All government requires secrecy, but self-government requires transparency. The core governmental activities needed to promote national security—diplomacy, military preparations and actions, and intelligence in all its forms—are never conducted according to a rule of full disclosure. But governments that derive their just powers from the consent of the governed jeopardize their own legitimacy if they withhold or distort information the governed need to assess the governance to which they are consenting. Governmental secrecy not only impedes the formation of sound public judgment, but can facilitate abuses of power. Arthur Schlesinger, Jr., once noted that although secrecy is sometimes necessary, it more frequently served “to protect the executive branch from accountability for its incompetence and its venality, its follies, errors and crimes.”

Across more than two centuries, the United States has repeatedly confronted the challenge of relying on state secrecy when the public interest requires it, while resisting the misuse of secrecy to advance the private interests of those in power. Today, more than ever, this is a fraught issue in American life. Thanks to Edward Snowden, Bradley Manning, and a host of others, the American public is learning about highly secret matters, like the techniques of domestic surveillance, with which it is vitally concerned. Yet we are also being warned by responsible officials that exposing those secrets is damaging our ability to fight terrorism, thwart nuclear proliferation, compete with aspiring powers like Russia and China, and protect our other interests around the globe. The result is a fierce national debate about defining and enforcing the limits of secrecy. Rahul Sagar’s Secrets and Leaks: The Dilemma of State Secrecy, which emerged out of a Harvard doctoral dissertation, could not be more timely.

* * *

Sagar, who now teaches politics at Princeton, begins at the beginning, with the American Founding. He persuasively refutes the view, shared by Schlesinger and others, that the framers of the Constitution favored nearly unlimited transparency. Exploring 18th-century republican thought, he finds a wealth of evidence demonstrating that the leading thinkers of the day took it for granted that statecraft requires, in the words of Scottish Enlightenment philosopher Francis Hutcheson, “secret and speedy execution” if it is to succeed. The practical men who devised our Constitution—in secret proceedings behind closed doors—designed institutions in which secrecy had a place even as they understood it could generate a host of evils.

Having examined our intellectual and legal traditions, Sagar takes up the central question: which institutions in the framers’ system have the power to ensure that secrecy is employed only for proper ends? The Catch-22 is that identifying abuses of secrecy requires access to the very material being kept secret. That is the fundamental “dilemma” noted in Sagar’s subtitle, and the framers were almost entirely silent about how to resolve it.

* * *

Two centuries of historical experience have provided work-arounds, but none is altogether satisfactory. The third branch of government—our judicial system—is one potentially powerful check on executive secrecy. The courts, after all, possess the authority to review executive branch compliance with the laws mandating transparency. Thus, the Freedom of Information Act requires disclosure of government documents on request, but also exempts nine categories of information, creating plenty of ambiguity and room for litigation. When the government has been challenged, courts have been extraordinarily deferential to the executive branch, ruling that it alone has the responsibility and the competence to determine what kinds of disclosures might cause harm.

A fascinating case in point discussed by Sagar is the 1980 Freedom of Information Act suit brought by activist Morton Halperin, who asked the CIA to disclose the identity of some private attorneys it retained, as well as the fees it paid them. The CIA argued that neither the names nor sums could be disclosed, lest it “give leads to information about covert activities that constitute intelligence methods.” Explaining how such information could lead to that result, the agency maintained that “[i]f a large bill is incurred in a covert operation, a trained intelligence analyst could reason from the size of the legal bill to the size and nature of the operation.” This seems implausible, and on its face even an abuse of the government’s power to withhold information from the public. Federal judge Malcolm Wilkey ruled for the CIA, however, explaining that “a court, lacking expertise in the substantive matters at hand, must give substantial weight to agency statements.” Such judicial deference—rooted in the fact that the separation of powers entails a division of labor and competencies—has limited the judiciary’s ability to patrol the boundaries of executive branch secrecy.

Congress could serve as a check on it, too, but Sagar shows how its own institutional handicaps compromise legislators’ ability to do so. Though vested with considerable power regarding national security and foreign affairs, Congress has a very limited capacity to monitor, let alone discipline, secret operations of the sprawling executive branch. The president and his subordinates, in turn, are justifiably apprehensive about sharing highly sensitive matters with a body composed of so many members. As a result, they regularly invoke executive privilege to avoid an accidental, or intentional, security breach.

Sagar entertains various reform proposals to expand the scope of congressional oversight, like sharply curtailing the deployment of executive privilege or creating a core group in Congress empowered to oversee everything the executive branch does. He rejects the former on the grounds that anything less than absolute secrecy would sometimes imperil national security and Americans’ lives, as in 1979’s “Canadian caper” in Iran, the subject of the movie Argo. And he rejects the latter approach for not really addressing the fundamental dilemma at all. Permitting only a small number of elected representatives to know all executive branch secrets will not allay fears about the abuse of secrecy. Rather, the members of this core group will themselves become the objects of suspicion insofar as “their own conduct as overseers is shielded from public view.” In the end, Sagar concludes that given the executive branch’s “stranglehold over the flow of national security information, there is little reason to believe that lawmakers will be able to take the lead in uncovering policies and actions that the president has decided to conceal” (emphasis in the original).

* * *

This brings Sagar to the most interesting and original section of his fascinating book: the press and its interaction with “whistleblowers.” He examines whether leaks can be a useful “regulatory mechanism” for constraining untoward executive secrecy. The dilemma is that while leaks can serve as an “effective and credible” means of policing the executive, their unlawful nature undermines their legitimacy. A system where leaks are not condemned but condoned would have serious consequences and be fundamentally anti-democratic. The parties involved in leaks—government bureaucrats who illicitly disclose confidential information, and journalists who disseminate it—are self-appointed, private arbiters. Allowing leakers and journalists to determine what constitutes an illegitimate government secret, argues Sagar, “violates the democratic ideal that such decisions should be made by persons or institutions that have been directly or indirectly endorsed by citizens.” When, instead, the decisions about which secrets are kept and which revealed are made by government officials exceeding their authority, by the journalists they leak to, and by the publishers who run the journalists’ revelations, the leaks “constitute a form of usurpation.”

Sagar shows that whistleblowers often disclose wrongdoing within government, but their particular form of law-breaking is not always civil disobedience properly understood. He specifies five conditions that whistleblowers must meet if they are to be lauded, not condemned, for violating the web of regulations and laws in which they are situated. First, their revelations must disclose genuine wrongdoing, which is more often contestable than clear. In particular, a disagreement over policy is not the same as an accusation of illegality, though all too often self-appointed whistleblowers collapse the distinction. Second, the evidence a whistleblower adduces to demonstrate his claim of governmental misconduct must be “clear and convincing,” not “fragmentary.” Third, an unauthorized disclosure “should not impose an undue or disproportionate burden on national security.” Putting out “the truth at any cost” is illegitimate in circumstances where it does more harm than good, which can be the case when the country is under threat or at war. Fourth, the whistleblower should “utilize the least drastic means of disclosure” by exhausting all avenues for complaint within the federal government before going outside it to the press. Finally, anonymity is a perversion of democratic norms, making it “difficult for the public to discern whose interests the disclosure is serving, and to take appropriate steps to counter the possibility of manipulation.”

* * *

This five-part test can be criticized, but Sagar methodically considers and cogently answers various objections. In actual practice, the complexities of governance mean that his test may not equip us in every case to make crisp judgments about the ethics of disclosure. But it provides a useful starting place for thinking about whistleblowing. In particular, his framework condemns the behavior we see on display today from Edward Snowden and Bradley Manning, both of whom fall far short of meeting Sagar’s qualifications for justifiable leaks. Snowden meets only one of the five criteria: he disclosed his own identity. But he chose to do even that from Hong Kong and Russia, locations beyond the reach of American law. This was civil disobedience without the consequences, which is not civil disobedience at all. (Agreeing to subject oneself to the possibility of prosecution and punishment should have been Sagar’s sixth condition. Choosing a life in exile as a fugitive from justice has nothing in common with civil disobedience.) Considering the immense damage Snowden has done to our national security, and the way he did it, Snowden is less whistleblower than traitor. Some of the journalists who were awarded Pulitzer prizes for obtaining and disseminating Snowden’s revelations have assiduously—and to my mind, disingenuously—sidestepped this distinction, calling Snowden simply a journalistic “source.” Others among them—adherents of an extreme left-wing brand of libertarianism—have preferred to laud him as a hero.

Manning also does not fare particularly well on Sagar’s grid. If he had stopped with the video he provided to WikiLeaks of two Apache helicopters firing their cannons at targets in Iraq, he would have had at least a prima facie case to be considered a whistleblower, even though the Pentagon’s subsequent legal reviews found no evidence of illegal conduct by American forces in this incident. But Manning did not stop there. His anonymous and indiscriminate leaking of a vast trove of sensitive documents is precisely what turned him into the malefactor now serving a long military prison sentence.

The Snowden and Manning cases are relatively easy to assess, but other leaks—some of them far less damaging to our security and also quite informative to the public—are today the daily substance of our national news outlets. At the same time, the extent of governmental secrecy has become an urgent subject of discussion and debate. Although Rahul Sagar’s Secrets and Leaks bears too many traces of a doctoral dissertation to attract a broad audience, it is a thoroughly researched, thoughtfully considered work that clarifies an unsolvable dilemma at the heart of democratic governance.