CITIZENS FOR RESPONSIBILITY AND ETHICS IN WASHINGTON, APPELLANT v. UNITED STATES DEPARTMENT OF JUSTICE, APPELLEE
No. 18-5116
United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued December 4, 2018 Decided April 30, 2019
Appeal from the United States District Court for the District of Columbia (No. 1:17-cv-00432)
Anne L. Weismann argued the cause for the appellant. Alan B. Morrison and Adam J. Rappaport were with her on brief. Stuart McPhail entered an appearance.
Alex Abdo and Jameel Jaffer were on brief for the amici curiae The Knight First Amendment Institute at Columbia University, et al. in support of the appellant.
Brad Hinshelwood, Attorney, United States Department of Justice, argued the cause for the appellee. Michael S. Raab, Attorney, was with him on brief.
Before: HENDERSON and PILLARD, Circuit Judges, and GINSBURG, Senior Circuit Judge.
Opinion for the Court filed by Circuit Judge HENDERSON.
Dissenting opinion filed by Circuit Judge PILLARD.
I. BACKGROUND
The authority of the Office of Legal Counsel (OLC) is “nearly as old as the Republic itself.” Citizens for Responsibility & Ethics in Wash. v. U.S. Dep‘t of Justice (CREW I), 846 F.3d 1235, 1238 (D.C. Cir. 2017). Since the Judiciary Act of 1789, the United States Attorney General has had the authority “to give his advice and opinion upon questions of law when required by the President of the United States, or when requested by the heads of any of the departments, touching any matters that may concern their departments.” Judiciary Act of 1789, § 35, 1 Stat. 73, 93 (codified as amended at
Although the OLC frequently conveys its legal advice to executive agencies through informal means, it sometimes does so through “formal written opinions.” See Memorandum from David J. Barron, Acting Assistant Attorney Gen., to Attorneys of the Office of Legal Counsel, Best Practices for OLC Legal Advice and Written Opinions 1–2 (July 16, 2010) (hereinafter Best Practices Memo). Formal written opinions “take the form of signed memoranda, issued to an Executive Branch official who has requested the [OLC]‘s opinion.” Id. at 2. The OLC considers its formal written opinions to be “one particularly important form of controlling legal advice.” Id. at 1. Indeed, a former head of the OLC has described its formal written opinions and informal advice as “authoritative” and “binding by custom and practice in the executive branch.” Josh Gerstein, Official: FOIA Worries Dampen Requests for Formal Legal Opinions, Politico: Under the Radar (Nov. 5, 2015), https://www.politico.com/blogs/under-the-radar/2015/11/official-foia-worries-dampen-requests-for-formal-legal-opinions-215567.
The OLC publishes some, but not all, of its formal written opinions. See Best Practices Memo 5. In deciding whether to publish a formal written opinion, the OLC considers “the potential importance of the opinion to other agencies or officials in the Executive Branch; the likelihood that similar questions may arise in the future; the historical importance of the opinion or the context in which it arose; and the potential significance of the opinion to the [OLC]‘s overall jurisprudence.” Id. “In applying these factors, the [OLC] operates from the presumption that it should make its significant opinions fully and promptly available to the public.” Id. The OLC then weighs those factors against “countervailing considerations” such as whether publication “would reveal classified or other sensitive information relating to national security[,] . . . would interfere with federal law enforcement efforts or is prohibited by law.” Id. Additionally, the OLC “decline[s] to publish opinions when doing so is necessary to preserve internal Executive Branch deliberative processes or protect the confidentiality of information covered by the attorney-client relationship between OLC and other executive offices.” Id. at 5-6.
Invoking FOIA, CREW seeks to compel disclosure of the OLC‘s unpublished formal written opinions. Importantly, CREW does not rely on FOIA‘s “most familiar provision“—
While CREW I was pending, we also decided EFF. 739 F.3d 1. In EFF, we addressed a claim brought under FOIA‘s “reactive” provision seeking disclosure of a formal written opinion the OLC had sent to the FBI. Id. at 4–6. The court held that the opinion was exempt from disclosure by the deliberative process privilege. Id. at 9–10. As relevant here, it determined that, notwithstanding the opinion at issue bore some “indicia of a binding legal decision“—namely, that it was “controlling (insofar as agencies customarily follow OLC advice that they request), precedential, and can be withdrawn“—it did not constitute the FBI‘s “working law” because the OLC “does not speak with authority on the FBI‘s policy.” Id. at 9. Instead, the court concluded, the OLC opinion was “precisely the sort of ‘advisory opinion . . . comprising part of a process by which governmental decisions and policies are formulated’ that is covered by the deliberative process privilege.” Id. at 10 (alteration in original) (quoting Pub. Citizen, Inc. v. Office of Mgmt. & Budget, 598 F.3d 865, 875 (D.C. Cir. 2010)).
Following our decisions in EFF and CREW I, CREW sent a letter to the OLC in February 2017 renewing its request that the OLC disclose all of its formal written opinions and accompanying indices under FOIA‘s reading-room provision. The OLC did not respond to the renewed request, prompting CREW to sue again, this time under FOIA‘s judicial-review provision,
The district court dismissed CREW‘s new complaint under Federal Rule of Civil Procedure 12(b)(6) for failure to state a claim upon which relief can be granted. Citizens for Responsibility & Ethics in Wash., 298 F. Supp. 3d at 156. It observed that CREW‘s complaint “is premised on a universal claim” for all of the OLC‘s formal written opinions and corresponding indices. Id. at 154. “Accordingly,” the court reasoned, “if the [Department of Justice] can identify any formal written opinions that are not subject to FOIA disclosure, CREW‘s universal claim fails.” Id. It then concluded that our EFF decision “dooms CREW‘s complaint as currently articulated, because it establishes that at least one of OLC‘s formal written opinions—the opinion in EFF—is exempt from FOIA disclosure.” Id. at 155. The court, however, acknowledged that CREW might state a viable claim if it “amend[ed] its complaint to allege that some specific subset of OLC‘s formal written opinions [is] being unlawfully withheld.” Id. at 156. The court therefore gave CREW leave to amend its complaint. Id. CREW declined to amend its complaint, however, opting instead to appeal.
Notably, CREW is not the only advocacy group seeking to compel disclosure of the OLC‘s formal written opinions in toto. Campaign for Accountability (CfA), amicus in this appeal, filed a similar suit under FOIA‘s reading-room provision. See Campaign for Accountability v. U.S. Dep‘t of Justice, 278 F. Supp. 3d 303, 305 (D.D.C. 2017). There, as here, the district court
II. ANALYSIS
The Court reviews de novo the dismissal of a complaint under Federal Rule of Civil Procedure 12(b)(6) for failure to state a claim upon which relief can be granted. Coburn v. Evercore Tr. Co., N.A., 844 F.3d 965, 968 (D.C. Cir. 2016). To state a claim under FOIA, CREW must plausibly allege “that an agency has (1) ‘improperly‘; (2) ‘withheld‘; (3) ‘agency records.‘” Competitive Enter. Inst. v. Office of Sci. & Tech. Policy, 827 F.3d 145, 147 (D.C. Cir. 2016) (quoting Kissinger v. Reporters Comm. for Freedom of the Press, 445 U.S. 136, 150 (1980)). In this case, there is no dispute that the formal written opinions the OLC has declined to publish are “withheld” “agency records.” The only remaining question is whether the OLC has withheld those opinions “improperly.”
An agency withholds its records “improperly” if it fails to comply with one of FOIA‘s “mandatory disclosure requirements.” U.S. Dep‘t of Justice v. Tax Analysts, 492 U.S. 136, 150 (1989). Relevant here, FOIA‘s reading-room provision mandates that an agency disclose certain enumerated categories of records. See
CREW‘s complaint makes no such allegation. It instead alleges only that the OLC‘s formal written opinions are “controlling,” “authoritative” and “binding.” As EFF noted, these descriptors alone are insufficient to render an OLC opinion the “working law” of an agency; that OLC opinions are “controlling (insofar as agencies customarily follow OLC advice that they request), precedential, and can be withdrawn . . . does not overcome the fact that OLC does not speak with authority on the [agency‘s] policy.” Id.1
Importantly,
CREW and the dissent raise four primary objections to our analysis. First, CREW contends that our approach erroneously places on CREW the burden of identifying a subset of the OLC‘s formal written opinions that constitutes the “working law” of an agency and is therefore subject to disclosure under FOIA‘s reading-room provision. Granted, as CREW emphasizes, FOIA places the burden “on the agency to sustain its action,”
We impose this standard in assessing claims brought under FOIA‘s reactive provision. Section 552(a)(3) imposes a mandatory disclosure requirement on agencies “upon any request for records” that “reasonably describes such records” and complies with procedural regulations.
Although FOIA‘s reading-room provision differs from its reactive provision, the analytical mechanics remain the same. Whether an agency must disclose records under FOIA‘s reading-room provision turns not on the existence of a plaintiff‘s request but on the nature of the records—whether they fall within one of
Second, the dissent argues that in any event, CREW‘s complaint alleges sufficient factual material to state a plausible claim that the OLC‘s formal written opinions are subject to disclosure under FOIA‘s reading-room provision. Dissent 2-4. Revealingly, however, the dissent does not rest its analysis solely on the factual allegations in CREW‘s complaint—that the OLC‘s formal written opinions are “controlling,” “authoritative” and “binding“; instead, the dissent supplements those allegations with others not contained in the complaint. For example, the dissent presents one of the OLC‘s formal written opinions regarding the Defense of Marriage Act as an example of an opinion that may be subject to disclosure under FOIA‘s reading-room provision. Id. at 3–4. Regardless of the plausibility of the complaint the dissent describes, that is not the complaint CREW filed in the district court. We are therefore left with assessing the sufficiency of CREW‘s actual allegations that the OLC‘s formal written opinions are “controlling,” “authoritative” and “binding.” As we have explained, these allegations alone are insufficient to present a plausible claim that the OLC‘s formal written opinions fall within one of the reading-room provision‘s enumerated categories.
Third, the dissent argues that we draw too much from EFF and thus require CREW to plead around potential FOIA exemptions. Id. at 4–6. The dissent, however, wrongly treats EFF as only an exemption case. Although EFF ultimately held that an OLC formal written opinion is exempt from disclosure, the decision adopted the broader rule that the OLC‘s
Fourth, CREW complains that requiring it to identify a subset of the OLC‘s formal written opinions subject to FOIA‘s reading-room provision is unfair as a matter of policy because it “imposes a burden of proof that is almost impossible for CREW to meet.” But the purported unfairness CREW faces is self-inflicted. CREW declined to avail itself of other measures at its disposal, not the least of which was acceptance of the district court‘s invitation to amend its complaint as amicus CfA has done. See Am. Compl. ¶¶ 33–49, Campaign for Accountability, No. 1:16-cv-1068 (amended complaint identifies several subcategories of OLC‘s formal written opinions CfA believes are subject to FOIA‘s reading-room provision).4 CREW also would have had no difficulty carrying its pleading burden, and thereby putting the OLC to its burden to sustain its action, had it proceeded under FOIA‘s reactive provision,
So ordered.
Pillard, Circuit Judge, dissenting: It is not certain at this stage how much—if any—of OLC‘s output might ultimately be subject to disclosure under FOIA‘s reading-room provision. But “a well-pleaded complaint may proceed even if it strikes a savvy judge that actual proof of . . . facts [supporting relief] is improbable, and ‘that a recovery is very remote and unlikely.‘” Bell Atl. Corp. v. Twombly, 550 U.S. 544, 556 (2007) (quoting Scheuer v. Rhodes, 416 U.S. 232, 236 (1974)). The sole issue before us is the threshold question whether CREW has alleged enough to survive a motion to dismiss. Because I believe that it has, I would reverse the contrary judgment of the district court.
*
*
*
Under Federal Rule of Civil Procedure 8, a complaint must contain “a short and plain statement of the claim showing that the pleader is entitled to relief.” Fed. R. Civ. P. 8(a)(2). “[T]he pleadings must ‘give the defendants fair notice of what the claim is and the grounds upon which it rests,’ but the Rule ‘does not require detailed factual allegations.‘” Jones v. Kirchner, 835 F.3d 74, 79 (D.C. Cir. 2016) (quoting Twombly, 550 U.S. at 555, and Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009)). CREW plausibly alleges that at least some OLC opinions fall within FOIA‘s reading-room provision, because they are either “final opinions . . . made in the adjudication of cases” or “interpretations which have been adopted by the agency and are not published in the Federal Register,”
CREW has described—often in OLC‘s own words—that Office‘s adjudicative process and the authoritative nature of its opinions. The facts that CREW alleges to support its claim that OLC issues “final opinions” in the “adjudication of cases” describe OLC‘s role as an authoritative decider of disputes between and among entities within the executive branch. According to OLC‘s own website and Best Practices Memo, both of which are publicly accessible and are referenced and quoted in CREW‘s complaint, OLC provides controlling advice in interagency disputes when asked. J.A. 9, 11 (Compl. ¶¶ 18-19, 27,
CREW also alleges facts sufficient to raise its parallel claim that OLC renders “interpretations which have been adopted by the agency” within the meaning of Section 552(a)(2). The complaint plausibly alleges that another role of OLC is to provide legal advice that is “authoritative” and “binding by custom and practice in the executive branch“—statements of what the law permits or requires that “[p]eople are supposed to and do follow.” J.A. 9 (Compl. ¶¶ 18-19) (internal quotation marks omitted). The complaint alleges that certain of OLC‘s opinions are “controlling” interpretations, and even “may effectively be the final word on controlling law” in some situations. Id.
The Best Practices Memo bolsters those allegations by explaining that OLC provides opinions to independent agencies only if they have agreed to be bound by—i.e., will adopt—OLC‘s advice. For purposes of a motion to dismiss for failure to state a claim, we treat as true CREW‘s factual allegations that client agencies accept OLC opinions as “controlling,” “authoritative,” and “binding.” There is extra reason to do so here insofar as the allegations quote and track published descriptions by OLC veterans of the Office‘s work as including a category of legal interpretations that agencies believe they must, and do, adopt. Again, CREW‘s legal theory may ultimately fail. It may be that the agreement of one part of the executive branch (the agency “client“) to follow a legal interpretation offered by another part of the branch (OLC) is not what Section 552(a)(2) means by “interpretations which have been adopted by the agency.” But CREW plausibly alleges a regular category of OLC‘s work product that, it contends, matches the statutory description—sufficiently teeing up its claim for the district court.
By way of illustration, CREW has pointed to publicly available opinions that—although not disclosed under the reading-room provision—plausibly are subject to its requirements. One of the opinions it identified, for example, addressed whether the Defense of Marriage Act (DOMA) prevented the nonbiological child of a civil union from qualifying for insurance benefits under the Social Security Act. The government published that opinion and has made it electronically available to the public on its website, but the government‘s litigation position is that it was not required to do so—whether pursuant to the reading-room provision or otherwise. CREW disagrees, because in seeking OLC‘s advice on the DOMA question, the Social Security Administration had agreed to be bound by whatever interpretation the Office made. See OLC, Whether the Defense of Marriage Act Precludes the Nonbiological Child of a Member of a Vermont Civil Union from Qualifying for Child‘s Insurance Benefits Under the Social Security Act 243 n.1 (Oct. 16, 2007), https://www.justice.gov/file/451616/download. CREW accordingly argues that OLC‘s opinion was an “interpretation[] which ha[d] been adopted by the agency,” and so was subject to Section 552(a)(2). Contrary
The court rejects CREW‘s claim by reasoning, in effect, that CREW has asked for more relief than it can get. It points to an OLC opinion that arguably provided a “controlling legal interpretation[],” J.A. 5, but that this court held in Electronic Frontier Foundation v. U.S. Department of Justice, 739 F.3d 1 (D.C. Cir. 2014) (EFF), was protected from disclosure by the deliberative process privilege, id. at 4. I agree that EFF shows that there is a subcategory of opinions (encompassing at least one, and likely many more) that need not be disclosed under the reading-room provision, whether because they have not been adopted by the receiving agency, or are subject to a FOIA exemption, or both. But the majority makes too much soup from one oyster. EFF could only defeat CREW‘s merits claim if we were certain that every one of the Office‘s opinions would be shielded from disclosure for the reasons that were dispositive in EFF. The government does not so claim, however, and my colleagues do not so hold. The identification of a single opinion that could be withheld even were plaintiff‘s legal theory correct is no basis upon which to dismiss the complaint for failure to state a claim. Indeed, the majority itself is “skeptical of the Department of Justice‘s position that none of the OLC‘s formal written opinions constitutes the ‘working law’ of an agency subject to disclosure under FOIA‘s reading-room provision.” Op. 9 n.2. That alone should be enough, in light of the government‘s affirmative legal obligation under the reading-room provision, to entitle the claim to proceed.
The majority faults CREW for asking for “all” formal written OLC opinions, rather than making a more tailored request. See, e.g., Op. 8-9. But just because we might conceive of more exact ways to ask for the documents does not mean that CREW‘s pleading falls short. The complaint makes clear that CREW is seeking those documents that fall within the definition of what the government is required to publish under the reading-room provision. In its complaint, CREW alleges that “OLC has refused to produce . . . its formal written opinions setting forth controlling legal interpretations,” J.A. 5 (Compl. ¶ 2), as well as its “final opinions made in the adjudication of cases,” J.A. 10-11 (Compl. ¶ 26). Taking the complaint in the light most favorable to CREW, it is not requesting anything that FOIA exempts. And, even if the terms of the complaint could be read to sweep in the OLC opinion from EFF, an overambitious remedial request does not defeat a plausibly alleged legal claim.
By requiring more detail from CREW at this stage, the court effectively forces CREW to anticipate and plead around any FOIA-exemption defense the government might raise. But a potential defense not yet asserted is no ground upon which to dismiss a complaint. It is firmly established that “a plaintiff is not required to negate an affirmative defense in his complaint.” Flying Food Grp., Inc. v. NLRB, 471 F.3d 178, 183 (D.C. Cir. 2006) (internal quotation marks omitted); accord Gomez v. Toledo, 446 U.S. 635, 640 (1980). At the Rule 12(b)(6) stage, CREW‘s burden is limited to plausibly pleading that at least some OLC opinions are “working law” and are therefore covered by the reading-room provision—a burden it has neatly carried. It is the government‘s burden, in its turn, to allege and show that some or all of the documents sought are subject to a FOIA exemption.
The majority‘s position is puzzling from a practical standpoint as well. It is hard to see what more might be required to state CREW‘s claim. The parallel litigation in Campaign for Accountability v. U.S. Department of Justice underscores the point. As the majority notes, unlike CREW, plaintiff Campaign for Accountability did amend its complaint to describe subcategories of OLC opinions that it believes are not exempt from FOIA and so must be disclosed. See Am. Compl. 12-19, Campaign for Accountability v. U.S. Dep‘t of Justice, Case No. 1:16-cv-01068, ECF No. 22. Those subcategories are “[o]pinions resolving interagency disputes,” “[o]pinions issued to independent agencies,” “[o]pinions interpreting non-discretionary legal obligations,” “[o]pinions finding that particular statutes are unconstitutional and that therefore agencies need not comply with them,” and “[o]pinions adjudicating or determining private rights.” Id. at 13, 15-16, 18-19. The government there again moved to dismiss the amended complaint, arguing that neither OLC‘s binding opinions generally nor the subcategories the plaintiff identified in its amended complaint are subject to publication under the reading-room provision. See Mem. Supp. Def.‘s Renewed Mot. Dismiss Am. Compl. i-ii, Campaign for Accountability, Case No. 1:16-cv-01068, ECF No. 29-1. The claim posed by the amended complaint in Campaign for Accountability is not materially different from the claim CREW‘s complaint advances.
Contrary to the majority‘s view, the obstacle CREW challenges is not “self-inflicted,” Op. 13. The alternatives the majority proposes—amending the complaint, or seeking documents under FOIA‘s reactive provision, Section 552(a)(3)—are unresponsive to CREW‘s claim that OLC is not complying with its affirmative legal duties under the reading-room provision, Section 552(a)(2). As just noted, Campaign for Accountability effectively reiterates the same claim that CREW already adequately pleads. And the suggestion that plaintiff should proceed via FOIA‘s reactive provision is also off the mark: Shunting plaintiff down a different statutory path that requires it to request particular documents is no answer to its claim for reading-room access to electronic versions of Section 552(a)(2) documents without any FOIA request. The point of the reading-room provision, after all, is to put the burden on agencies to make their “working law” readily available, without request, to anyone who might want to read it.
In sum, the government‘s position is that the body of OLC documents that are subject to the reading-room provision is a null set; CREW‘s position is that it is not. The grounds of CREW‘s claim are clear enough from the face of its complaint. It may or may not be overly ambitious in its request, and the government may or may not be overly protective in its position that no OLC opinions need be published under the reading-room provision. In either event, the claim is adequately pleaded.
*
*
*
“Determining whether a complaint states a plausible claim for relief [is] . . . a context-specific task that requires the reviewing court to draw on its judicial experience and common sense.” Iqbal, 556 U.S. at 679. To survive a motion to dismiss, the complaint need only plausibly allege facts sufficient to ground its challenge to the government‘s zero-disclosure position. I believe it does so. The Rule 8 pleading standard requires no further allegations. Because I conclude that CREW‘s allegations plainly suffice to state