finding that the deliberative process privilege is “ ‘predicated on the recognition that the quality of administrative decision-making would be seriously undermined if agencies were forced to operate in a fish bowl.’ ”
How later courts described this case
- finding that the deliberative process privilege is “ ‘predicated on the recognition that the quality of administrative decision-making would be seriously undermined if agencies were forced to operate in a fish bowl.’ ”
- noting that because the Government will rarely come forward with information permitting “either an inference of confidentiality or an inference of nonconfidentiality ... the presumption of confidentiality is irre-buttable”
- emphasizing that the application of Exemption 5 in the context of the deliberative process privilege “depends on the factual content and purpose of the requested document” as opposed to other variables
- taking categorical approach to FOIA cases as suggested by the Supreme Court in Reporters’ Committee, supra, by creating presumption that FBI generally affords confidentiality to its sources
Written by the judges who cited it.
The opinion
SILBERMAN, Circuit Judge,
concurring in the denial of rehearing en banc, with whom WILLIAMS, Circuit Judge, joins:
Appellant’s primary argument for rehearing is based on the notion that our opinion — and Schmerler, on which it is based — rely on an unjustifiably strong presumption that when the FBI interviews witnesses in a criminal investigation, those witnesses are a “confidential source” within the meaning of Exemption 7(D). It is argued that this presumption is inconsistent with the undisputed proposition that the government has the burden of proof in FOIA cases. But the presumption is directed, strictly speaking, not at the burden of proof, but at the quantum of evidence necessary to carry the burden. Cf. Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248 , 101 S.Ct. 1089 , 67 L.Ed.2d 207 (1981) (discussing allocation of burden of proof in Title VII cases and how an inference creates a presumption in the prima facie case). Fixing the burden of proof on a party is plainly not inconsistent with a substantive rule identifying specific facts that will carry the burden.
Confidentiality, of course, is a relative term. When a witness tells an FBI agent something relevant to a criminal investigation the witness assumes, absent special circumstances, that other FBI and law enforcement personnel will have access to the information, which, in a sense, makes the divulgence less “confidential” than if the agent would swear not to tell a soul. But it cannot be seriously argued that one’s status as a “confidential source” should turn on the precise scope of that expectation of confidentiality.
There is always some element of confidentiality in this setting unless, at a minimum, both the agent and the witness are wholly indifferent to publication. Our holding in Schmerler, which we follow here, is merely that the element of confidentiality typically present in such interviews satisfies the “confidential source” standard. Otherwise, the FBI would routinely be compelled to produce evidence as to the particular expectations of the interviewee and the agent, and we would be obliged on a case-by-ease basis to try to determine just how much confidentiality qualifies as a “confidential source.” By identifying the set of circumstances that support a strong inference of confidentiality, subject to the possibility of exception in an extraordinary case, we seek to avoid ad hoc decisionmaking. Cf. Schmerler v. Federal Bureau of Investigation, 696 F.Supp. 717 (D.D.C.1988) (relying on a number of unique factors), rev’d, 900 F.2d 333 (D.C. Cir.1990). Because at least some element of confidentiality is virtually always present, United States Department of Justice v. Reporters Committee for Freedom of the Press, 489 U.S. 749 , 109 S.Ct. 1468, 1483 , 103 L.Ed.2d 774 (1989) counsels that we treat such questions generically.