 17 & n. 6. The FBI records, however, neither address nor cast light on the national security harms that the CIA relied on before the district court in asserting its entitlement to FOIA Exemption (b)(1): inter alia , that confirmation of the existence or nonexistence of responsive records reasonably could be expected to cause damage to the national security by disclosing “whether or not the CIA has or had an intelligence interest in [Dr.] Florez or his associates,” “whether or not the CIA engaged in intelligence operations involving him, and the location of those operations,” and whether “the CIA maintained any human intelligence sources related to an interest in [Dr.] Florez”—the public revelation of which could jeopardize both human intelligence sources and the Agency's credibility in maintaining them. See J.A. 49-52 (emphases added).
As pertinent here, FOIA Exemption (b)(1), see 5 U.S.C. § 552(b)(1), one of the two separate exemptions on which the CIA's Glomar response was based, requires a showing that confirming or denying the existence of responsive records “reasonably could be expected to result in damage to the national security,” Exec. Order No. 13526, 75 Fed. Reg. 707, 707 (Dec. 29, 2009).
These are significant national security concerns. As the D.C. Circuit has repeatedly noted, sources abroad, fearing retaliation against themselves or family and friends, “often refuse to aid the CIA absent assurances of confidentiality” that logically must be honored “even decades after the death of the foreign national,” lest the Government's substantial interest in the effective operation of its foreign intelligence service be thwarted. Wolf v. C.I.A. , 473 F.3d 370, 376–77 (D.C.Cir.2007) ; see also Fitzgibbon v. C.I.A. , 911 F.2d 755, 761 (D.C.Cir.1990) (noting that “[i]f potentially valuable intelligence sources come to think that the Agency will be unable to maintain the confidentiality of its relationship to them, many could well refuse to supply information to the Agency in the first place” (emphasis omitted) (quoting Sims v. C.I.A. , 471 U.S. 159, 175, 105 S.Ct. 1881, 85 L.Ed.2d 173 (1985) )). The FBI documents here—which, notwithstanding the majority's claim to the contrary, disclose little regarding Dr. Florez and say nothing at all about any connection to the CIA—are simply not helpful in assessing the logic and plausibility of such concerns in the present case. Indeed, the majority's claim to the contrary hinges on reframing the CIA's asserted national security interest as involving a generalized concern with “masking the government's intelligence interest (if any) in Dr. Florez”—a reframing that, however subtle, simply gainsays the significant national security concerns associated with and peculiar to the effective operation of this country's foreign intelligence service. Maj. Op. at 184–85 (emphasis added).
At any rate, all this is somewhat beside the point. For as it turns out, only one of the two exemptions on which the district court based its decision (each of which provides a separate and independent basis for a Glomar response) even requires a showing that confirming or denying the existence of responsive records reasonably could be expected to result in damage to the national security. See 5 U.S.C. § 552(b)(1) ; see also Exec. Order No. 13526, 75 Fed. Reg. 707, 707 (Dec. 29, 2009). The other, FOIA Exemption (b)(3), provides simply that FOIA's disclosure requirements do not apply to matters “specifically exempted from disclosure by statute.” 5 U.S.C. § 552(b)(3). As the Supreme Court has stated, this exemption “applies to records that any other statute exempts from disclosure, thus offering Congress an established, streamlined method to authorize the withholding of specific records that FOIA would not otherwise protect.” Milner v. Dep't of Navy , 562 U.S. 562, 131 S.Ct. 1259, 1271, 179 L.Ed.2d 268 (2011) (citation omitted). The sole issue for decision regarding the applicability of FOIA Exemption (b)(3), moreover, as we have noted, “is the existence of a relevant statute and the inclusion of withheld material within the statute's coverage.” Wilner , 592 F.3d at 72 (quoting Ass'n of Retired R.R. Workers, Inc. v. U.S. R.R. Ret. Bd. , 830 F.2d 331, 336 (D.C.Cir.1987) ).
Both statutory provisions on which the district court relied—Section 102(A)(i)(1) of the National Security Act of 1947, as amended, 50 U.S.C. § 3024(i)(1), pursuant to which the Director of National Intelligence “shall protect intelligence sources and methods from unauthorized disclosure,” and Section 6 of the Central Intelligence Agency Act of 1949, as amended, 50 U.S.C. § 3507, exempting the CIA specifically from the provisions of any law requiring the disclosure of the functions of Agency personnel—qualify as statutes of exemption (a point Florez does not dispute on appeal). In such circumstances, our inquiry is limited to the simple question whether the withheld information “falls within the statute.” Larson, 565 F.3d at 868 ; accord Wilner, 592 F.3d at 72. Moreover, regarding the National Security Act exemption, in particular, the Supreme Court has recognized that its plain language (requiring the Director to protect intelligence sources and methods from unauthorized disclosure) “may not be squared with any limiting definition” such as one protecting “only confidential or nonpublic intelligence sources.” Sims , 471 U.S. at 169–70, 173, 105 S.Ct. 1881 (noting that Congress intended National Security Act exemption “to protect the secrecy and integrity of the intelligence process” and that this Act commits broad power to the Director “to control the disclosure of intelligence sources”). Given that the FBI disclosures here do not even discuss the CIA or its activities, it is hard to fathom how these disclosures could possibly impact the logic and plausibility of the CIA 's representation that a Glomar response is appropriate pursuant to FOIA Exemption (b)(3)—that “acknowledging the existence or nonexistence of [CIA] records ... would reveal information” likely to lead to unauthorized disclosures regarding its sources and methods and its clandestine intelligence activities. J.A. 55; see also Gardels , 689 F.2d at 1105 (regarding Exemption (b)(3), “[t]he test is not whether the court personally agrees in full with the CIA's evaluation of the danger—rather, the issue is whether on the whole record the Agency's judgment objectively survives the test of reasonableness, good faith, specificity, and plausibility in this field of foreign intelligence in which the CIA is expert and given by Congress a special role.”). The majority does not plumb the mystery of how these documents could be relevant to the CIA's Exemption (b)(3) showing, or even attempt to do so.
The majority seriously downplays the significance of the CIA's reliance on FOIA Exemption (b)(3), contending that “the CIA has proffered a single general rationale with respect to both Exemptions.” Maj. Op. 18 n.6. This is incorrect. The CIA's Exemption (b)(3) rationale focuses on the specific statutory provision at issue and the materials it protects—not on the potential effects of disclosure on the national security. The declaration of Martha M. Lutz, Chief of the CIA's Litigation Support Unit, states specifically regarding the claimed exemption under the National Security Act, for instance, that “acknowledging the existence or nonexistence of records reflecting a classified connection to the CIA would reveal information that concerns intelligence sources and methods, which the National Security Act is designed to protect.” J.A. 55. Once the National Security Act has been invoked pursuant to FOIA Exemption (b)(3), our inquiry focuses narrowly on the question “whether the withheld material relates to intelligence sources and methods.” Larson v. Dep't of State , 565 F.3d 857, 865 (D.C.Cir.2009) ; see also Fitzgibbon, 911 F.2d at 761 (noting that Exemption (b)(3) “differs from other FOIA exemptions in that its applicability depends less on the detailed factual contents of specific documents; the sole issue for decision is the existence of a relevant statute and the inclusion of withheld material within the statute's coverage” (quoting Ass'n of Retired R.R. Workers, 830 F.2d at 336 )). The majority simply elides this distinction between Exemptions (b)(1) and (b)(3).
See also Wolf , 473 F.3d at 378 (noting in the context of FOIA Exemption (b)(3) that the Supreme Court “gives even greater deference to CIA assertions of harm to intelligence sources and methods under the National Security Act” then in the context of the (b)(1) exemption); Students Against Genocide v. Dep't of State , 257 F.3d 828, 835 (D.C.Cir.2001) (same); Fitzgibbon , 911 F.2d at 766 (same).
The majority contends that the FBI documents reflect that “the FBI maintained an active interest in Dr. Florez for well over a decade.” Maj. Op. at 185. Even if this were the case, it in no way undercuts the CIA's assertion that information as to the existence or nonexistence of documents in its possession reflecting a classified connection to the CIA constitutes information concerning intelligence sources and methods that is exempt from FOIA disclosure by the National Security Act—an assertion to which this Court owes deference. See Wolf , 473 F.3d at 378 ; Students Against Genocide , 257 F.3d at 836.
At any rate, after careful review, not the “casual[ ]” one the majority inexplicably charges me with, see Maj. Op. at 185, I read the FBI disclosures very differently. Far from establishing the FBI's “active” interest in Dr. Florez and providing a “wealth of information” about him, see Maj. Op. at 185, these documents, which total under 60 pages amassed over ten years, offer precious little insight into Dr. Florez beyond basic biographical information and speculation about his personal life. Nearly half of the documents divulge no information about him whatsoever. Some of the more reliable information concerning him, moreover, appears to come from two newspaper clippings among the documents, one announcing his appointment as chargé d'affaires in Washington, D.C. and the other his defection from the Castro regime.
To be sure, the FBI maintained a file on Dr. Florez and copied certain agencies (not including, of course, the CIA) on certain documents. Bureau offices on occasion exchanged whatever information they had about him, particularly (and unsurprisingly, given the FBI's domestic counterintelligence orientation) when Dr. Florez was stationed in Washington, D.C. and, years later, when he decided to defect to the United States. The fact that the Bureau periodically communicated the scant information it had concerning Dr. Florez internally and with specified external agencies, however, hardly reveals the kind of defined, concrete interest that could even conceivably be of “appreciable probative value” in assessing whether the CIA, a separate agency with interests, sources, and methods of its own, met its burden in justifying its Glomar response. Id . at 184.
The majority does not even endeavor to set out the string of logical inferences that are necessary to establish the relevance of this material to the CIA's assertion of FOIA Exemptions (b)(1) and (b)(3). It cannot, and for a simple reason. The mere fact that the FBI maintained some interest in a foreign diplomat who was stationed in this country for a period and who eventually defected here—the only non-conjectural conclusion that one may reach from these documents—is simply not germane to the question whether the CIA's invocation of FOIA Exemptions (b)(1) and (b)(3) is adequately supported.
The FBI documents say nothing about the CIA's interest or lack of interest in Dr. Florez, nor do they address Agency sources and methods or the functions of Agency personnel—the very matters protected from disclosure by the exemptions that the CIA has asserted. I thus cannot agree that these records even potentially affect the conclusion, already drawn by the district court, that the CIA has adequately shown that the claimed exemptions apply by providing “explanations of potential harm to national security” that are “both ‘logical’ and ‘plausible’ ” and by demonstrating that “acknowledging the existence or nonexistence of the records [Florez] seeks could reasonably be expected to lead to the unauthorized disclosure of intelligence sources and methods as well as clandestine intelligence activities,” which are at the core of the functions of Agency personnel. Florez v. C.I.A. , 2015 WL 728190, *6, *8 (S.D.N.Y. Feb. 19, 2015) (quoting Wilner , 592 F.3d at 73 ).
* * *
The majority cannot explain how these FBI documents, which do not even mention the CIA, much less any relationship between the CIA and Dr. Florez, are relevant to assessing the logic and plausibility of the Agency's justification for its Glomar response. Faced with this difficulty, the majority faults me for even pointing out that the FBI documents do not mention the CIA, asserting that to do so is inconsistent with my real position that the disclosures of one federal agency “are never relevant and must be wholly disregarded” in assessing the propriety of another agency's Glomar response. Maj. Op. 186(emphasis added). I take no such position, however, as to hypothetical cases not presently before this panel, nor do I pronounce, as the majority repeatedly charges, “a rule limiting the evidence a district court may consider in a Glomar inquiry.” Id . at 187. I do no more than conclude (contrary to the majority) that these FBI documents, suggesting (albeit without particulars, and to a limited degree) that the FBI maintained some interest in Dr. Florez for some period of time, are simply not relevant to the question whether the CIA's justification for its Glomar response in this case is plausible and makes sense.
The majority also charges me with “exclusive reliance on the official acknowledgment doctrine,” id . at 187, which it says I use to “propagat[e] a per se rule barring consideration of third party disclosures on the sufficiency of an agency's Glomar response,” id . at 186 n. 8. Again, I urge no such rule. That said, however, I cannot agree that this doctrine is not properly considered in assessing the question whether these FBI documents are relevant to the CIA's rationale for its Glomar response. Indeed, my conclusion that they are not is only reinforced through more general consideration of the FOIA statute and Glomar doctrine itself.
The FOIA statute recognizes different agencies' divergent missions, interests, and methods, see 5 U.S.C. § 552 (requiring disclosure on an agency-by-agency basis), and it is well established that the disclosure of material by one agency will not be attributed to another, so as to forestall the second agency's recourse to appropriate FOIA exemptions. See Wilson v. C.I.A. , 586 F.3d 171, 186 (2d Cir.2009) (“[T]he law will not infer official disclosure of information classified by the CIA from ... release of information by another agency, or even by Congress.”); see also Moore v. C.I.A. , 666 F.3d 1330, 1333 n. 4 (D.C.Cir.2011) (release of a document “by the FBI ” cannot constitute “an official acknowledgment by the CIA ”); Frugone , 169 F.3d at 774–75 (upholding the CIA's ability to make a Glomar response despite official disclosure of the same information by the Office of Personnel Management). In the Glomar context specifically, moreover, courts have long recognized the danger in “requiring [an agency] to break its silence” as a result of “statements made by another agency.” Frugone , 169 F.3d at 775. Agencies have different missions. “[I]t is logical to conclude” regarding a foreign intelligence service like the CIA, for instance, “that the need to assure confidentiality” to human intelligence sources and to foster confidence in such assurances vis-à -vis past, present, and future sources may require “neither confirming nor denying the existence of records” regarding foreign nationals—whether or not they be subjects of interest or persons with whom the Agency maintained a relation—for many years. Wolf , 473 F.3d at 377 ; see also Fitzgibbon , 911 F.2d at 763-64 (noting “compelling interest” in protecting both national security information and “the appearance of confidentiality so essential to the effective operation of our foreign intelligence service” (quoting Sims , 471 U.S. at 175, 105 S.Ct. 1881 )). Other agencies may not share this concern at all, or to the same degree.
The majority acknowledges that, provided an agency has established a proper basis for a Glomar response, an agency is not required to break its silence “as a result of ‘statements made by another agency.’ ” Maj. Op. at 186(quoting Frugone , 169 F.3d at 775 ). The majority asserts that this precedent is inapplicable here, however, because the majority is not “imput [ing] the FBI's decision to disclose information about Dr. Florez to the CIA, or suggest[ing] that the FBI Disclosures necessarily preclude the CIA's right to assert a Glomar response.” Maj. Op. at 186–87. Instead, the majority asserts, “we simply conclude that the FBI Disclosures are relevant evidence—unavailable to the District Court at the time of its initial decision—bearing upon the sufficiency of the justifications set forth by the CIA in support of its Glomar response.” Maj. Op. at 187.
But how can these documents be relevant, given that they do not even mention the CIA, when the declarations supporting the CIA's claimed exemptions are centrally concerned with harms associated with revealing the existence or nonexistence of documents reflecting a classified connection to the CIA , in its role as the United States' foreign intelligence service? With respect, the majority is cavalier, I conclude, in its dismissal of the official acknowledgement doctrine's relevance to this case. If, on remand, the district court were to determine that the FBI documents render illogical or implausible the CIA's affidavits (how the district court could reach such a conclusion, given the FBI disclosures themselves, I cannot say), that conclusion would produce the “anomalous result” of one agency's revelations obligating disclosure of classified material by another. Frugone , 169 F.3d at 775. The majority's error in deeming these irrelevant documents germane thus appears to invite by the back door what the official acknowledgement doctrine prohibits at the front.
The “official acknowledgment” case law, as I read it, has two animating principles. The cases ordering disclosure on the basis of official acknowledgment tend to emphasize the first of these: that an agency, having already disclosed certain classified information, cannot later refuse to confirm or deny the existence or nonexistence of that information. See, e.g. , N.Y. Times Co. v. Dep't of Justice , 756 F.3d 100, 122 (2d Cir.2014) ; Wolf , 473 F.3d at 379. The cases rejecting arguments of official acknowledgment, however, at least when those arguments rely on disclosures by a third party (often another agency), point to the harm of using one agency's disclosures as a “FOIA backdoor” to obligate another agency to reveal classified information. See, e.g. , Wilson , 586 F.3d at 186 ; Frugone , 169 F.3d at 775.
To be sure, these courts did not address the novel theory of relevance before us here. Nevertheless, they uniformly resisted any temptation to reconsider the responding agency's justification for its Glomar response in light of third party disclosures. As the D.C. Circuit put it in Frugone, in discussing the National Security Act exemption, “[c]ommon sense suggests that [those charged with the protection of intelligence sources and methods] must have authority to maintain secrecy commensurate with [this statutory] responsibility,” notwithstanding another agency's disclosures. 169 F.3d at 775. Here, although styled as an evidentiary matter distinct from the “official acknowledgment” doctrine, the majority's theory of relevance implicates the same concern: that Congress could not have intended the “anomalous result” that disclosures by one agency could open the door to compelled disclosure by another. See id .
With respect, I simply cannot see the basis on which these FBI documents are relevant to the Glomar inquiry that the district court has already undertaken. As we have said, “[i]n evaluating an agency's Glomar response, a court must accord ‘substantial weight’ to the agency's affidavits, ‘provided [that] the justifications for nondisclosure are not controverted by contrary evidence in the record or by evidence of ... bad faith.’ ” Wilner , 592 F.3d at 68 (quoting Minier v. C.I.A. , 88 F.3d 796, 800 (9th Cir.1996) ); see also Wolf , 473 F.3d at 374 (noting that courts “conducting de novo review in the context of national security concerns ... ‘must accord substantial weight to an agency's affidavit concerning the details of the classified status of the disputed record’ ” (quoting Miller v. Casey , 730 F.2d 773, 776 (D.C. Cir. 1984) )). There is no evidence of bad faith here, and the FBI documents are in no way contrary to the justifications for nondisclosure already proffered by the CIA. In such circumstances, a court “should not conduct a more detailed inquiry to test the agency's judgment and expertise or to evaluate whether the court agrees with the agency's opinions.” Larson , 565 F.3d at 865 ; accord Wilner , 592 F.3d at 76 ; see also Students Against Genocide , 257 F.3d at 835 (noting that “the assessment of harm to intelligence sources and methods is entrusted to the Director of Central Intelligence, not to the courts”). To rely on these documents as a basis for remand is to “second-guess the predictive judgments made by the government's intelligence agencies” in precisely the manner we have in the past eschewed. Wilner , 592 F.3d at 76 (quoting Larson , 565 F.3d at 865 ).
I conclude, for substantially the reasons set out in the district court's careful and thorough opinion, that the CIA met its burden in justifying its Glomar response. Because I can discern no basis on which the FBI disclosures draw into question the CIA's explanations as to why information regarding the existence or nonexistence of records in its possession is exempted from disclosure by FOIA, I respectfully dissent from the majority's decision to remand and would, instead, affirm.