President Donald Trump’s former National Security Advisor John Bolton would have been confident. His indulgent The Room Where it Happened: A White House Memoir pitted him against the administration in a not infrequent battle over material that is published by former officials recounting their giddy days in high office. On June 17, the US government filed a civil suit seeking a preliminary injunction ahead of the planned release of the memoir on June 23, and a “constructive trust” arising from all profits issuing from the publication of the work.

Bolton had, as Jack Goldsmith and Marty Lederman point out, signed two separate, fundamentally similar non-disclosure agreements, “corresponding to two different sets of Specialized Compartmented Information programs to which he was afforded access.”

Publishing sensitive national security information in the US context is governed by that driest of documents known as Standard Form 312. Bolton undertook that he would “never divulge classified information to anyone unless: (a) [he has] officially verified that the recipient has been properly authorized by the United States Government to receive it; or (b) [he has] been given prior written notice of authorization from the United States Government … that such disclosure is permitted.” The second feature of the agreement is that Bolton agreed that, should he be “uncertain about the classification status” of any information in question, he would “confirm from an authorized official that the information is unclassified before [he] may disclose it.”

This was not all. To further supress information that would otherwise make it into the public domain is Standard Form 4414, which covers “Special Access Programs”, referred to in the field as sensitive compartmented information (SCI). The policing authority in this case is the National Security Council, which required Bolton to submit to review “any writing … that contains or purports to contain any SCI or descriptive of activities that produce or relate to SCI or that I have reason to believe derived from SCI, that I contemplate disclosing to any person not authorized to have access to SCI or that I have prepared for public disclosure.”

Judge Royce Lamberth of the US District Court for the District of Columbia was not convinced by arguments made by the administration for a preliminary injunction halting the memoir’s publication. But this did not necessarily make Bolton an endearing defendant. The judge admitted that “Bolton’s unilateral conduct raises grave national security concerns” but found that “the horse is out of the barn”.

Ultimately, Bolton’s decision to go forth with the publication without final clearance from the intelligence censors was incautious but irreversible. The judge even conceded that “Bolton may indeed have caused the country irreparable harm.” The point was rapid, vast distribution and spread, assisted by the nature of technology. In “the Internet age, even a handful of copies in circulation could irrevocably destroy confidentiality.” All was required was for a determined individual, armed with the contents of such a publication, to “publish [it] far and wide from his local coffee shop.” Resigned, the judge conceded that “the damage is done. There is no restoring the status quo.”

To that end, any injunction “would be so toothless”. The other obvious point – that over 200,000 copies of the book had already been shipped domestically, with thousands of copies being exported to booksellers in Europe, India and the Middle East – rendered the need for such a restraint moot. “By the looks of it,” mused the judge, “the horse is not just out of the barn – it is out of the country.”

The Bolton episode underscores the very legitimacy of the prepublication review process. Former CIA operative John Kiriakou makes the unimpeachable point that such documents, however sympathetic their authors, need to get into open circulation. The republic needs the oxygen of revelation. The process of review, he attests, is “deeply flawed and frequently political.” As Kiriakou reminds us, such a system of suppression drew breath from the case of Victor Marchetti, who worked as an analyst at the CIA between 1955 and 1969. Serialised versions of his book reflecting on the grand old days were slated to run in Esquire. The CIA took issue, filed a temporary injunction against publication of the book citing the presence of classified information and the naming of undercover operatives. The case made its way to the US Supreme Court, which held that the initial judgment in favour of an injunction was sound. The non-disclosure regime was appropriate. “We find the contract constitutional and otherwise reasonable and lawful.” What followed was an arduous process of review, cutting and redaction, with Marchetti seeking clearance, and the CIA being miserly in concession.

Not all was lost for former members of the intelligence community and publishers. Texts might still make it into circulation, provided they were cleared, and done so within 60 days by the relevant prepublication board. Those not cleared might see profits confiscated. But this did not address the issue of zealous overclassification, unnecessary redaction and violations of the 60 day rule.

The battle against the very constitutionality of the prepublication review system has begun in earnest. On January 27, 2020, the Knight First Amendment Institute at Columbia University and the ACLU filed a Freedom of Information request seeking records related to the review of the manuscripts of 25 former federal officials, among them Bolton’s memoir. In April 2, 2019, the Knight First Amendment Institute filed a lawsuit challenging the very constitutionality of prepublication review. Along with the ACLU, the action was undertaken on behalf of five former public servants arguing that the prepublication system spanning the CIA, the Defence Department, the National Security Agency and the Office of the Director of National Intelligence, violated the First Amendment right “to convey and of the public to hear, in a timely manner, the opinions of former government employees on issues of public importance.”

The action further argued that the prepublication process violated the First Amendment in not providing former employees “with fair notice of they can and cannot publish without prior review”, one that also invited “arbitrary and discriminatory enforcement by censors.”

On April 16, 2020, the District Court in Maryland found in favour of the government, holding the prepublication review system to be constitutional. Judge George Hazel found that the ACLU and Knight First Amendment Institute had standing to challenge the review process, but felt governed by the forty-year old Supreme Court case of Snepp v. United States. The defects of the prepublication system, be it in terms of vagueness on classification, the certainty of review standards, and the absence of procedural safeguards, had little bearing on the question of constitutionality.

The plaintiffs have duly appealed. Among their arguments is the fact that Snepp focused on remedies rather than First Amendment principles, sidestepping the very merits of the CIA review system. The limits of government authority in imposing prepublication review obligations also remained untested. The reasoning of Snepp has also aged, both in terms of the law of pre-restraint on employee obligations and the factual environment. As the Knight First Amendment Institute urges, “We need to hear these voices [of former employees of the intelligence services], but if we want to hear them, we have to fix the obstacle course that prepublication review has become.”