The Office of Legal Counsel (OLC) in the Department of Justice is the authoritative legal counsel for the executive branch, and its binding, formal opinions form a precedential body of law guiding executive actions in parallel to judicial doctrine.^1×1. See Daphna Renan, The Law Presidents Make, 103 Va. L. Rev. 805, 815 (2017). OLC may, when requested, also dispense advice informally in a method akin to “talking shop,” id. at 835, that is nonetheless still binding by custom, id. at 847 n.177. OLC serves a critical role by providing carefully reasoned legal guidance to executive actors while insulating debate over some of the President’s most important decisions from Congress, the judiciary, and the public.^2×2. For example, OLC advised President Kennedy on his ability to institute a blockade in the Bay of Pigs. See Authority of the President to Blockade Cuba, 1 Op. O.L.C. Supp. 195 (2013) (originally issued Jan. 25, 1961). This insularity is exacerbated by the fact that many of the questions OLC addresses are never litigated. See Memorandum from David J. Barron, Acting Assistant Att’y Gen., Office of Legal Counsel, to Att’ys of the Office of Legal Counsel, Re: Best Practices for OLC Legal Advice and Written Opinions 1 (July 16, 2010), https://www.justice.gov/sites/default/files/olc/legacy/2010/08/26/olc-legal-advice-opinions.pdf [https://perma.cc/G8Q3-5BCD] [hereinafter Best Practices Memo]. Even if a court disagrees with an OLC opinion, the Department of Justice often refuses to prosecute criminal acts committed in reliance on OLC opinions, permitting OLC to dispense “get-out-of-jail-free cards.” Jack Goldsmith, The Terror Presidency 97 (2007). Because it sits within the executive branch, OLC favors a broader vision of presidential power than Congress, the judiciary, or legal academia.^3×3. See Goldsmith, supra note 2, at 37. Publication of OLC’s formal opinions thus plays a crucial role in preserving the separation of powers and promoting transparency and accountability in executive decisionmaking.^4×4. See Cornelia T.L. Pillard, The Unfulfilled Promise of the Constitution in Executive Hands, 103 Mich. L. Rev. 676, 677, 749–50 (2005). Additionally, the prospect of disclosure provides attorneys within OLC a key incentive to maintain integrity in the face of pressure to please executive actors seeking legal approval for policy decisions. See Trevor W. Morrison, Stare Decisis in the Office of Legal Counsel, 110 Colum. L. Rev. 1448, 1470 (2010). Recently, in Citizens for Responsibility & Ethics in Washington v. U.S. Department of Justice^5×5. 922 F.3d 480 (D.C. Cir. 2019). (CREW II), the D.C. Circuit dismissed Citizens for Responsibility and Ethics in Washington’s (CREW) claim that the “reading room” provision of the Freedom of Information Act^6×6. 5 U.S.C. § 552(a)(2) (2018). (FOIA) required OLC to publish its opinions because CREW had not identified which opinions had been adopted by other agencies.^7×7. See CREW II, 922 F.3d at 486–87. The D.C. Circuit’s holding reinforces a trend against disclosure of OLC opinions under FOIA and increases the risk that OLC “secret law” will grow entrenched without congressional intervention.

FOIA’s recognition of the need for transparency in the types of executive decisions guided by OLC has been widely affirmed. Congress passed FOIA to “prevent the creation of ‘secret law,’”^8×8. NLRB v. Sears, Roebuck & Co., 421 U.S. 132, 138 (1975). that is, unpublished law made within the executive branch to guide policy decisions that often have profound public impact.^9×9. See Dakota S. Rudesill, Coming to Terms with Secret Law, 7 Harv. Nat’l Sec. J. 241, 283 (2015). FOIA’s reading room provision obliges agencies to proactively publish certain categories of documents, including “working law,” or documents that carry the force of law and guide agency policy.^10×10. CREW II, 922 F.3d at 486 (quoting Sears, 421 U.S. at 153). Working law consists of the final opinions reached in the adjudication of cases as well as statements regarding the policies and interpretations that have been adopted by an agency.^11×11. Id. Several categories of documents are exempted, including records that are part of an agency’s deliberative process.^12×12. See 5 U.S.C. § 552(b) (2018). The deliberative process privilege exempts papers “reflect[ing] the agency’s group thinking in . . . determining what its law shall be.” Sears, 421 U.S. at 153 (citation omitted). FOIA’s more commonly employed reactive provision permits interested parties to request specific documents from an agency.^13×13. 5 U.S.C. § 552(a)(3); see also CREW II, 922 F.3d at 434. On a bipartisan basis, the Supreme Court and past Presidents have endorsed FOIA’s goal of transparency: the Court has established a “strong presumption in favor of disclosure” of agency documents,^14×14. U.S. Dep’t of State v. Ray, 502 U.S. 164, 173 (1991). and both President Obama and President George W. Bush directed federal agencies to increase transparency through FOIA.^15×15. See Improving Agency Disclosure of Information, Exec. Order No. 13,392, 70 Fed. Reg. 75,373 (Dec. 14, 2005); Memorandum of January 21, 2009, 74 Fed. Reg. 4,683 (Jan. 26, 2009). OLC opinions guide agency decisions by providing binding legal advice to executive actors who seek OLC’s assistance in addressing complex legal issues or settling legal questions disputed by executive agencies, thus implicating secret law concerns.^16×16. See Best Practices Memo, supra note 2, at 1–2.

In 2013, CREW^17×17. CREW is a nonprofit advocating for government transparency, accountability, and ethics. See About Us, CREW, https://www.citizensforethics.org/who-we-are [https://perma.cc/P3E7-WHPP]. unsuccessfully petitioned OLC for disclosure of all of its formal opinions under FOIA’s reading room provision.^18×18. CREW v. U.S. Dep’t of Justice, 164 F. Supp. 3d 145, 148 (D.D.C. 2016). After its request was denied, CREW sued OLC in Citizens for Responsibility & Ethics in Washington v. U.S. Department of Justice^19×19. 846 F.3d 1235 (D.C. Cir. 2017). (CREW I), alleging that the denial was arbitrary and capricious under the Administrative Procedure Act^20×20. 5 U.S.C. §§ 551, 553–559, 701–706 (2018); id § 706(2)(A). CREW initially also brought a claim under the reading room provision of FOIA but subsequently dropped this element of its complaint. See CREW, 164 F. Supp. 3d at 147–49. (APA). CREW sought to enjoin OLC to publish all existing and future formal opinions and their accompanying indices.^21×21. CREW I, 846 F.3d at 1240. Writing for the district court in CREW I, Judge Mehta dismissed the suit because it was improperly brought under the APA instead of FOIA.^22×22. See CREW, 164 F. Supp. 3d at 147. Remedy under the APA is available only as a last resort,^23×23. See id. at 151. and Judge Mehta believed that FOIA’s reading room provision afforded CREW an adequate remedy.^24×24. See id. at 155–56. Judge Tatel affirmed for the D.C. Circuit,^25×25. CREW I, 846 F.3d at 1235. Judge Tatel was joined by Judge Sentelle and Judge Wilkins. noting that the court’s determination that the suit should be brought under FOIA was “entirely distinct” from finding that CREW was entitled to relief.^26×26. Id. at 1246. Judge Tatel further stated that FOIA did not entitle CREW to force OLC to publish its opinions because judicial review was limited to documents withheld from specific plaintiffs, not the public. Id. at 1243–44. CREW II did not address this issue. However, in Animal Legal Defense Fund v. U.S. Department of Agriculture, 935 F.3d 858 (9th Cir. 2019), the Ninth Circuit explicitly rejected the D.C. Circuit’s reading of the remedy available under the reading provision as a “red herring” that “renders the reading room provision into precatory language.” Id. at 875. Following the court’s suggestion in CREW I, CREW sued again to enjoin publication under FOIA’s reading room provision after OLC once again denied CREW’s request to disclose all of its unpublished written opinions and accompanying indices.^27×27. See CREW v. U.S. Dep’t of Justice, 298 F. Supp. 3d 151, 153 (D.D.C. 2018).

Writing for the district court in CREW II, Judge McFadden dismissed the suit.^28×28. Id. Judge McFadden found CREW’s argument that the binding nature of OLC opinions rendered them the working law of the requesting agencies to be factually insufficient based on Electronic Frontier Foundation v. U.S. Department of Justice^29×29. 739 F.3d 1 (D.C. Cir. 2014); see CREW, 298 F. Supp. 3d at 154–55. (EFF). In EFF, the D.C. Circuit held that an OLC opinion concerning the FBI’s practice of obtaining phone records without a warrant was merely legal advice and thus exempt from the reading room provision by the deliberative process privilege.^30×30. See EFF, 739 F.3d at 9–10. Judge McFadden relied on EFF to establish that an OLC opinion could not be the working law of the requesting agency unless the agency actively adopted the opinion because OLC had no authority to set an agency’s policy.^31×31. CREW, 298 F. Supp. 3d at 155. Judge McFadden further noted that the attorney-client privilege would also exempt some OLC opinions from the reading room provision.^32×32. See id. Consequently, he found that CREW’s plea for publication of all OLC opinions was overbroad and denied CREW’s request for discovery to determine the extent of OLC’s disclosure obligations.^33×33. See id. at 156. In dismissing the suit under Federal Rule of Civil Procedure 12(b)(6) for failing to make sufficient factual allegations to state a plausible claim,^34×34. See id. Judge McFadden gave CREW leave to file an amended complaint seeking disclosure of an eligible subset of written opinions.^35×35. See id. A similar case seeking disclosure of OLC’s formal written opinions, Campaign for Accountability v. U.S. Department of Justice, 278 F. Supp. 3d 303 (D.D.C. 2017), was also dismissed with leave to amend. See id. at 325. Campaign for Accountability did choose to amend its complaint, and the case was still pending as of the time of this case comment. See CREW II, 922 F.3d at 486. CREW declined to amend its complaint and directly appealed.^36×36. See CREW II, 922 F.3d at 485.

Judge Henderson affirmed for the D.C. Circuit,^37×37. Judge Henderson was joined by Judge Ginsburg. CREW’s subsequent petition for en banc review was denied. Citizens for Responsibility and Ethics in Wash. v. U.S. Dep’t of Justice, No. 18-5116 (D.C. Cir. Aug. 14, 2019) (per curiam). similarly relying on EFF for the proposition that the binding nature of formal opinions was insufficient to establish that they were working law.^38×38. See CREW II, 922 F.3d at 486–87. EFF had emphasized that OLC’s power to draw the legal boundaries of an agency’s decision did not enable it to adopt a particular policy on the agency’s behalf.^39×39. See id. at 486 (citing Elec. Frontier Found. v. U.S. Dep’t of Justice, 739 F.3d 1, 9 (D.C. Cir. 2014)). Because CREW had not alleged any additional facts beyond the binding nature of formal opinions, the court found that CREW had failed to state a plausible claim of relief under Rule 12(b)(6).^40×40. See id. at 487. The court further noted that requiring plaintiffs to identify a subset of opinions that was not exempt from the reading room provision was not overly burdensome; rather, it aligned with the reactive provision’s requirement that parties describe the documents requested.^41×41. See id. at 489.

Judge Pillard^42×42. Judge Pillard served as Deputy Assistant Attorney General in OLC under President Clinton. Cornelia T.L. Pillard, U.S. Ct. of Appeals D.C. Cir., https://www.cadc.uscourts.gov/internet/home.nsf/Content/VL+-+Judges+-+NP [https://perma.cc/EDY6-U5T6]. dissented from the court’s dismissal because she believed that CREW had plausibly alleged that a portion of OLC opinions constituted the working law of requesting agencies and was subject to mandatory publication under the reading room provision.^43×43. See CREW II, 922 F.3d at 490 (Pillard, J., dissenting). She found that CREW had plausibly established that some OLC opinions had been adopted by requesting agencies because they were binding.^44×44. See id. at 491 (citing 5 U.S.C. § 552(a)(2)(B) (2018)). Judge Pillard also believed that CREW had plausibly alleged that some opinions were final opinions reached in adjudications because OLC decided interagency disputes through an adversarial system.^45×45. See id. at 490–91 (citing 5 U.S.C. § 552(a)(2)(A)). Judge Pillard characterized the majority’s reliance on EFF as making “too much soup from one oyster” by dismissing the complaint after identifying a single opinion exempt from the reading room provision.^46×46. Id. at 492. Finally, she expressed concern that requiring plaintiffs to plead for publication of an eligible subset of opinions “effectively forces [plaintiffs] to anticipate and plead around any FOIA-exemption defense the government might raise.”^47×47. Id. Thus, she would have allowed the case to survive the motion to dismiss and proceed to discovery.^48×48. See id. at 493–94.

CREW II deals a strong blow to efforts to ensure transparency and accountability in executive decisionmaking by reinforcing a trend away from disclosure under FOIA’s reading room provision. OLC proactively publishes “only a fraction” of its opinions,^49×49. Morrison, supra note 4, at 1477. and parties seeking to access unpublished opinions must navigate the long and arduous path of filing individual requests under FOIA’s reactive provision. FOIA’s reading room provision could have provided a critical pathway toward greater publication of OLC opinions. However, by requiring plaintiffs to identify which types of OLC opinions are publishable without the benefit of discovery, the D.C. Circuit builds on EFF to reduce the likelihood that plaintiffs will be able to survive the pleading stage and achieve publication of OLC opinions. This reduces transparency in the immediate term and also incentivizes the use of formal opinions by providing the executive branch greater assurance that formal opinions will not be involuntarily disclosed. Due to the precedential nature of OLC opinions, CREW II increases the risk that OLC secret law will strengthen unless Congress imposes additional disclosure requirements on OLC.

FOIA’s application to OLC has been mostly limited to the reactive provision, which imposes significant barriers to disclosure. OLC generally views its opinions as privileged and exempt from the reading room provision’s mandatory publication requirement.^50×50. See OLC FOIA Electronic Reading Room, U.S. Dep’t Just., https://www.justice.gov/olc/olc-foia-electronic-reading-room [https://perma.cc/746N-DGTL]. Although it voluntarily publishes “a fraction of all of its written opinions,” many of these opinions are not published for years.^51×51. Morrison, supra note 4, at 1477. Because OLC does not publish a comprehensive list of its opinions, it is “difficult to know what fraction of all OLC opinions are publicly available,” and the published opinions are likely “not a representative sample” because opinions discussing politically controversial or highly classified topics are unlikely to be published. See id. at 1478. This slow, limited publication often forces parties seeking access to OLC’s records to submit specific document requests under FOIA’s reactive provision, where they face a difficult path. Making the request is the first hurdle; parties are required to specifically describe the documents sought although the titles and subjects of OLC opinions may not be public.^52×52. See CREW II, 922 F.3d at 487–88. Even when granted, FOIA requests for OLC work product face potentially lengthy delays, and opinions may not be made publicly available beyond disclosure to the requesting party.^53×53. See, e.g., Harold Hongju Koh, Protecting the Office of Legal Counsel from Itself, 15 Cardozo L. Rev. 513, 517–18 (1993). Complex requests for OLC work products face an average delay of six months, and at least one request has been pending since 2012.^54×54. See FOIA: Create a Basic Report, Off. Info. Pol’y, U.S. Dep’t Just., https://www.foia.gov/data.html?foia/Services/RequestDisposition.jsp?agencyName=DOJ&requestYear=2017 [https://perma.cc/W8R3-ZQUP] (inquiry conducted by selecting “Processing Time” and “Complex Requests,” “Department of Justice,” and FY 2018). This is three months shorter than the longest-pending FOIA request in the Department of Justice, set by the FBI. See id. (inquiry conducted by selecting “Processing Time” and “Ten Oldest Requests,” ”Department of Justice,” and FY 2018). The complexity of a FOIA request is determined by the extent of the request and the anti-cipated number of “steps to process” the request. See FOIA, How Do I Make a FOIA Request, https://www.foia.gov/how-to.html [https://perma.cc/6QRF-6NDR]. Consequently, the legal analysis behind some of the executive branch’s most crucial decisions may stay secret for long periods, which, at a minimum, temporarily shields executive decisionmakers from having to fully justify their actions.^55×55. Cf. Jennifer Shkabatur, Transparency With(out) Accountability: Open Government in the United States, 31 Yale L. & Pol’y Rev. 79, 89–90 (2012) (arguing that delaying processing FOIA requests limits agency accountability and that requiring specific document descriptions imposes knowledge barriers). This delay also hides precedential opinions that continue to influence present executive decisionmaking.^56×56. For example, OLC opined on whether a President could pardon himself in the final week of President Nixon’s presidency. See Presidential or Legislative Pardon of the President, 1 Op. O.L.C. Supp. 370 (Aug. 5, 1974). This question arose again during the impeachment of President Clinton, but the OLC opinion was not published until 2013. See Can President Clinton Pardon Himself?, Slate (Dec. 30, 1998, 6:46 PM), https://perma.cc/A2S8-VG4H]. This is the topic of litigation in Francis v. U.S. Department of Justice, No. 19-cv-01317 (W.D. Wash. filed Aug. 21, 2019), a suit brought under FOIA by the Knight Institute to obtain OLC opinions more than twenty-five years old. Complaint at 1–2, Francis, No. 19-cv-01317 (W.D. Wash. filed Aug. 21, 2019). In light of the challenges to obtaining OLC opinions under the reactive provision, the reading room provision presented as a strong candidate for ensuring transparency and accountability within the executive branch before EFF and CREW II.

Following CREW II, however, plaintiffs seeking to require OLC to proactively publish its opinions under the reading room provision face a significant hurdle. EFF had previously curtailed the reading room provision’s applicability to OLC opinions by holding that OLC did not have the authority to establish the working law of the agencies receiving its opinions.^57×57. See 739 F.3d 1, 9–10 (D.C. Cir. 2014). As Judge Pillard noted, CREW II further limited the reading room provision by requiring plaintiffs to plead for the publication of specific subsets of opinions that agencies have adopted as their working law without the benefit of discovery.^58×58. See CREW II, 922 F.3d at 492 (Pillard, J., dissenting). This requires plaintiffs litigating under the reading room provision to perform the same gymnastic exercise as parties requesting records under the reactive provision by forcing plaintiffs to navigate a vast information disparity and specifically describe the documents they are seeking to have published.^59×59. See id. at 487–88 (majority opinion). This pleading standard disadvantages plaintiffs; as empirical studies following Twombly and Iqbal have demonstrated, raising the pleading burden increases the likelihood that lawsuits will fail at the motion to dismiss stage for failure to plead sufficient facts.^60×60. See, e.g., Scott Dodson, A New Look: Dismissal Rates of Federal Civil Claims, 96 Judicature 127, 132–33 (2012). The negative effects of CREW II have already been partially realized in Campaign for Accountability v. U.S. Department of Justice,^61×61. 280 F. Supp. 3d 112 (D.D.C. 2017). where the plaintiff was pressed to amend which subsets of opinions it alleged were subject to the reading room provision in order to escape the Department of Justice’s motion to dismiss.^62×62. Plaintiff’s Opposition to the Government’s Renewed Motion to Dismiss at 1, Campaign for Accountability v. U.S. Dep’t of Justice, No. 16-cv-01068 (D.D.C. Mar. 6, 2018). Campaign for Accountability sought the publication of opinions issued to independent agencies. It cited OLC’s requirement that independent agencies consent to being bound by OLC opinions for the proposition that independent agencies had affirmatively adopted OLC opinions as their working law. See id. at 30 n.11. Without the benefit of discovery, the plaintiff was unaware that this was the same consent process OLC used for all agencies, and it dropped this claim to survive the government’s motion to dismiss. See id. The heightened pleading burden imposed by CREW II thus adds to existing barriers to obtaining OLC opinions through FOIA.

In addition to impairing efforts to ensure transparency of past opinions, CREW II may incentivize greater future use of formal opinions.^63×63. See Renan, supra note 1, at 855. Formal opinions may be favored for their thoroughness, precedential weight, credibility-bestowing power,^64×64. See id. at 820, 854. However, OLC’s credibility-bestowing power has waned. See id. at 866–67. Its memos authorizing torture after 9/11 were widely condemned and ultimately revoked. See id. In 2019, its memo disagreeing with the Inspector General of the Intelligence Community’s decision that a whistleblower report concerning President Trump’s communications with Ukraine needed to be sent to Congress was publicly condemned by more than sixty federal Inspectors General. See Letter from Council of the Inspectors Gen. on Integrity & Efficiency to Steven A. Engel, Assistant Att’y Gen. (Oct. 22, 2019), https://www.ignet.gov/sites/default/files/files/CIGIE_Letter_to_OLC_Whistleblower_Disclosure.pdf [https://perma.cc/QXR9-3JSR]. and ability to grant executive actors near immunity if their decisions are challenged in court.^65×65. Goldsmith, supra note 2, at 97. However, the risk that these opinions will be involuntarily disclosed may encourage executive actors to rely more on informal opinions from OLC^66×66. Such advice is often delivered through oral presentations or e-mails. Best Practices Memo, supra note 2, at 2. or the legal apparatuses within agencies and the White House.^67×67. See Goldsmith, supra note 2, at 91; Renan, supra note 1, at 828. CREW II reduces the risk of involuntary disclosure through FOIA, and thus, could lead to increased usage of formal opinions. Increased reliance on formal opinions would raise the likelihood of entrenchment of secret law within the executive branch because formal opinions have precedential weight and are given greater authority in court.^68×68. See Goldsmith, supra note 2, at 97. Movement toward compelled disclosure through FOIA may not necessarily result in increased transparency of future legal decisions within the executive branch if it disincentivizes the use of formal opinions. But it would have the important effects of limiting the precedential role of OLC’s future guidance and increasing transparency of the existing body of opinions.

CREW II reinforces a trend against the disclosure of OLC opinions, preventing Congress and the public from accessing the binding legal advice underpinning many of the executive branch’s actions. Given the D.C. Circuit’s reluctance to facilitate the publication of OLC opinions through the reading room provision, the responsibility of promoting transparency and accountability within OLC may fall to Congress.^69×69. Cf. Jameel Jaffer & Brett Max Kaufman, A Resurgence of Secret Law, 126 Yale L.J.F. 242 (2016), https://www.yalelawjournal.org/pdf/JafferKaufmanforPosting_f5dj1p2t.pdf [https://perma.cc/JY4X-GNB4] (urging Congress to reaffirm the working law doctrine in light of courts’ limitation of disclosure obligations under the working law doctrine). The D.C. Circuit’s decisions in EFF and CREW II fortify the insular nature of OLC and contrast with congressional intent to establish broad disclosure of agency law through FOIA^70×70. See U.S. Dep’t of State v. Ray, 502 U.S. 164, 173 (1991) (noting FOIA’s “strong presumption in favor of disclosure”). as well as recent legislative efforts to ensure greater transparency within OLC.^71×71. See SUNLIGHT Act of 2019, H.R. 4556, 116th Cong. (2019) (proposing mandating OLC to publish its opinions). Senator Feingold previously proposed the OLC Reporting Act of 2008, S. 3501, 110th Cong. (2008), which would have required the Attorney General to inform Congress of any OLC opinions finding federal statutes unconstitutional, inapplicable, or without effect. OLC found that the OLC Reporting Act would be unconstitutional. See Constitutionality of the OLC Reporting Act of 2008, 32 Op. O.L.C. 14 (2008). Especially as OLC opinions continue to be a flashpoint between Congress and the President,^72×72. See, e.g., Testimonial Immunity Before Congress of the Former Counsel to the President, O.L.C. slip op. at 1 (May 20, 2019), https://www.justice.gov/sites/default/files/opinions/attachments/2019/11/04/2019-05-20-test-immun-fmr-whc-2_1.pdf [https://perma.cc/MS87-G3AY] (finding that White House Counsel Don McGahn was not required to testify before Congress). it may be prudent for Congress to at least require OLC to disclose its formal opinions to Congress. In the absence of such reform, CREW II strengthens the specter of secret law, pressing against the constitutional limits of executive power and shielding some of the Executive’s most important legal decisions from Congress and the public.="https:>="https:>