Campaign for Accountability v. United States Department of Justice

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued April 11, 2025 Decided October 17, 2025
No. 24-5163
C AMPAIGN FOR ACCOUNTABILITY,
APPELLEE
v.
UNITED S TATES D EPARTMENT OF J USTICE,
APPELLANT
Consolidated with 24-5170
Appeals from the United States District Court
for the District of Columbia
(No. 1:16-cv-01068)
Daniel Tenny, Attorney, U.S. Department of Justice,
argued the cause for appellant/cross-appellee. With him on the
briefs were Brian M. Boynton, Principal Deputy Assistant
Attorney General, at the time the brief was filed, Brett A.
Shumate, Acting Assistant Attorney General, and Michael S.
Raab, Attorney.

-- 1 of 40 --

2
Alex Abdo argued the cause for appellee/cross-appellant.
With him on the briefs were Anna Diakun, Jackson Busch, and
Jameel Jaffer. Stephanie Krent entered an appearance.
Before: S RINIVASAN, Chief Judge, R AO and P AN , Circuit
Judges.
Opinion for the Court filed by Circuit Judge P AN .
Opinion concurring in the judgment filed by Circuit Judge
R AO .
P AN , Circuit Judge: When Executive Branch officials
need legal advice on matters of importance, they turn to the
Office of Legal Counsel (OLC), a component of the
Department of Justice (DOJ). Since 1950, OLC has provided
guidance to Presidents, Cabinet members, and myriad
government agencies. It has opined on a wide variety of critical
issues — like whether the government may apply “enhanced
interrogation techniques” to suspected terrorists; launch drone
strikes to kill American citizens abroad; and extend federal
benefits to children of same-sex couples. See infra note 1. As
the Executive Branch’s preeminent legal advisor, OLC helps
the President and other government officers to “faithfully
execute[]” the law. U.S. Const. art. II, § 3.
When an agency or official submits a legal question to
OLC for resolution, the Office researches and analyzes the
matter, and may issue a formal, written opinion that answers
the question presented. As a matter of custom, Executive
Branch agencies treat OLC’s legal conclusions as binding. Yet
the public rarely sees those opinions: OLC publishes only a
chosen few.
Campaign for Accountability (CfA) is a non-profit
watchdog group that “uses research, litigation, and

-- 2 of 40 --

3
communications to expose misconduct and malfeasance in
public life.” Compl. ¶ 11 (J.A. 65). It seeks to bring more of
OLC’s opinions into the daylight. Invoking the Freedom of
Information Act (FOIA), CfA sued the Justice Department,
initially seeking disclosure of all OLC opinions. After several
rounds of litigation in the district court, only three categories
of opinions remain in contention: Opinions that (1) resolve
interagency disputes, (2) concern the adjudication or
determination of private rights, and (3) interpret non-
discretionary legal duties.
The district court ruled in DOJ’s favor as to CfA’s requests
for opinions concerning private rights and non-discretionary
legal duties, holding that such opinions are not disclosable
under FOIA. But the court ruled for CfA on its claim that it is
entitled to see opinions resolving interagency disputes. Both
parties appealed. We conclude that FOIA does not require the
disclosure of any of the OLC opinions that CfA seeks. We
therefore affirm in part and reverse in part the judgment of the
district court.
I.
A.
“Under our system of government, Congress makes laws
and the President, acting at times through agencies . . . ,
‘faithfully execute[s]’ them.” Util. Air Regul. Grp. v. EPA, 573
U.S. 302, 327 (2014) (alteration in original) (quoting U.S.
Const. art. II, § 3). In implementing federal laws and
administering the day-to-day operations of the federal
government, the President, Cabinet secretaries, and agencies of
the Executive Branch often encounter challenging legal issues,
such as whether proposed policies or actions conform with the
requirements of the law. Since 1789, Congress has tasked the
Attorney General with providing advice about such legal

-- 3 of 40 --

4
questions to the President and the heads of executive
departments. See 28 U.S.C. §§ 511–512; Judiciary Act of
1789, ch. 20, § 35, 1 Stat. 73. The Attorney General has
delegated that advisory role to an elite office within the Justice
Department: OLC. See 28 C.F.R. § 0.25(a).
OLC provides formal and informal counsel to members of
the Executive Branch. It routinely issues formal written
opinions, which are a “particularly important form of
controlling legal advice.” Memorandum from David J. Barron,
Acting Assistant Attorney General, to Attorneys of the Office
of Legal Counsel 1 (July 16, 2010) (Best Practices Mem.) (J.A.
298). Often, OLC addresses “very difficult and unsettled
issues of law.” Stip. ¶ 6 (J.A. 291). Its advice can shift millions
of dollars in federal funds, determine the viability of a
regulation, or even decide matters of life and death.1 Because
OLC “frequently” weighs in “on issues of first impression that
are unlikely to be resolved by the courts,” its advice “may
1 See Disposition of Proceeds from the Sale of Government
Buildings Acquired with Social Security Trust Funds, 34 Op. O.L.C.
263 (2010); Permissibility of Small Business Administration
Regulations Implementing the Historically Underutilized Business
Zone, 8(a) Business Development, and Service-Disabled Veteran-
Owned Small Business Concern Programs, 2009 WL 2870163
(O.L.C. 2009); Memorandum from David J. Barron, Acting
Assistant Att’y Gen., Re: Applicability of Federal Criminal Laws
and the Constitution to Contemplated Lethal Operations Against
Shaykh Anwar al-Aulaqi (July 16, 2010), https://perma.cc/DYT8-
JWJ7; see also Memorandum from Jay S. Bybee, Assistant Att’y
Gen., Interrogation of al Qaeda Operative (2002),
https://perma.cc/2ZSR-N36W; 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, 31 Op. O.L.C. 243 (2007).

-- 4 of 40 --

5
effectively be the final word on the controlling law.” Best
Practices Mem. 1 (J.A. 298).
As a centralized legal adviser, OLC “occasionally” is called
upon to resolve legal debates between different agencies within
the Executive Branch. Stip. ¶ 7 (J.A. 292); see also Exec.
Order No. 12,146, 44 Fed. Reg. 42657, 42658 (July 18, 1979)
(encouraging agencies that “are unable to resolve a legal
dispute between them . . . to submit the dispute to the Attorney
General”). Exercising the Attorney General’s authority, OLC
“ask[s] each side to submit [] a memorandum” of its views,
“take[s] care to consider fully and address impartially the
points raised on both sides,” and weighs both traditional legal
authorities and its own precedents. Best Practices Mem. 2–4
(J.A. 299–301). As a matter of custom, the parties to an
interagency dispute treat as dispositive an OLC opinion that
settles their disagreement.
The Office also issues opinions in a range of other
circumstances. As relevant to this case, OLC helps agencies
understand their “non-discretionary legal obligations” by
explaining to them what they are required to do by federal law,
see Compl. ¶¶ 41–44 (J.A. 77–78); and the Office helps
agencies determine the “private rights” of individuals who are
affected by their policies, see id. ¶¶ 47–49 (J.A. 80). For
example, one OLC opinion informed the Social Security
Administration that the Defense of Marriage Act would not
prevent the agency from providing benefits to the non-
biological child of a same-sex civil union. See 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, 31
Op. O.L.C. 243 (2007) (J.A. 142–46). That opinion explained
to the agency how it should interpret a statute that it was

-- 5 of 40 --

6
required to implement, while also guiding the agency’s policy
on the provision of benefits to certain individuals.
In preparing formal written opinions, OLC follows an
“ordinary process” set forth in its Best Practices Memo. Stip.
¶ 13 (J.A. 293); see Best Practices Mem. (J.A. 298–303).
Typically, the process begins with a request for advice from
one or more agencies. OLC evaluates the request to ensure that
“[t]he legal question presented [is] focused and concrete; OLC
generally avoids providing a general survey of an area of law
or issuing broad, abstract legal opinions.” Best Practices Mem.
2–3 (J.A. 299–300). In addition, OLC’s opinions “should
address legal questions prospectively; OLC avoids opining on
the legality of past conduct.” Id. at 3 (J.A. 300). If OLC
proceeds with formulating an opinion, it will solicit the views
of interested agencies, research the issue, consult governing
legal authorities and its own precedents, and issue a final,
written opinion. See id. at 2–5 (J.A. 299–302). In arriving at
its conclusions, OLC “must provide advice based on its best
understanding of what the law requires — not simply an
advocate’s defense of the contemplated action or position
proposed by an agency or the Administration.” Id. at 1 (J.A.
298). At the same time, because OLC’s “responsibilities also
include facilitating the work of the Executive Branch and the
objectives of the President, consistent with the law,” “unlike a
court, OLC will, where possible and appropriate, seek to
recommend lawful alternatives to Executive Branch proposals
that it decides would be unlawful.” Id. at 2 (J.A. 299).
Executive Branch officials consider final OLC opinions to
be “controlling legal advice.” Best Practices Mem. 1 (J.A.
298). In practice, that means that agencies customarily accept
OLC’s guidance. Id.; see also Elec. Frontier Found. v. Dep’t
of Just. (EFF), 739 F.3d 1, 9 (D.C. Cir. 2014) (noting that OLC
opinions are “controlling (insofar as agencies customarily

-- 6 of 40 --

7
follow OLC advice that they request)”). To ensure the
requesting agency’s adherence to that custom, OLC will “issue
[its] opinion only if [it] ha[s] received in writing from that
agency an agreement that it will conform its conduct to
[OLC’s] conclusion.” Best Practices Mem. 3 (J.A. 300).2
OLC may publicly release its written opinions but usually
declines to do so. It bases its decision to publish on the nature
of the opinion, often considering whether the opinion is
significant enough, whether it implicates national security, and
whether publication would compromise the confidentiality of
attorney-client information. Best Practices Mem. 5–6 (J.A.
302–03). Although mostly hidden from public view, OLC’s
opinions, together with those issued by the Attorneys General,
may “comprise the largest body of official interpretation of the
Constitution and statutes outside the volumes of the federal
court reporters.” Trevor W. Morrison, Stare Decisis in the
Office of Legal Counsel, 110 Colum. L. Rev. 1448, 1451
(2010) (quoting John O. McGinnis, Models of the Opinion
Function of the Attorney General: A Normative, Descriptive,
and Historical Prolegomenon, 15 Cardozo L. Rev. 375, 376
(1993)).
B.
In 1966, Congress enacted FOIA, empowering “citizens to
know what their Government is up to.” Nat’l Archives & Recs.
Admin. v. Favish, 541 U.S. 157, 171 (2004) (cleaned up).
Specifically, FOIA “provide[s] a statutory right of public
access to documents and records held by agencies of the federal
government.” Pratt v. Webster, 673 F.2d 408, 413 (D.C. Cir.
2 OLC has clarified, and CfA does not dispute, that the Office
enforces this requirement with respect to independent and non-
independent Executive Branch agencies.

-- 7 of 40 --

8
1982). The statute’s “basic purpose . . . is to ensure an informed
citizenry, vital to the functioning of a democratic society,
needed to check against corruption and to hold the governors
accountable to the governed.” NLRB v. Robbins Tire & Rubber
Co., 437 U.S. 214, 242 (1978). In FOIA, Congress “sought to
reach a workable balance between the right of the public to
know and the need of the Government to keep information in
confidence to the extent necessary without permitting
indiscriminate secrecy.” John Doe Agency v. John Doe Corp.,
493 U.S. 146, 152 (1989) (cleaned up).
A plaintiff may sue an agency for violating FOIA “upon a
showing that an agency has (1) ‘improperly’; (2) ‘withheld’;
(3) ‘agency records.’” Kissinger v. Reps. Comm. for Freedom
of the Press, 445 U.S. 136, 150 (1980) (quoting 5 U.S.C.
§ 552(a)(4)(B)). “An agency withholds its records
‘improperly’ if it fails to comply with one of FOIA’s
‘mandatory disclosure requirements.’” Citizens for Resp. &
Ethics in Washington v. Dep’t of Just. (CREW II), 922 F.3d
480, 486 (D.C. Cir. 2019) (quoting Dep’t of Just. v. Tax
Analysts, 492 U.S. 136, 150 (1989)).
FOIA requires agencies to disclose records in three ways,
which are subject to certain exceptions or exclusions:
First, FOIA requires agencies to publish certain kinds of
documents in the Federal Register. See 5 U.S.C. § 552(a)(1).
These include, for example, “rules of procedure” and
“substantive rules of general applicability adopted as
authorized by law.” Id. § 552(a)(1)(C)–(D).
Second, FOIA directs agencies to disclose and index
certain records proactively, by making them “available for
public inspection” in an electronic reading room. 5 U.S.C.
§ 552(a)(2). Subsection (A) of the so-called reading-room
provision pertains to “final opinions . . . made in the

-- 8 of 40 --

9
adjudication of cases.” Id. § 552(a)(2)(A). And subsection (B)
concerns “those statements of policy and interpretations which
have been adopted by the agency and are not published in the
Federal Register.” Id. § 552(a)(2)(B). Agencies must make all
documents covered by either subsection available to the public,
“even absent a specific request.” CREW II, 922 F.3d at 484
(citing 5 U.S.C. § 552(a)(2)).
Third, the statute calls for agencies to disclose “all other
records” “upon request.” Kennecott Utah Copper Corp. v.
Dep’t of Interior, 88 F.3d 1191, 1202 (D.C. Cir. 1996) (citing
5 U.S.C. § 552(a)(3)). This is “FOIA’s ‘most familiar
provision.’” CREW II, 922 F.3d at 484 (quoting Citizens for
Resp. & Ethics in Washington v. Dep’t of Just. (CREW I), 846
F.3d 1235, 1240 (D.C. Cir. 2017)). The majority of our
opinions interpreting FOIA address “FOIA requests” made by
members of the public under subsection (a)(3).
An agency may withhold a document that is otherwise
subject to disclosure under one of FOIA’s nine exemptions.
See 5 U.S.C. § 552(b)(1)–(9). As potentially relevant here,
Exemption 5, id. § 552(b)(5), incorporates the deliberative
process privilege, which shields from disclosure pre-decisional
and deliberative “documents reflecting advisory opinions,
recommendations and deliberations comprising part of a
process by which governmental decisions and policies are
formulated,” Fish & Wildlife Serv. v. Sierra Club, Inc., 592
U.S. 261, 267 (2021) (quoting NLRB v. Sears, Roebuck & Co.,
421 U.S. 132, 150 (1975)).
The deliberative process privilege does not protect from
disclosure an agency’s “working law,” “i.e. the ‘reasons which
[supplied] the basis for an agency policy actually adopted.’”
EFF, 739 F.3d at 7 (alteration in original) (quoting Sears, 421
U.S. at 152–53). To qualify as working law, the agency records

-- 9 of 40 --

10
in question must “have the force and effect of law.” Sears, 421
U.S. at 153 (cleaned up). Working law encompasses the
documents that fall under subsection (A) and subsection (B) of
the reading-room provision: “‘final opinions . . . made in the
adjudication of cases’ and ‘statements of policy and
interpretations which have been adopted by the agency.’”
CREW II, 922 F.3d at 486 (quoting 5 U.S.C. § 552(a)(2)(A)–
(B)). Such documents are subject to disclosure under FOIA
and generally do not fall under the deliberative process
privilege because they reflect post-decisional, final agency
determinations. See Sears, 421 U.S. at 153–54 (“We should be
reluctant . . . to construe Exemption 5 to apply to the
documents described in 5 U.S.C. § 552(a)(2); and with respect
at least to ‘final opinions,’ which not only invariably explain
agency action already taken or an agency decision already
made, but also constitute ‘final dispositions’ of matters by an
agency, we hold that Exemption 5 can never apply.” (cleaned
up)).
We have previously considered whether an OLC legal
opinion that is prepared at the behest of a client agency is
disclosable under FOIA. In EFF, we addressed whether OLC
was required to disclose an opinion about the FBI’s past
information-gathering practices in response to a FOIA request
under 5 U.S.C. § 552(a)(3), while in CREW II, we considered
an argument that the reading-room provision required OLC to
disclose all its formal written opinions. See EFF, 739 F.3d at
4; CREW II, 922 F.3d at 486. In EFF, OLC had failed to
comply with subsection (a)(3), yet we held that the requested
opinion was shielded from disclosure by the deliberative
process privilege because it had not been adopted by the
requesting agency as its working law. See EFF, 739 F.3d at 8–
10. In CREW II, we built on EFF to conclude that OLC
opinions are not categorically disclosable under the reading-
room provision, even if they are generally considered

-- 10 of 40 --

11
“controlling, authoritative and binding” within the Executive
Branch. CREW II, 922 F.3d at 486 (cleaned up) (citing EFF,
739 F.3d at 9). As we explained in EFF, because OLC does
not have “the authority to establish the ‘working law’ of [an
agency],” its opinion cannot be the working law of an agency
unless the agency “‘adopt[s]’ what OLC offer[s].” EFF, 739
F.3d. at 8–10. In other words, “[a]n OLC opinion [that includes
statements of policy and interpretations] qualifies as the
working law of an agency only if the agency has adopted the
opinion as its own.” CREW II, 922 F.3d at 486 (cleaned up)
(citing EFF, 739 F.3d at 9).
C.
In June 2016, CfA sued the Justice Department, seeking
disclosure of all OLC’s past and future formal written opinions
under FOIA’s reading-room provision, 5 U.S.C. § 552(a)(2).
At that time, we had not yet issued our decision in CREW II,
which addresses an identical claim. The district court relied on
our holding in EFF to reject CfA’s sweeping request and to
dismiss its Complaint for failure to state a claim under Federal
Rule of Civil Procedure 12(b)(6). See Campaign for
Accountability v. Dep’t of Just. (CfA I), 278 F. Supp. 3d 303,
320–23 (D.D.C. 2017).
CfA then filed a narrower Amended Complaint. See
Amend. Compl. (Compl.) (J.A. 62–85). As relevant here, it
asserted that three sets of OLC opinions are subject to
disclosure: (1) “opinions resolving interagency disputes”;
(2) “opinions interpreting non-discretionary legal obligations”;
and (3) “opinions adjudicating or determining private rights.”
Id. ¶¶ 35–38, 41–44, 47–49 (J.A. 74–75, 77–78, 80) (cleaned
up).
DOJ filed a new motion to dismiss, which the district court
granted in part and denied in part. Campaign for

-- 11 of 40 --

12
Accountability v. Dep’t of Just. (CfA II), 486 F. Supp. 3d 424,
426 (D.D.C. 2020). The district court dismissed CfA’s claims
concerning OLC opinions that concern non-discretionary
obligations and private rights. See id. at 441–45. It held that
such opinions are not subject to disclosure as “working law”
because they cannot be classified as either “final opinions . . .
made in the adjudication of cases” under subsection (A) of the
reading-room provision, or “statements of policy and
interpretations which have been adopted by the agency” under
subsection (B). See id. at 437 (citing 5 U.S.C. § 552(a)(2)(A)–
(B)). Specifically, the district court reasoned that opinions
interpreting non-discretionary duties do not “necessarily
announce[] the client agency’s working law from the moment
the opinion is issued.” Id. at 442 (emphasis in original).
Rather, an agency would “need[] to adopt OLC’s legal
interpretation by acting upon it in the context of its adjudication
or policy development in order for that opinion to be plausibly
considered the [agency’s] working law.” Id. (citing EFF, 739
F.3d at 9). The court also explained that opinions concerning
private rights may, “[a]t most,” be “plausibly conceived of as
‘binding’ and ‘controlling’ statements of relevant law that the
client agency may adopt and apply to a particular case
adjudicating private rights.” Id. at 443 (emphasis in original)
(quoting EFF, 739 F.3d at 9). Such opinions would become
disclosable as working law only after the agency adopted the
OLC’s interpretation as its own. Id.
The district court took a different view about OLC opinions
that resolve interagency disputes. It held that CfA had
plausibly alleged that such opinions fall under the reading-
room provision and must be disclosed. CfA II, 486 F. Supp. 3d
at 426. First, the court concluded that such opinions are
plausibly “final opinions . . . made in the adjudication of cases”
under subsection (A) because OLC’s “process for resolving
such disputes is adjudicative in nature” and a “case” need not

-- 12 of 40 --

13
“involve private parties.” Id. at 439 (citing 5 U.S.C.
§ 552(a)(2)(A)). Second, the court ruled that interagency-
dispute opinions are plausibly “‘statements of policy and
interpretations’ that have been adopted ex ante by at least one
of the disputing agencies.” Id. at 426 (quoting 5 U.S.C.
§ 552(a)(2)(B)). Agencies may be viewed as adopting such
opinions in advance, the court reasoned, because an agency
must provide written assurance “that it will conform its conduct
to [OLC’s] conclusion” before OLC will issue an opinion. Id.
at 440 (quoting Best Practices Mem. 3 (J.A. 300)).
After the district court ruled on the motions to dismiss, the
case was transferred to a different district court judge.
Proceeding on a stipulated record, the parties filed cross-
motions for summary judgment concerning whether OLC
opinions that resolve interagency disputes must be disclosed as
working law under FOIA’s reading-room provision. The
district court granted summary judgment in favor of CfA,
holding that such opinions must be disclosed. See Campaign
for Accountability v. Dep’t of Just. (CfA III), 732 F. Supp. 3d
63 (D.D.C. 2024). The court ruled that OLC opinions that
resolve interagency disputes are “final opinions . . . made in the
adjudication of cases” under subsection (A) of the reading-
room provision, essentially because OLC’s resolution process
“bears many of the hallmarks of adversarial adjudication.” Id.
at 70–76 (quoting 5 U.S.C. § 552(a)(2)(A)).
The district court rejected OLC’s argument that the
opinions are “categorically protected” by the deliberative
process privilege, noting that the deliberative process privilege
and the reading-room provision are “generally mutually
exclusive.” CfA III, 732 F. Supp. 3d at 71, 75–76 (citing Sears,
421 U.S. at 153). The court declined to reach whether opinions
resolving interagency disputes also qualify as “statements of
policy and interpretations which have been adopted by the

-- 13 of 40 --

14
agency” under subsection (B) of the reading-room provision.
Id. at 65 (citing 5 U.S.C. § 552(a)(2)(B)).
DOJ timely appealed the court’s ruling that OLC must
disclose opinions resolving interagency disputes; CfA timely
cross-appealed the court’s dismissal of its claims regarding
opinions on non-discretionary duties and private rights. We
have jurisdiction under 28 U.S.C. § 1291.
II.
A.
We “review[] 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.” CREW II, 922 F.3d
at 486. The same goes for “decisions granting summary
judgment in FOIA cases.” EFF, 739 F.3d at 7. We also review
de novo the district court’s interpretation of FOIA. See Jud.
Watch, Inc. v. FBI, 522 F.3d 364, 367 (D.C. Cir. 2008).
B.
Although OLC does not contest CfA’s standing to bring
this suit, “we have an independent obligation to assure
ourselves of our jurisdiction.” Waterkeeper All., Inc. v. Regan,
41 F.4th 654, 659 (D.C. Cir. 2022).
CfA has standing to challenge OLC’s compliance with
FOIA if it can “demonstrate (i) that [it] has suffered or likely
will suffer an injury in fact, (ii) that the injury likely was caused
or will be caused by the defendant, and (iii) that the injury
likely would be redressed by the requested judicial relief.”
FDA v. All. for Hippocratic Med., 602 U.S. 367, 380 (2024).
An injury must be both “‘concrete,’ meaning that it must be
real and not abstract” and “particularized,” meaning it “must

-- 14 of 40 --

15
affect ‘the plaintiff in a personal and individual way’ and not
be a generalized grievance.” Id. at 381 (cleaned up) (first
quoting TransUnion LLC v. Ramirez, 594 U.S. 413, 424
(2021); then quoting Lujan v. Defs. of Wildlife, 504 U.S. 555,
560 (1992)). Supreme Court precedents permit a plaintiff to
establish standing based on a concrete and particularized
“informational injury,” which we have interpreted to mean that
the plaintiff can show “(1) it has been deprived of information
that a statute requires the government to disclose to it, and (2)
it suffers, by being denied access to that information, the type
of harm Congress sought to prevent by requiring disclosure.”
Nat’l Sec. Archive v. CIA, 104 F.4th 267, 272 (D.C. Cir. 2024)
(cleaned up); see also TransUnion, 594 U.S. at 441 (citing FEC
v. Akins, 524 U.S. 11 (1998) and Pub. Citizen v. Dep’t of
Justice, 491 U.S. 440 (1989)).
Here, the standing question is straightforward because “the
Supreme Court has held that an informational injury is
sufficient to satisfy standing under FOIA.” Nat’l Sec. Archive,
104 F.4th at 272 (citing Pub. Citizen, 491 U.S. at 449). A FOIA
plaintiff need only show “that they sought and were denied
specific agency records.” Pub. Citizen, 491 U.S. at 449; see
also Zivotofsky ex rel. Ari Z. v. Sec’y of State, 444 F.3d 614,
617 (D.C. Cir. 2006) (noting in the context of 5 U.S.C.
§ 552(a)(3) that “[a]nyone whose request for specific
information has been denied has standing to bring an action”).
CfA has made that showing: Its Amended Complaint alleges
that CfA invoked FOIA to request the disclosure of the OLC
opinions at issue, and that OLC denied the request.
We do not think the analysis changes because CfA relies on
the reading-room provision, which makes certain records
accessible to all, even absent a request for disclosure. Although
we recognized in Prisology, Inc. v. Federal Bureau of Prisons
that failure to publish records as required by the reading-room

-- 15 of 40 --

16
provision is a “harm common to everyone” and cannot, by
itself, form the basis of standing, a plaintiff suing to enforce the
reading-room provision still has standing if the plaintiff “made
a request of the agency and the agency denied the request.” 852
F.3d 1114, 1116–17 (D.C. Cir. 2017). Prisology specifically
acknowledged that standing was established in another case
where the plaintiff made a request under the reading-room
provision for OLC opinions and “claimed that the agency’s
refusal to make documents available . . . harmed its ‘core
programmatic activities,’ which included research from public
government records.” Id. at 1117 (quoting Amend. Compl.
¶¶ 6, 11, Citizens for Resp. & Ethics in Washington v. Dep’t of
Just., 164 F. Supp. 3d 145 (D.D.C. 2016) (No. 13-01291)).
CfA makes that precise claim here. Compl. ¶ 52 (J.A. 81–82)
(alleging that lack of disclosure “has harmed, and continues to
harm, CfA in carrying out its core programmatic activities,”
including researching government records).
We respectfully disagree with our concurring colleague,
who would hold that CfA lacks standing because it has failed
to demonstrate a particularized injury. In her view, when a
FOIA plaintiff requests (and is denied) access to OLC opinions,
the court’s standing analysis differs based on the type of OLC
opinions at issue. If a plaintiff requests (and is denied) OLC
opinions under subsection (a)(3), the plaintiff has suffered an
“informational injury” that is sufficiently particularized. See
Concurring Op. 4 (citing Prisology, 852 F.3d at 1117 for the
proposition that “under the record request provision our
caselaw allows standing based on an agency’s denial of a
request for records”). On that much we agree. But when a
plaintiff requests (and is denied) OLC opinions under the
reading-room provision, subsection (a)(2) — as is the case here
— our colleague believes that more is required. According to
our colleague, because a successful subsection (a)(2) request
would require OLC to make certain opinions “available for

-- 16 of 40 --

17
public inspection” in an electronic reading room, 5 U.S.C.
§ 552(a)(2) (emphasis added), the plaintiff’s injury is best
understood as a “generalized grievance,” Concurring Op. 5. In
other words, because OLC has withheld the requested records
from the public it has not withheld the requested records “from
the complainant.” Id. at 6 (emphasis in original) (citing 5
U.S.C. § 552(a)(4)(B)). Our colleague would therefore require
CfA to show more than an “informational injury”: She believes
that CfA must also identify a “downstream” or “adverse” effect
from OLC’s failure to publish the requested opinions under
subsection (a)(2). Id. at 5.
We disagree with our colleague’s analysis for three
reasons. First, as previously explained, Prisology did not turn
on whether records are sought under subsection (a)(2) versus
subsection (a)(3); rather, the pertinent inquiry was whether the
plaintiff expressly requested (and was denied) access to the
records. Prisology, 852 F.3d at 1117. Because the plaintiff in
that case “made no request of the Bureau of Prisons before
bringing suit and therefore received no denial from that
agency,” he had not suffered the type of “informational injury”
that is necessary to establish standing. Id. (distinguishing the
plaintiff’s case from those in which “the plaintiff made a
request of the agency and the agency denied the request”).
Second, our colleague’s contention that CfA has alleged a
mere “public” or “generalized grievance” because subsection
(a)(2) records must be made “available for public inspection”
in an electronic reading room, 5 U.S.C. § 552(a)(2) (emphasis
added), is incorrect. The fact that many people could be
similarly injured does not render the claim an impermissible
generalized grievance: “Where a harm is concrete, though
widely shared, the Court has found injury in fact.” Akins, 524
U.S. at 11 (cleaned up); see also Pub. Citizen, 491 U.S. at 449–
50 (“The fact that other citizens or groups of citizens might

-- 17 of 40 --

18
make the same complaint after unsuccessfully demanding
disclosure . . . does not lessen [their] asserted injury.”);
Massachusetts v. EPA, 549 U.S. 497, 525 n.24 (2007)
(“Standing is not to be denied simply because many people
suffer the same injury.” (cleaned up)).
And third, our colleague’s reliance on TransUnion is
misplaced. The plaintiffs in that case, like the plaintiff in
Prisology, “did not allege that they failed to receive any
required information.” TransUnion, 594 U.S. at 441. Indeed,
the Supreme Court specifically distinguished TransUnion from
cases that “involve[] denial of information subject to public-
disclosure or sunshine laws that entitle all members of the
public to certain information” — i.e., laws like FOIA. Id.
(emphasis added). The Supreme Court has “never suggested
that those requesting information under [FOIA] need show
more than that they sought and were denied specific agency
records. There is no reason for a different rule here.” Pub.
Citizen, 491 U.S. at 449–50.
Because CfA alleges that it asked for particular agency
records under the reading-room provision, that OLC denied its
request, and that the denial harms CfA’s ability to carry out its
“core programmatic activities,” Compl. ¶ 52 (J.A. 81–82), it
has established standing.
C.
The district court dismissed CfA’s claims that OLC was
required to disclose its legal opinions related to the
adjudication of private rights or the interpretation of non-
discretionary duties; but it held that OLC must disclose
opinions resolving interagency disputes. In our view, none of
the documents requested by CfA are “working law” subject to
disclosure under FOIA’s reading-room provision.

-- 18 of 40 --

19
As previously discussed, OLC opinions qualify as the
working law of an agency under the relevant subsections of the
reading-room provision if they take the form of “final
opinions . . . in the adjudication of cases,” or “statements of
policy and interpretations which have been adopted by the
agency.” CREW II, 922 F.3d at 486 (alteration in original)
(quoting 5 U.S.C. § 552(a)(2)(A)–(B)). Because none of the
categories of opinions at issue here meet either of those
requirements, CfA failed to adequately allege or show that
OLC violated the reading-room provision when it withheld the
requested documents.
1. Subsection (A): “Made in the Adjudication of
Cases”
CfA contends that two of the three categories of OLC
opinions at issue — those resolving interagency disputes and
those involving private rights — are subject to disclosure
because they encompass “final opinions . . . made in the
adjudication of cases” under subsection (A) of FOIA’s reading-
room provision. 5 U.S.C. § 552(a)(2)(A). As to opinions that
resolve interagency disputes, CfA argues that such opinions are
“final” and essentially “adjudicat[e]” “cases” between
agencies. Meanwhile, OLC opinions concerning private rights
are final and made in the client agency’s adjudication of the
cases before it, according to CfA, because OLC’s advice
controls the agency’s resolution of those cases. We disagree
on both counts. OLC’s process for issuing legal opinions to
Executive Branch agencies and officials is not tantamount to
an “adjudication of cases” because the opinion process
generally involves no “final disposition,” yields only
prospective advice, and does not resolve a “case.” Id.
The definition of “adjudication” under subsection (A) is
found in the Administrative Procedure Act (APA), which

-- 19 of 40 --

20
encompasses FOIA. Sears, 421 U.S. at 158; Chiquita Brands
Int’l Inc. v. SEC, 805 F.3d 289, 295–96 (D.C. Cir. 2015).
Under the APA, an “adjudication” is an “agency process for the
formulation of an order.” 5 U.S.C. § 551(7). In turn, an
“order” is “the whole or a part of a final disposition . . . of an
agency in a matter other than rule making but including
licensing.” Id. § 551(6).
The Supreme Court has held that a “final disposition” must
“have some determinate consequences for the party to the
proceeding.” Int’l Tel. & Tel. Corp. v. Loc. 134, Int’l Bhd. of
Elec. Workers, AFL-CIO (ITT), 419 U.S. 428, 443 (1975). A
decision that “did not finally decide anything” is not a “final
disposition.” Sears, 421 U.S. at 159 (citing ITT, 419 U.S. 428).
For instance, an NLRB investigatory proceeding is not an
“adjudication” because the resulting decision, “standing alone,
binds no one” and at the conclusion of the proceeding, the
agency “does not order anybody to do anything.” ITT, 419 U.S.
at 443–44. Nor does an “ordinary litigation decision,” such as
an agency’s decision on how to respond to a subpoena, qualify
as a final disposition. Watts v. SEC, 482 F.3d 501, 506 (D.C.
Cir. 2007); see also id. (observing that an “internal agency
process for reaching a decision . . . is not typically or
comfortably described as an ‘adjudication’ (even given the
broad scope of formal and informal adjudications under the
APA)”).
Here, DOJ acknowledges that “[e]very OLC opinion is . . . ,
by custom and practice of the Executive Branch, treated as
controlling, authoritative, and binding.” Gov’t Reply 10. But
OLC’s control over an agency’s legal interpretations does not
give it authority to direct or implement the agency’s policies.
See EFF, 739 F.3d at 9–10. In short, OLC opinions are not
“orders” — they do not “order anybody to do anything.” ITT,
419 U.S. at 443. Instead, an OLC opinion offers advice. For

-- 20 of 40 --

21
example, it might give a client agency a menu of lawful
options, see Best Practices Mem. 2 (J.A. 299); or it might
“describe[] the legal parameters of what the [agency] is
permitted to do,” but leave the agency “free to decline to adopt
the [practice] deemed legally permissible in the OLC Opinion,”
EFF, 739 F.3d at 10 (emphasis omitted). One typical opinion,
cited by CfA, concluded that the Environmental Protection
Agency may assess civil penalties against federal agencies, but
it did not require the agency to do so. See Clean Air Act Op.,
21 Op. O.L.C. 109, 109 (J.A. 481). Because an OLC opinion
is not a “final disposition” or an “order,” it does not involve
“adjudication.”
Nor does an opinion that memorializes OLC’s advice
resolve any “case.” In interpreting subsection (A), we have
“emphasized that the ability of a third party to participate as a
party and to obtain personal relief in a proceeding bears
significantly on the determination whether, for purposes of
FOIA’s affirmative disclosure requirement, the proceeding
amounts to an adjudication of a case culminating in a final
order.” Am. Immigr. Laws. Ass’n v. Exec. Off. for Immigr.
Rev., 830 F.3d 667, 679 (D.C. Cir. 2016) (cleaned up). When
an agency seeks advice from OLC about a matter that affects
the rights of private individuals, those third parties do not
participate in OLC’s decision-making process and they obtain
no “personal relief” from it. Id. Nor can the agency itself be
considered a party before OLC that obtains “personal relief.”
Id. (noting that the reading-room provision “refer[s] to final
opinions resulting from proceedings in which a party has a right
to set the agency decision-making process in motion and obtain
a determination concerning the statute or other laws the agency
is charged with interpreting and administering” (cleaned up)).
The prospective nature of OLC’s advice confirms that it
does not engage in the “adjudication of cases.” One of the

-- 21 of 40 --

22
“principal distinctions between rulemaking and adjudication”
is that “rulemaking governs only the future, whereas
adjudications ‘immediately bind parties by retroactively
applying law to their past actions.’” ITServe All., Inc. v. DHS,
71 F.4th 1028, 1035 (D.C. Cir. 2023) (quoting Safari Club Int’l
v. Zinke, 878 F.3d 316, 333 (D.C. Cir. 2017) and citing Bowen
v. Georgetown Univ. Hosp., 488 U.S. 204, 216 (1988) (Scalia,
J., concurring)). Indeed, “it is black-letter administrative law
that adjudications are inherently retroactive.” Cath. Health
Initiatives Iowa Corp. v. Sebelius, 718 F.3d 914, 921 (D.C. Cir.
2013). OLC’s opinions, however, “address legal questions
prospectively; OLC avoids opining on the legality of past
conduct (though from time to time [it] may issue prospective
opinions that confirm or memorialize past advice or that
necessarily bear on past conduct in addressing an ongoing legal
issue).” Best Practices Mem. 3 (J.A. 300). OLC opinions
therefore are not “final opinions . . . made in the adjudication
of cases” under subsection (A) of the reading-room provision.
Consistent with the foregoing principles, the district court
correctly dismissed CfA’s claims under subsection (A) with
respect to opinions that concern the adjudication of private
rights. As the district court recognized, OLC lacks “authority
to ‘adjudicat[e]’ the rights of the private individuals whom
OLC’s client agencies regulate.” CfA II, 486 F. Supp. 3d at 443
(alteration and emphasis in original) (quoting 5 U.S.C.
§ 552(a)(2)(A)). Even if, as CfA argues, OLC is involved in
the agency’s adjudicative process and even if its opinion
controls the agency’s legal conclusion, the “responsible
decisionmaker in an agency’s decision-making process”
remains the agency that, as a matter of law and fact, actually
carries out the adjudication. Bristol-Meyers Co. v. FTC, 598
F.2d 18, 25 (D.C. Cir. 1978).

-- 22 of 40 --

23
The district court erred, however, when it concluded on
summary judgment that OLC opinions that resolve interagency
disputes must be disclosed because OLC’s opinion process
“bears many of the hallmarks of adversarial adjudication.” CfA
III, 723 F. Supp. 3d at 73. The district court reasoned that (1)
OLC issues “written opinions only for [interagency] disputes
that are ‘concrete and ongoing,’” and (2) “the process by which
the OLC researches, drafts, and finalizes its opinions that
resolve interagency disputes is standardized, thorough, and
bears many of the hallmarks of adversarial adjudication.” Id.
(quoting Best Practices Mem. 3 (J.A. 300)). In particular, the
court noted that OLC “solicits a ‘detailed’ brief” from each
party to the dispute, “functions like a neutral decisionmaker,”
applies both its own precedents and governing legal authorities,
and then treats the final opinion as precedential. Id. (quoting
Best Practices Mem. 3–5 (J.A. 300–02)). But even if OLC’s
role and procedures bear some resemblance to “adversarial
adjudication,” OLC does not actually engage in the
“adjudication of cases,” as defined by the APA and our
caselaw.
In sum, OLC opinions resolving interagency disputes or
concerning private rights are not subject to disclosure as “final
opinions . . . made in the adjudication of cases.” 5 U.S.C.
§ 552(a)(2)(A).
2. Subsection (B): “Adopted by the Agency”
CfA also contends that all three categories of opinions at
issue must be disclosed under subsection (B) of FOIA’s
reading-room provision because they are “statements of policy

-- 23 of 40 --

24
and interpretations which have been adopted by” OLC’s client
agencies. 5 U.S.C. § 552(a)(2)(B). We again disagree.3
In CREW II and EFF, we established that “[a]n OLC
opinion [that includes statements of policy and interpretations]
qualifies as the working law of an agency only if the agency
has adopted the opinion as its own.” CREW II, 922 F.3d at 486
(cleaned up) (citing EFF, 739 F.3d at 9). Because “OLC [does]
not have the authority to establish the ‘working law’ of
[agencies],” an OLC opinion “amounts to advice offered by
OLC for consideration by officials of the” agency and “is not
the law of an agency unless the agency adopts it.” EFF, 739
F.3d at 8. Importantly, an OLC opinion is not adopted simply
because it is “controlling,” “authoritative,” and “binding,” as
those qualities “do[] not overcome the fact that OLC does not
speak with authority on the agency’s policy.” CREW II, 922
F.3d at 486 (cleaned up) (quoting EFF, 739 F.3d at 9). Rather,
in considering whether an agency has adopted legal advice as
its working law, courts focus on what the agency
decisionmaker has done with that advice. See Sears, 421 U.S.
at 152–53 (observing that “the reasons which did supply the
3 The district court did not rely on subsection (B) when it
concluded on summary judgment that OLC opinions resolving
interagency disputes must be disclosed under the reading-room
provision. See CfA III, 732 F. Supp. 3d at 70 (citing 5 U.S.C.
§ 552(a)(2)(A)–(B)). But CfA asks us to affirm on that ground. See
Polm Fam. Found., Inc. v. United States, 644 F.3d 406, 408 (D.C.
Cir. 2011) (“A prevailing party may defend its judgment on any
ground properly raised below whether or not that ground was relied
upon, rejected, or even considered by the District Court.” (cleaned
up)). CfA also asks us to reverse the district court’s dismissal of its
claims related to opinions concerning non-discretionary legal duties
and the determination of private rights based on subsection (B). See
CfA II, 486 F. Supp. 3d at 441–45.

-- 24 of 40 --

25
basis for an agency policy actually adopted . . . if expressed
within the agency, constitute the ‘working law’ of the
agency”). An agency adopts memoranda when “in practice,
[the memoranda] are authoritative Agency decisions in the
cases to which they are addressed and . . . in practice, [they]
also guide subsequent Agency rulings.” Schlefer v. United
States, 702 F.2d 233, 244 (D.C. Cir. 1983) (citing 5 U.S.C.
§ 552(a)(2) and Sears, 421 U.S. at 149).
As a general matter, OLC opinions are not automatically
“adopted by the agency” under subsection (B) of the reading-
room provision. 5 U.S.C. § 552(a)(2)(B). Even though all
OLC opinions are considered authoritative within the
Executive Branch, the requesting agency does not necessarily
adopt every opinion as its own working law. As already
discussed, OLC opinions generally tell agencies what they can
or cannot do, not what they must do. An agency does not adopt
such an opinion until it actually takes action in accordance with
OLC’s advice. See Schlefer, 702 F.2d at 244. Only at that point
could the opinion become the agency’s working law.
Moreover, if OLC determines that an agency’s proposed
action is unlawful, it may “seek to recommend lawful
alternatives.” Best Practices Mem. 2 (J.A. 299). In one
instance, OLC resolved an interagency dispute by concluding
that EPA could permissibly choose between three
interpretations of a particular statute. See Scope of the
Environmental Protection Agency’s Discretion to Adopt Any
One of Three Alternative Interpretations of the Mitchell-Conte
Amendment to the Clean Air Act, 13 Op. O.L.C. 105, 105
(1989). Such an opinion tells the agency only what it “is
permitted to do” and does not control its policy. EFF, 739 F.3d
at 10 (emphasis in original). That type of menu-of-choices
memorandum cannot be adopted as the agency’s working law
without further agency action because it does not state a policy

-- 25 of 40 --

26
that can “have the force and effect of law.” Sears, 421 U.S. at
153. Thus, as a rule, no category of OLC opinions is
automatically and always “adopted by the agency” under
subsection (B) of the reading-room provision.
CfA’s argument to the contrary draws on the district
court’s suggestion, made when ruling on the government’s
motion to dismiss, that opinions resolving interagency disputes
might be adopted by the requesting agencies ex ante — i.e.,
before the opinions are issued. See CfA II, 486 F. Supp. 3d at
440. CfA posits that OLC’s practice of securing up-front
commitments from the requesting agencies that they will treat
OLC’s advice as controlling serves to effectuate an ex ante
adoption of any ensuing opinion.
We find CfA’s theory unpersuasive. Legal advice does not
become working law when it is solicited by the agency, even if
the agency is supposed to follow the advice. See Murphy v.
Dep’t of Army, 613 F.2d 1151, 1153–54 & n.9 (D.C. Cir. 1979).
Instead, the test is whether the agency ultimately applies the
advice as its working law. See Schlefer, 702 F.2d at 244. That
standard does not allow for ex ante adoption.
Indeed, our prior cases recognizing an agency’s adoption
of legal advice as working law focus on the agency’s actions
after the advice was given. For example, in Taxation With
Representation Fund v. IRS, we held that the agency
decisionmaker “transform[ed]” “merely advisory documents”
from the agency’s counsel into “documents that reflect the
agency’s current position on a given issue” by distributing
them “to subordinate attorneys and other staff persons within

-- 26 of 40 --

27
the agency to be used as interpretative guides and research
tools.” 646 F.2d 666, 683 (D.C. Cir. 1981).4
Moreover, the “ex ante adoption” theory conflicts with
EFF by improperly focusing on the controlling, authoritative,
or binding nature of OLC opinions. Even if the requesting
agencies “necessarily agree that their legal interpretations and
policy positions will be controlled by and conformed to the
OLC’s resolution,” CfA Br. 27, that is just another way of
saying that OLC’s opinions are “controlling.” We squarely
held in EFF that a “controlling” or “authoritative” OLC
opinion is not working law until the agency actually adopts the
opinion. See EFF, 739 F.3d at 9. That is because OLC itself
is not “authorized to make decisions about” the agency’s
policies. Id. For these reasons, OLC’s opinions are not
4 See also Coastal States Gas Corp. v. Dep’t of Energy, 617 F.2d
854, 869 (D.C. Cir. 1980) (holding that the deliberative process
privilege does not extend to certain documents that “in practice
represent interpretations of established policy on which the agency
relies in discharging its regulatory responsibilities”); Tax Analysts v.
IRS (Tax Analysts I), 117 F.3d 607, 617 (D.C. Cir. 1997) (holding
that “[t]he legal conclusions the [IRS] Office of Chief Counsel
provides to field personnel constitute agency law, even if those
conclusions are not formally binding,” as the documents are
“‘routinely used’ and relied upon by field personnel” (quoting
Coastal States, 617 F.2d at 869)); Tax Analysts v. IRS (Tax Analysts
II), 294 F.3d 71, 81 (D.C. Cir. 2002) (reaching the same conclusion
about similar advice memoranda that “simply explain and apply
established policy” (quoting Coastal States, 617 F.2d at 869));
Schlefer, 702 F.2d at 244 (holding that opinions interpreting statutes
administered by the agency are not subject to the deliberative process
privilege because they “are authoritative Agency decisions in the
cases to which they are addressed and . . . also guide subsequent
Agency rulings”).

-- 27 of 40 --

28
categorically adopted, ex ante or otherwise, by OLC’s client
agencies.5
The district court correctly applied the foregoing
principles when it rejected CfA’s arguments that the remaining
categories of OLC opinions constitute working law under
subsection (B) of the reading-room provision. CfA
nevertheless argues that (1) opinions interpreting non-
discretionary legal duties “set agency policy by conclusively
interpreting the agencies’ legal obligations” such that
“agencies that solicit opinions in this category must necessarily
adopt the OLC’s interpretations of their legal obligations as
their own,” CfA Br. 39; and (2) “[w]hen an agency asks the
OLC to decide a legal issue relevant to an adjudication [of
private rights], it effectively cedes to the OLC the decision-
making authority over that aspect of the adjudication, and the
agency then applies the resulting legal decision,” CfA Reply
18. In our view, those are additional flavors of the “ex ante
adoption” theory. Even if agencies believe that they “must
necessarily” accept OLC’s legal interpretations, they do not
categorically adopt OLC opinions. To be sure, in individual
instances, an agency might “in practice” treat an OLC opinion
as “an authoritative Agency decision[] in the case[] to which [it
is] addressed,” and use it to “guide subsequent Agency
rulings.” Schlefer, 702 F.2d at 244. That would be an
5 CfA also contends, in passing, that OLC’s opinions resolving
interagency disputes are subject to disclosure because they are
“statements of policy and interpretations which have been adopted”
by OLC itself. See 5 U.S.C. § 552(a)(2)(B). Because that argument
was not properly raised, we will not address it. See Schneider v.
Kissinger, 412 F.3d 190, 200 n.1 (D.C. Cir. 2005) (“It is not enough
to mention a possible argument in the most skeletal way, leaving the
court to do counsel’s work, create the ossature for the argument, and
put flesh on its bones.”).

-- 28 of 40 --

29
“adoption” of the opinion in the relevant sense, but it would
occur only after the agency treated that individual opinion as
its working law.
3. Deliberative Process Privilege
Finally, the government asserts that the requested
documents are exempt from disclosure under the deliberative
process privilege embodied in FOIA Exemption 5. We need
not reach that issue. In CREW II, we considered an assertion
of the deliberative process privilege in the context of a FOIA
request for documents under the reading-room provision. We
declined to consider whether the privilege applied where the
plaintiff failed to meet the “antecedent requirement” of
plausibly alleging that FOIA mandated the disclosure of the
requested documents. 922 F.3d at 486, 487 (even if “FOIA
places the burden ‘on the agency to sustain its action,’” “the
OLC’s ultimate burden of proof does not alter the antecedent
requirement that [the plaintiff] plead a plausible
claim” (quoting 5 U.S.C. § 552(a)(4)(B))).
Thus, at the motion to dismiss stage, “regardless of the
OLC’s ultimate burden of proof, [the plaintiff] must first allege
factual matter supporting a plausible claim that the OLC
‘improperly’ withheld its formal written opinions.” CREW II,
922 F.3d at 487 (quoting Tax Analysts, 492 U.S. at 150). That
requires the plaintiff to “allege sufficient factual material about
the opinions that — if taken as true — would place them into
one of § 552(a)(2)’s enumerated categories” that require
disclosure. Id. at 488. “Then, and only then, would the OLC
bear its burden to justify withholding its formal written
opinions.” Id. Because we conclude that the reading-room
provision does not apply to the OLC opinions at issue here, and
that CfA’s claims therefore must be dismissed, we do not reach

-- 29 of 40 --

30
the government’s arguments regarding the deliberative process
privilege.
* * *
For the foregoing reasons, we reverse the district court’s
entry of summary judgment in favor of CfA with respect to the
disclosure of opinions that resolve interagency disputes and
remand for the district court to dismiss that claim. We affirm
the district court’s judgment insofar as it granted OLC’s motion
to dismiss CfA’s claims regarding the adjudication or
determination of private rights and the interpretation of non-
discretionary legal duties.
So ordered.

-- 30 of 40 --

RAO, Circuit Judge, concurring in the judgment: I write
separately to explain why Campaign for Accountability
(“CfA”) lacks standing to bring this suit. The Freedom of
Information Act creates a public right to disclosure of certain
records in an electronic reading room. CfA alleges that the
Office of Legal Counsel (“OLC”) is not following the law
because it failed to make its legal opinions available in a
reading room. To establish standing for an alleged reading
room violation, CfA was required to demonstrate a
particularized injury and some downstream, adverse impact
from the agency’s nondisclosure. See Prisology, Inc. v. Fed.
Bureau of Prisons, 852 F.3d 1114, 1116–18 (D.C. Cir. 2017);
TransUnion LLC v. Ramirez, 141 S. Ct. 2190, 2214 (2021).
CfA failed to demonstrate a particularized injury. CfA
merely raised a generalized grievance that OLC is violating the
law and only vaguely gestured at organizational injuries. We
lack jurisdiction to entertain CfA’s suit, which at bottom
presents an undifferentiated interest in enforcement of the laws
that is insufficient to invoke the Article III judicial power.
I.
The Office of Legal Counsel is a specialized component of
the Department of Justice that provides legal advice and
opinions to assist the President, the Attorney General, and
Executive Branch agencies in the execution of the laws. OLC
at times issues formal opinions of law, under authority
delegated from the Attorney General, but more frequently OLC
provides informal opinions and legal advice on constitutional,
statutory, and regulatory matters. See Citizens for Resp. &
Ethics in Wash. v. U.S. Dep’t of Just., 922 F.3d 480, 483–84
(D.C. Cir. 2019).
CfA is a nonprofit organization that “uses research,
litigation, and communications to expose misconduct and
malfeasance in public life.” J.A. 65. CfA requested OLC make

-- 31 of 40 --

2
certain types of legal opinions publicly available in an
electronic reading room. In its request for disclosure, CfA
relied exclusively on the Freedom of Information Act
(“FOIA”) reading room provision, which requires that agencies
“make available for public inspection in an electronic format”
enumerated categories of records. 5 U.S.C. § 552(a)(2) (the
“reading room provision”). CfA explicitly stated it was not
making a FOIA request for particular agency “records.” See 5
U.S.C. § 552(a)(3) (the “record request provision”). OLC
denied the reading room request, asserting that none of its
opinions were required to be disclosed under the reading room
provision.
CfA brought suit to compel OLC to release the opinions.
In its operative complaint, CfA claimed injury from OLC’s
refusal to provide the opinions in a reading room. CfA also
alleged the denial harmed CfA’s “core programmatic
activities.” The district court held that OLC must disclose
opinions that resolve interagency disputes, but not the other
categories of opinions sought by CfA.
II.
The majority holds that each of CfA’s reading room claims
fails on the merits. I find the majority’s careful analysis
persuasive, but I would decide this case at the jurisdictional
threshold because CfA lacks standing to bring these claims.
A.
Article III of the Constitution limits the jurisdiction of
federal courts to “Cases” and “Controversies.” U.S. Const. art.
III, § 2, cl. 1. Plaintiffs seeking to vindicate their rights in
federal court must demonstrate standing, establishing they have
suffered an injury in fact that is caused by the defendant and
redressable by the court. Lujan v. Defs. of Wildlife, 504 U.S.

-- 32 of 40 --

3
555, 560–61 (1992). This “irreducible constitutional
minimum” ensures that a proper party is seeking relief that may
be properly issued by a federal court. See id.; William
Baude & Samuel L. Bray, Proper Parties, Proper Relief, 137
Harv. L. Rev. 153, 154–56 (2023).
As particularly relevant here, the Supreme Court has long
held that the federal courts have no power to entertain suits
based only on a plaintiff’s general interest in enforcement of
the laws. When a plaintiff sues the government, standing
requirements ensure that federal courts do not “operate as an
open forum for citizens to press general complaints” about
violations of federal law. FDA v. All. for Hippocratic Med., 144
S. Ct. 1540, 1554 (2024) (cleaned up); see also Allen v. Wright,
468 U.S. 737, 754 (1984); Schlesinger v. Reservists Comm. to
Stop the War, 418 U.S. 208, 220–23 & n.13 (1974). The injury
in fact element “screens out plaintiffs who might have only a
general legal, moral, ideological, or policy objection to a
particular government action.” All. for Hippocratic Med., 144
S. Ct. at 1556.
Such general objections are not the business of the federal
courts, which may decide only cases and controversies.
Standing is a jurisdictional limit that marks the peculiar
province of the judicial power, separate from the legislative and
executive powers. The “generalized interest of all citizens” in
enforcement of the laws does not suffice to establish standing.
Schlesinger, 418 U.S. at 217. Courts may vindicate the private
rights of plaintiffs who are harmed by the government, but “it
would exceed Article III’s limitations if … in the absence of
any showing of concrete injury, we were to entertain citizen
suits to vindicate the public’s nonconcrete interest in the proper
administration of the laws.” Summers v. Earth Island Inst., 555
U.S. 488, 497 (2009) (cleaned up). Standing ensures that courts
stay within their proper constitutional sphere.

-- 33 of 40 --

4
Put another way, federal courts must stay within Article III
limits, lest they intrude on the powers of the political
departments. “Vindicating the public interest (including the
public interest in Government observance of the Constitution
and laws) is the function of Congress and the Chief Executive.”
Lujan, 504 U.S. at 576; see id. at 574–76 (discussing cases).
Courts cannot issue advisory opinions, and they must refrain
from “exercis[ing] general legal oversight of the Legislative
and Executive Branches.” TransUnion, 141 S. Ct. at 2203. The
constitutional responsibility for faithful execution of the laws
belongs to the President, not the courts. See U.S. Const. art. II,
§§ 1 and 3. Congress cannot “convert the undifferentiated
public interest in executive officers’ compliance with the law
into an individual right vindicable in the courts” because that
would grant courts “authority over the governmental acts of
another and co-equal department.” Lujan, 504 U.S. at 577
(cleaned up).
These bedrock standing principles have particular
relevance to plaintiffs bringing suit under FOIA’s reading room
provision, which imposes a general obligation on agencies to
disclose certain records in an electronic reading room. In
Prisology, we held that an agency’s failure to publish records
in a reading room is a “harm common to everyone, a harm of
the sort Lujan described as not stating an Article III case or
controversy.” Prisology, 852 F.3d at 1115–17. We described
how the standing inquiry differed between the reading room
provision and FOIA’s more familiar record request provision.
See id. at 1117 (recognizing that under the record request
provision our caselaw allows standing based on an agency’s
denial of a request for records). Because an alleged reading
room violation is a generalized grievance, a plaintiff seeking to
establish standing must demonstrate a “particularized injury”
and cannot merely point to the agency’s failure to disclose the
records. Id. at 1116–17.

-- 34 of 40 --

5
Our standing analysis in Prisology was further reinforced
by TransUnion, in which the Supreme Court recognized that
an “informational injury that causes no adverse effects cannot
satisfy Article III.” 141 S. Ct. at 2214 (cleaned up); see Grae v.
Corr. Corp. of Am., 57 F.4th 567, 570–71 (6th Cir. 2023)
(collecting cases recognizing this principle in the context of
public disclosure laws). A plaintiff cannot invoke the
jurisdiction of the federal courts unless he has identified some
“‘downstream consequences’ from failing to receive the
required information.” TransUnion, 141 S. Ct. at 2214
(quoting Trichell v. Midland Credit Mgmt., Inc., 964 F.3d 990,
1004 (11th Cir. 2020)).
Under TransUnion and Prisology, a plaintiff suing to
enforce FOIA’s reading room provision must establish
standing by demonstrating some downstream harm from an
agency’s failure to publish records in a public reading room.
B.
CfA alleges that OLC is in violation of FOIA’s reading
room provision and seeks a judicial order requiring OLC to
publish certain categories of its formal written opinions. But a
violation of the reading room provision alone is nothing more
than a generalized grievance, and CfA has not identified a
“downstream” or “adverse” effect from OLC’s alleged failure
to publish its opinions. TransUnion, 141 S. Ct. at 2214. CfA
therefore lacks standing.
1.
As an initial matter, CfA’s bare allegation of a reading
room violation does not suffice for standing. The reading room
provision states: “Each agency, in accordance with published
rules, shall make available for public inspection in an electronic
format” enumerated categories of records. 5 U.S.C. § 552(a)(2)

-- 35 of 40 --

6
(emphasis added). This provision imposes a requirement on
agencies to make the specified categories of records available
to the public. It does not create an individual right, only a
general public right to certain records. FOIA’s judicial review
provision states that district courts have jurisdiction “to order
the production of any agency records improperly withheld from
the complainant.” 5 U.S.C. § 552(a)(4)(B) (emphasis added).
Congress provided for judicial review when records were
improperly withheld from a particular complainant, not from
the public at large. Assuming the judicial review provision
applies to private enforcement of the reading room provision,
FOIA reflects the general principles of Article III standing
because any alleged injury from a reading room violation must
be particular to the plaintiff bringing suit.
CfA therefore cannot rely solely on OLC’s refusal to
disclose the opinions CfA requested under the reading room
provision because this denial does not amount to a
particularized injury. As explained above, CfA expressly
disclaimed reliance on FOIA’s record request provision, so any
statutory entitlement to reading room records invoked by CfA’s
request is “common to everyone,” and OLC’s refusal to
disclose is not, without more, a harm particular to CfA.
Prisology, 852 F.3d at 1116–17 (“Even if we inferred an injury
to Prisology from the Bureau’s alleged failure to publish its
records electronically, this would not differentiate Prisology
from the public at large.”). Without an adverse effect particular
to CfA, any injury from nondisclosure “is plainly
undifferentiated and common to all members of the public.”
United States v. Richardson, 418 U.S. 166, 176–77 (1974)
(cleaned up).
My colleagues conclude that CfA has standing primarily
because OLC denied CfA’s request that certain OLC opinions
be published in an electronic reading room. See Majority Op.

-- 36 of 40 --

7
15–16. That conclusion, however, squarely conflicts with
Prisology, which held that simply withholding records from a
reading room is not sufficient to establish standing.1 852 F.3d
at 1116–17. A generalized grievance about the “proper
application” of a public disclosure law is insufficient to invoke
the jurisdiction of the federal courts. Lujan, 504 U.S. at 573–
74. CfA cannot establish standing by claiming nothing more
than a generalized injury from OLC’s alleged refusal to publish
opinions pursuant to FOIA’s reading room provision.
2.
Nor has CfA demonstrated that OLC’s violation of the
reading room provision caused a downstream, adverse effect
particular to CfA. See TransUnion, 141 S. Ct. at 2214;
Prisology, 852 F.3d at 1117.
While the Supreme Court has not yet explained what is
needed to show an adverse effect, this court’s standing
precedents inform certain minimum requirements. First, the
plaintiff must put forth particularized facts; “general averments
1 By contrast, for suits under FOIA’s record request provision, the
Supreme Court has observed that the denial of a request for agency
records suffices as an injury in fact. See Pub. Citizen v. U.S. Dep’t of
Just., 491 U.S. 440, 449 (1989). TransUnion explicitly left in place
precedents concerning informational injuries under FOIA’s record
request provision. 141 S. Ct. at 2214; see also All. for Hippocratic
Med., 144 S. Ct. at 1564 (indicating there may be informational
injuries stemming from federal laws that require agencies “to
disseminate … information upon request by members of the
public”). Nonetheless, we have recognized that the reasoning of
TransUnion—that plaintiffs alleging informational injury must show
downstream, adverse effects—is in tension with cases allowing
standing for the mere denial of a FOIA record request. See Nat’l Sec.
Archive v. CIA, 104 F.4th 267, 272 & n.1 (D.C. Cir. 2024).

-- 37 of 40 --

8
and conclusory allegations” are not enough. Coubaly v. Cargill
Inc., 144 F.4th 343, 348 (D.C. Cir. 2025). Second, when
challenging the government’s alleged noncompliance with a
public disclosure law, a plaintiff must demonstrate concrete
and particularized harms that go beyond the general public
interest in ensuring that the government follows the law.2 See
All. for Hippocratic Med., 144 S. Ct. at 1556; TransUnion, 141
S. Ct. at 2207, 2214; id. at 2217, 2219–20 (Thomas, J.,
dissenting) (explaining that a legal violation is insufficient for
standing in the context of public rights because at the founding,
the judicial power did not extend to a plaintiff seeking to
enforce “a duty owed broadly to the community” absent a
“showing of actual damages” to the plaintiff).
Here, because CfA relies on organizational standing, it
must demonstrate “discrete programmatic concerns” that are
“being directly and adversely affected” by nondisclosure; a
“mere setback to the organization’s abstract social interests is
not enough.” Viasat, Inc. v. FCC, 47 F.4th 769, 781 (D.C. Cir.
2022) (cleaned up). As other circuits have observed in applying
TransUnion to public disclosure lawsuits, CfA must
2 As I have observed in a different context, a plaintiff suing under a
government disclosure law must show downstream harm that “has a
close relationship to a harm that has traditionally been regarded as
providing a basis for a lawsuit in English or American courts.”
Maloney v. Carnahan, 45 F.4th 215, 222–23 (D.C. Cir. 2022) (Rao,
J., dissenting from the denial of rehearing en banc). I do not further
elaborate on the requirements of a downstream harm in this context
because CfA fails to show downstream consequences even under a
minimal conception of what that requires. Furthermore, I note that
an injury in fact may be easier to establish when one private party is
harmed by another party’s unlawful action—a more traditional type
of case than a suit to vindicate public rights of disclosure against the
government. See TransUnion, 141 S. Ct. at 2214; id. at 2217, 2219–
20 (Thomas, J., dissenting).

-- 38 of 40 --

9
demonstrate “an interest in using the information beyond
bringing [its] lawsuit.”3 Grae, 57 F.4th at 570–71 (quoting
Harty v. W. Point Realty, Inc., 28 F.4th 435, 444 (2d Cir.
2022)); see also Campaign Legal Ctr. v. Scott, 49 F.4th 931,
938–39 (5th Cir. 2022).
Perhaps because the government never questioned CfA’s
alleged injury in fact, the summary judgment briefing is devoid
of any demonstration of standing. Looking back to the
complaint, CfA alleged it “uses research, litigation, and
communications to expose misconduct and malfeasance in
public life” and, here, OLC’s nondisclosure “has harmed, and
continues to harm, CfA in carrying out its core programmatic
activities.” J.A. 65, 81.
These allegations fall well short of stating downstream and
adverse effects particular to CfA. CfA merely describes various
activities in which it generally engages and asserts the legal
conclusion of “harm” to its “core programmatic activities.” It
neither identifies which “discrete programmatic concerns” are
at stake here, nor alleges how such concerns are “directly and
adversely affected” by OLC’s failure to provide the specified
opinions in a public reading room. Viasat, 47 F.4th at 781.
CfA’s amended complaint lacks any factual allegations as to
3 When considering standing in other contexts, this court has
similarly looked at downstream or adverse effects for informational
injuries. See Animal Legal Def. Fund, Inc. v. Vilsack, 111 F.4th 1219,
1229 (D.C. Cir. 2024) (finding concrete injury alleged by member of
plaintiff association who desired the withheld information to “make
an informed purchasing decision”); cf. Frank v. Autovest, LLC, 961
F.3d 1185, 1188 (D.C. Cir. 2020) (plaintiff failed to show
“detrimental reliance” on misrepresentations that allegedly violated
the Federal Debt Collection Practices Act).

-- 39 of 40 --

10
how it has suffered downstream, adverse effects from OLC’s
nondisclosure.4
These requirements for standing should not present an
insurmountable hurdle to plaintiffs that have suffered a
concrete and particularized injury from an agency’s alleged
failure to comply with statutory requirements. But maintaining
a case in federal court requires more than CfA’s general
statements of its organizational activities and conclusory
allegations of harm to those activities.
* * *
The Article III courts are not a forum for opining on “legal
issues in response to citizens who might roam the country in
search of governmental wrongdoing.” All. for Hippocratic
Med., 144 S. Ct. at 1555 (cleaned up). Because CfA has failed
to demonstrate standing, its suit must be dismissed for lack of
jurisdiction. I therefore concur in the judgment dismissing
CfA’s claims.
4 The majority relies on a district court case decided prior to
Prisology to show that the court entertained reading room claims on
allegations of harm similar to CfA’s. See Majority Op. 16. But as we
explained in Prisology, that case did “not discuss[] standing.” 852
F.3d at 1117. This drive-by jurisdictional assumption cannot support
CfA’s standing.

-- 40 of 40 --