We brought our lawsuit on behalf of five clients, who you can read about here. And over the past several years, we’ve spoken with dozens of former government employees who have told us similar stories about the state of the prepublication review system. But we have also learned a great deal about the failings of that system through the government’s own documents.

The lawsuit is based in part on thousands of documents that the Knight Institute and the ACLU obtained in response to multi-year Freedom of Information Act litigation. We are publishing many of these documents today, and we’ll be publishing more in the coming months.

These documents illustrate many of the fundamental problems with the current prepublication review system.

First, the government’s documents show that the prepublication review system is a tangled mess. We already knew that agencies imposed prepublication review obligations through a patchwork of internal policies, regulations, and contracts. But the documents show us just how fractured and complicated the system really is.

When we filed our first FOIA request in March 2016, many of the documents that explain agencies’ prepublication review policies and processes were publicly available—but surprisingly, many weren’t. After we sued eighteen agencies under FOIA, we were able to uncover several of those hidden documents.

For example, the CIA produced an agency regulation titled “Agency Prepublication Review of Certain Material Prepared for Public Dissemination” that “sets forth CIA policies and procedures for the submission and review of material proposed for publication or public dissemination by current and former employees and contractors and other individuals obligated by the CIA secrecy agreement.” (Previously, the only publicly available version of this document was heavily redacted and more than eleven years old.)

Thanks to the documents we’ve shaken loose we’ve been able to fill in major holes concerning the agencies’ prepublication review regimes. For example, we now understand that the CIA’s regime includes the now-released regulation and at least three nondisclosure agreements—including two that appear to be in use across the intelligence community, and one that all CIA officers must sign as a condition of employment.

The cache of documents also makes clear that the prepublication review system is a bureaucratic morass. Submission and review standards, review timelines, and appeals processes vary widely across the agencies. What’s more, many authors must submit the same manuscript to more than one agency, and agencies conducting prepublication review often refer submissions to other agencies which have their own so-called “equities” in deciding whether to censor the material. As a general matter, this referrals process is almost entirely opaque to authors. Often, authors are not told to which agencies their manuscripts have been referred, and even if they are told, they are most likely not informed what standards these agencies will apply.

Second, the documents we obtained show that authors frequently have difficulty determining what must be submitted, because submission requirements are vague and sweep in a vast range of innocuous material.

An email exchange produced by the FBI is illustrative. In the exchange, a former FBI employee contacted the agency’s prepublication review office about a novel he had written about criminals who rob armored trucks. The story, he said, was “purely fictional.” It was set in 1988, before he had even entered the Bureau, and it took place in Atlanta, a city to which he had never been assigned to work. All of the characters’ names were made up. Though he had investigated armed robberies while working at the FBI twenty-three years prior, he did not include any information he acquired from the FBI in his story. In his view, the draft manuscript didn’t “fit[] within the purview of the pre-publication guidelines,” which require the submission of “knowledge gained through FBI employment or assignments related to the FBI.” But the FBI reviewers disagreed, asserting that the manuscript could still disclose “sensitive techniques or methods.” They told the author to submit.

Remarkably, the documents also show that the CIA at least has deliberately kept submission requirements shrouded. A document produced by the agency states that, as a matter of policy, the CIA’s Publications Review Board “will not provide a copy of a secrecy agreement or nondisclosure agreement to an author who requests one they signed,” even though these agreements “are typically not classified.”

Third, the government’s documents reveal that the system is overwhelmed. The amount of material being submitted for prepublication review has increased dramatically in recent decades. For example, when the CIA’s PRB formally came into existence in the 1970s, it reviewed approximately 1,000 pages per year. In 2014, the board reviewed over 150,000, averaging a rate of 400 per day.

As the number of submissions has increased, so too have the frequency and severity of delays. For example, in 1982, the average PRB review took just 13 days. Today, review takes much longer. According to a draft CIA Inspector General report provided to us, “with today’s volumes, complexity, and strive for immediacy, PRB is struggling with achieving timeliness, and to some extent thoroughness/quality.” The report also states that “book-length manuscripts received today are currently projected to take over a year because of the complexity and large book backlog.”

Charts produced to the ACLU and Knight Institute in FOIA litigation confirm that other agencies have seen similar increases.

Fourth, the documents confirm that favored officials are sometimes afforded special treatment, with their manuscripts fast-tracked or seemingly subjected to less scrutiny. For example, while it took NCIS-veteran Mark Fallon nearly eight months to get his book reviewed, it took former FBI Director James Comey and Secretary of State Hillary Clinton just seven weeks and eight weeks, respectively. In Clinton’s case, agency censors were explicitly warned: “The [review] office is under great pressure to turn this around quickly. If you are tardy in your response, you may get a high-level Department official call.”

We have found no official guidance setting forth factors to determine when an author should receive special treatment. It appears to be personal and arbitrary.

Fifth, the documents demonstrate that administrative appeals processes fail to afford authors effective recourse for erroneous censorship decisions. Delays are one problem. For example, an administrative appeal lodged in 2014 by former intelligence officer Michael Richter appears to have taken over seven months to adjudicate. This, despite the fact that the appeal “require[d] analysis of one sentence, in one paragraph, citing one document”—and the author’s insistence that he had “only learned of [the excised] information by virtue of having access to the New York Times website.”

Another problem is lack of transparency. Because authors who receive adverse prepublication review determinations rarely get an explanation—or at least a meaningful one—they are unable to effectively address censors’ concerns.

To make matters worse, authors are unable to directly appeal redactions mandated by other agencies to those agencies. Take Richter’s case, for example. As a former DIA and ODNI officer, he submitted his manuscript to the Defense Department. The manuscript was subsequently referred to the ODNI and the CIA, and it was the CIA that insisted on the redaction. However, Richter was unable to directly appeal the CIA’s redaction to the CIA. Instead, he was required to lodge an appeal with the Defense Department, which was permitted—though, it seems, not obliged—to ask the CIA to reconsider its decision.

Individually, each of these revelations may offer only a relatively small insight into the pathologies of a sprawling system, but, collectively, they are striking. For years, prominent voices from across a broad spectrum of political ideologies and government work experiences have observed that the prepublication review system is broken. Now, the government’s own documents confirm it.

“Recent events strongly suggest that the Trump administration is abusing the prepublication review process to keep the public from knowing critical information about the president and his administration in the lead up to the 2020 elections,” said Ramya Krishnan, Staff Attorney at the Knight First Amendment Institute at Columbia University. “If the administration is using prepublication review as a tool to favor loyalists and to silence its critics, the public has a right to know before the elections.”

Today’s lawsuit seeks records pertaining to the government’s prepublication review of former National Security Advisor John Bolton’s memoir as well as several other works by former senior Trump administration officials, including former acting FBI director Andrew McCabe, former acting Attorney General Matthew Whitaker, and former National Security Advisor H.R. McMaster, as well as former Obama administration officials, including former CIA Directors Michael Hayden and John Brennan, both forceful critics of the Trump administration. The plaintiffs submitted a FOIA request on January 27, 2020, and to date have not received responsive records.

“If this is how the administration is treating high-profile former officials in the spotlight, imagine how they're treating the millions of others who don't have the resources to fight back," said Brett Max Kaufman, Senior Staff Attorney at the American Civil Liberties Union.

According to today’s complaint, “The requested records will help the public to evaluate whether the Administration has strategically deployed prepublication review to amplify favorable speech and suppress critical speech. It is crucial that the electorate can make such a determination prior to the November 2020 election so it can assess whether the information cleared by the Administration is accurate or whether it reflects an improperly censored account of the Administration’s activities.”

“Former public officials are uniquely positioned to share insights into government activities,” said Meenakshi Krishnan, Legal Fellow at the Knight First Amendment Institute. “Any attempt to use prepublication review to suppress criticism in advance of an election raises significant First Amendment concerns.”

The groups are suing the Department of State, Department of Justice, Central Intelligence Agency, National Security Agency, Federal Bureau of Investigation, Office of the Director of National Intelligence, Department of Air Force, and Department of the Army.

A copy of today’s complaint can be found here.

On April 2, 2019, the Knight Institute and ACLU filed a lawsuit on behalf of five former public servants challenging the government’s system of prepublication review. The plaintiffs argue that the system violates the First and Fifth Amendments. On April 16, 2020, a district court granted the government’s motion to dismiss the case, holding that the plaintiffs had standing to pursue the challenge but that the prepublication review system is constitutional. The plaintiffs have appealed. More information about this lawsuit, Edgar v. Ratcliffe, is available here.

An overview of the key features of the prepublication review regimes across 17 intelligence agencies can be found here.

For more information, contact Lorraine Kenny, lorraine.kenny@knightcolumbia.org.

Over the last several months, the Knight Institute along with the ACLU have filed and litigated FOIA requests for records relating to “prepublication review”—the byzantine regulatory regime that requires millions of former public servants to submit their manuscripts to government censors before publication. We know that Comey’s book was submitted for prepublication review, and apparently “very little” was removed. However, we do not know how long the review took, what alterations the FBI (or any other agency) demanded that he make, and how his experience with prepublication review compares to the experience of the thousands of other former employees and contractors who submit manuscripts to government agencies each year. 

It is a remarkable and under-examined fact that the U.S. intelligence agencies require millions of former employees, contractors, and even interns to submit works for review prior to publication. It is a sprawling regime of prior restraints—and, as Jack Goldsmith and Oona Hathaway have observed, it is a system “racked with pathologies.” There is no executive-branch-wide policy on the review process, meaning that each agency has its own. Agency schemes comprise a tangle of regulations, policies, and non-disclosure agreements the requirements of which are often obscure and sometimes inconsistent. Many agencies impose prepublication-review obligations even on former employees who have never had access to sensitive information. The criteria for submission and review are vague and overbroad.  Censors’ decisions are arbitrary. And authors who receive an adverse determination lack effective recourse. 

Perhaps the complaint most often heard about the prepublication-review system is that review simply takes too long. Many authors have waited months or even years for agencies to review their manuscripts. For example, Nada Bakos, a former CIA analyst, submitted a manuscript for review back in 2015. When she finally received a response nearly two years later, she was told she had to revise or delete substantial portions of her work, even though she says it contained no classified information. Bakos initially expected to be able to publish in 2016; it now looks like her book may go to print in 2019. Mark Fallon, a 27-year veteran of the NCIS, and former chief investigator at the Guantanamo Bay detention center, had his book on government torture held up for over seven months last year. As a result of the delay, he missed a submission deadline and had to cancel a book tour. His book was cleared only after the Knight Institute and the ACLU intervened on his behalf. 

Comey does not seem to have encountered similar difficulties.  Comey signed his book deal with Flatiron in August of last year. Unless Comey already had a complete manuscript in hand at that time—highly unlikely, given the topics the book is said to address—the government’s censors could not have had the book for very long.  And prepublication review does not seem to have delayed the book’s publication at all.  The book was initially scheduled for release on May 1, but after interest in the book spiked, Flatiron moved the publication date up by two weeks. 

The efficient resolution of any prepublication review is a good thing, and that is no less true in Comey’s case. As his publishers note, there is a demand for the former FBI director’s voice to be heard in the “urgent conversation” now taking place about the bureau. As special counsel Robert Mueller’s investigation into Russian meddling in the 2016 election ramps up, so too do efforts to discredit it.  The insights and experiences of someone intimately involved with the events leading up to the investigation are welcome at this juncture. But it is worth pausing to consider the hurdles that Comey’s book would have had to clear—not only to understand its journey, but also to better understand the obstacles to writing that other former public servants must face.   

The FBI’s prepublication-review scheme suffers from all of the problems associated with prior restraints. The scheme consists of a hodgepodge of instruments, including a federal regulation, 28 C.F.R.  § 17.18; the FBI Employment Agreement; the FBI Prepublication Review Policy Guide; and various nondisclosure agreements.  It is difficult to make sense of the relationship between all these instruments. After close review, our understanding is that Section 17.18 requires all FBI employees and contractors granted access to Sensitive Compartmented Information to sign a nondisclosure agreement that includes a provision relating to prepublication review.  Although Section 17.18 does not identify the nondisclosure agreement, other sources indicate that it is Form 4414, which imposes a lifetime prepublication-review obligation.  Separately, the FBI Employment Agreement requires all FBI personnel (current and former) to submit to prepublication review in accordance with the Policy Guide.

The Policy Guide sets out submission and review criteria that go far beyond what is needed to identify classified information. It requires FBI employees and contractors to submit proposed disclosures containing “any knowledge gained through FBI employment or assignments related to the FBI.” Submissions are reviewed for a long list of “prohibited” information categories. Classified information is but one item of this list. Reviewers may also censor information that “the FBI would have discretion to withhold pursuant to civil discovery obligations, the Freedom of Information Act, or any other statute, law, or regulation.”

As one court noted in 2009, the mere fact that the FBI has discretion to withhold information under a statute such as the FOIA “cannot negate or override the First Amendment inquiry.” Yet the Policy Guide makes clear that the Bureau continues to treat it as if it does. This fact, coupled with the absence of a definite time limit for review, creates a serious risk of arbitrary and discriminatory enforcement, one that has been borne out.

Again, it is good news that Comey’s book did not suffer at the hands of this caprice.  The public will get to hear his story just as its appetite for it is at a peak.  But it is worth asking what stories the public will not hear because of the dysfunctional prepublication-review system—either because the system suppresses those stories or because the prospect of having to submit manuscripts for review dissuades some former public servants from writing at all.

Three years ago, the Senate Intelligence Committee released the summary of its torture report, replete with shocking details about the way the CIA once waged the war on terror. Yet it was only part of the story. Of the full 6,000-page report, only 525 pages were released to the public, and those were heavily redacted.

Today, there is much we still don’t know about the George W. Bush administration’s fateful decision to use torture as a weapon of war, including how the agencies applying the policy interacted with one another. But a new book could offer some insight—if the Trump administration ever permits it to be released.

The book, Unjustifiable Means, is a forthcoming memoir by Mark Fallon, a former official for the Defense Department and Naval Criminal Investigative Service. Fallon retired in 2010 after a distinguished 31-year career—much of it on the front lines in the fight against Al-Qaeda. Before 9/11, Fallon worked with the task force that investigated the first World Trade Center attack and was the leader of the NCIS team that probed the bombing of the U.S.S. Cole. After 9/11, he was NCIS chief of counterintelligence for Europe, Africa and the Middle East, and later served as the deputy commander and chief investigator of the DOD Criminal Investigation Task Force, which gathered evidence to prosecute terrorism suspects by military commissions at the U.S. prison in Guantánamo Bay. Given Fallon’s insider background, his book may be an incredibly important account of the policy failures that damaged our national security by helping boost jihadi recruitment, among other things.

Some worry that the censors may be more concerned about protecting individual reputations than improper disclosures.  

Yet the book may never be released—or, if it is, may appear only in a heavily redacted, bowdlerized form. Fallon’s story is currently undergoing a pre-publication security clearance review. Ten agencies, including the Departments of Defense and Justice and the CIA, the three agencies most responsible for the torture fiasco, are combing through it. And some, myself included, worry that the censors may be more concerned about protecting individual reputations than improper disclosures. Fallon was initially promised this process would take no longer than six weeks; it has now dragged on for more than seven months, with no end in sight.

As someone who has closely followed—and participated—in the war on terror, I’m dismayed by the prospect of Fallon’s book being censored. When I was the Navy’s general counsel, Fallon was one of several NCIS officials who came to my office in November 2002 to present me with his concerns. He, along with NCIS Director David Brant and others, had obtained evidence that some Guantánamo detainees were being subjected to unlawful physical abuse during interrogations. If unchecked, they felt, the abuse would rise to the level of torture. Leaving aside whether torture was legal, they also felt that the harsh interrogation techniques being used were inherently counterproductive and demonstrated shocking incompetence. Lastly, they told me, the brutality was rumored to have been authorized “at the highest level” of the Bush administration.

The NCIS team in my office was composed of the types of officials that our nation values most: tough, courageous, capable and experienced people who are eager to combat the enemy and—no less important—faithful to the laws and values that define our country. None of them wanted to condone illegality, incompetence or dishonor, no matter how high the rank of the official ordering otherwise. By refusing to participate in and speaking out against the abuse of enemy captives, Fallon and his NCIS colleagues were defending their integrity and our nation. 

Fallon, who believes that it is as important to learn from our mistakes as from our successes, continues to serve our nation. Unjustifiable Means delves into how key officials were able to manipulate governmental processes to adopt counterproductive policies; it recounts the leadership, ethical and moral challenges that he and others faced in telling truth to power and illuminates how torture contributed to the proliferation of the global violent extremism that jeopardizes our national security.

These lessons are critical. The Obama administration commendably banned torture, but it was mistaken in believing that we could erase the temptation to use these harsh measures without meaningful accountability. Now, we are led by a president who is an unabashed torture enthusiast, is appointing architects or apologists of the Bush-era interrogation policies to key government positions, has abandoned U.S. leadership on human rights and treats police brutality as a laughing matter. The slide toward brutishness continues.

Unjustifiable Means can help thwart that slide—if the Trump administration would release it already, in its unredacted form.