Vol. 25 No. 4 (April 2015) pp. 56-61

SECRETS AND LEAKS: THE DILEMMA OF STATE SECRECY by Rahul Sagar. Princeton, NJ: Princeton University Press. 2013. 304pp. Cloth $35.00. ISBN: 978-0-691-14987-5. E-book $35.00. ISBN: 978-1-400-84820-1.

Reviewed by Daniel Hoffman, retired from Johnson C. Smith University. Email:

Rahul Sagar’s book is timely, in view of debates about alleged abuses connected with the War on Terror, the invasion of Iraq, and the massive leaks by Chelsea Manning and Edward Snowden. It is well-written and copiously footnoted, with an extensive bibliography of historical, legal and political science sources.

Sagar’s Introduction, Who Watches the Watchers, emphasizes the difficulty of designing an effective regulatory framework to prevent the abuse of state secrecy. “[S]o long as there is state secrecy, our ability to guard against its misuse depends not so much on the checks and balances established by the Constitution as on the virtues and vices of those men and women who secretly take the law into their own hands in order to either open our eyes or close our minds” (p. 7).

Though this statement discounts the importance of constitutional checks, Chapter 1, The Problem, places the Constitution at the heart of the dilemma. For Sagar, the crucial aspect of the Constitution is the sweeping powers it vests in the president, on whose virtue we must then depend.

To support the proposition that the Framers subscribed not to a “principle of disclosure” but instead to one of secrecy, Sagar cites “republican” writers from Renaissance Italy and the absolutist Stuart monarchy, who taught that the flourishing of a state in a world of international and domestic conflict depends on a regular and secure practice of secrecy. The capaciousness of his use of “republican” is evident in the book’s opening epigram—a quote from Machiavelli’s THE PRINCE. Machiavelli indeed taught that “virtu” was the prime ingredient for princely success, and the sole restraint on his actions. He returns on the final page, where Sagar exhorts us to “forgo platitudinous calls for ‘transparency’ and quixotic endeavors to tame ‘the prince’” (p. 204).

Sagar next considers the implications of the constitutional separation of powers. He argues that only the executive is endowed with the information, unity and energy needed to make national security decisions wisely and quickly, and to keep them confidential. Sharing the relevant information with Congress is unsafe, since individual members are prone to leak secrets confided to them, even when the body has opted for secrecy. Thus, disclosure to Congress cannot be deemed constitutionally required. Courts, meanwhile, lack expertise and must be deferential when national security is at stake.

After convincingly showing, by many episodes over many years, that “neither Congress nor the courts have intervened strongly” against excesses of executive secrecy (p. 48), Sagar goes on in chapters 2, Should We Rely on Judges, and 3, Should We Rely on Congress, to argue that they are structurally incapable of doing so. Due to want of information, of incentives, and of effective sanctions, they cannot reliably compel disclosure. Where national security is concerned—and Sagar insists it is for the President alone to determine when that is the case—the separation of powers is hopelessly ineffective. Besides, he argues, it is a luxury we cannot afford; the other branches have [*56] wisely accepted doctrines such as the state secrets privilege and executive privilege, recognizing that “necessity knows no law” (p. 82). In the end, it seems, those doctrines are grounded more in perceived necessity than in constitutional law.

Sagar challenges critics of secrecy to prove that; overall, the harms caused by abuses of secrecy exceed those caused by unauthorized disclosures (p. 93). Having deftly shifted the burden of proof to the other side, he needs to say little about the harms or benefits of the many secrets and leaks he offers as examples. Instead, he falls back upon the claim that, because the other branches lack the expertise and institutional capacity to keep secrets, mandated sharing of information with them cannot be safe.

His historical and institutional analyses lead Sagar to the very plausible conclusion that the most effective practical check against abuses is the leaking of the secrets. This prompts the questions, Should the Law Condone Unauthorized Disclosures (Chapter 4), Should We Rely on Whistleblowers (Chapter 5), and Should We Trust Leakers (Chapter 6). Sagar provides insightful analysis of the options, incentives and risks that would-be leakers encounter. Unless the leaks are beneficial to the president, prospects for the leaker are generally bleak, suggesting that leaks might not be excessively common. Yet Sagar offers a normative claim ostensibly based on democratic theory: only the president is elected nationwide and empowered to determine the national interest on behalf of the People; thus, would-be leakers have no business substituting their judgment for his.

Sagar repeatedly recognizes that secrecy can be abused, but his vision of the virtuous presidency leads him to a very narrow definition of abuse. Not just any unlawful act will qualify, but only unlawful acts that deliberately pursue selfish rather than truly national interests—as if presidents do not generally believe that what is good for them is good for the country. Excluding Congress and the public from vital decisions in which they have a right to participate is not itself an abuse (pp. 127 -130). Given this narrow definition, Sagar easily concludes that, while on rare occasions leaks may be necessary to expose the gravest abuses, for the most part leaks are unjustified and potentially harmful. His foremost worry is that leaks can sometimes cause gigantic harms--which cannot themselves be publicly detailed, for fear of causing even more damage. Necessarily, his examples are hypothetical.

Though disabled from examining the motives and judgments of secret-keepers, Sagar speculates at length about the possible bad motives and judgments of leakers. By their “usurpation” of the president’s personal authority to determine what must be secret (p. 114), leakers make themselves subject to criminal sanctions, unless they can meet a stringent five-point test to justify their conduct. The disclosure must: (1) concern an abuse of public authority, as narrowly defined; (2) there must be clear and convincing evidence of wrongdoing (difficult or impossible to come by); (3) the leak must not pose a disproportionate threat to public safety (also hard to know); (4) the means of disclosure must be the least drastic possible (publication to the world being the most drastic); and, (5) leakers must identify themselves and be prepared to undergo the predictable formal and informal sanctions. Thus, leakers must accept the burden of proving that the president’s motives were improper, and, in addition, of proving that their own motives in leaking information are not biased by “sectional,” “partisan,” or “personal” motives.

The policy recommendations in Chapter 6 and in the Conclusion, Bitter Medicine, [*57] accordingly focus on steps to reduce the incidence of leaking. In particular, Sagar exhorts the press to employ greater self-restraint, and advocates for “an independent and well-funded organization dedicated to scrutinizing media performance, which could name and shame reporters and editors who misuse anonymous sources, and the publishers who condone such behavior” (p. 201).

SECRETS AND LEAKS has many rhetorical earmarks of a carefully balanced argument; yet its positions on questions of historical interpretation, constitutional theory and democratic theory are extreme, joining a growing recent body of neo-monarchist literature. Central here is the Machiavellian emphasis on voter-determined presidential virtue as the sole and sufficient safeguard. (Full disclosure: I have long argued that our Constitution imposes lawful restraints on the president, withholding monarchical powers.)

Sagar’s treatment of the founding period is highly selective. Historians conventionally report the conflict in the 1790s between Hamilton and Madison over the risks and benefits of executive power, concluding that the issues have remained unresolved. Sagar, however, barely examines Madison’s side of the argument, using contributions by Hamilton and Jay to THE FEDERALIST PAPERS to characterize the Constitution as Hamiltonian. While he provides ample evidence that the Framers often resorted to secrecy, those actions alone cannot establish valid legal precedents. Neither the full range of historical facts nor the opposing interpretations developed by me (Hoffman 1981, 1984) and other writers, such as Louis Fisher and Heidi Kitrosser, are seriously addressed.

For example, Sagar relies heavily on the so-called Jay Treaty precedent (as did Chief Justice Burger in UNITED STATES V. NIXON), when George Washington refused to share requested papers with the House of Representatives, which nevertheless, moved by partisan loyalties and fears of war, granted funds to implement the Treaty. Sagar seems unaware that this “precedent” was actually a sudden, unilateral, politically motivated departure from the previous practice, in which the president would forward requested documents, but sometimes ask Congress to keep them confidential. Moreover, the House formally protested this innovation, and President John Adams soon repudiated it in the XYZ Affair. Yet Sagar expects us to accept on faith that George Washington was virtuous, so that it must have been irresponsible to criticize his conduct. Nor does he recognize how the ramming through of Jay’s Treaty exacerbated the partisan divide, leading to ongoing popular and press protests that in turn provoked the notorious Sedition Law of 1798. This dynamic of secrecy leading to protests leading to repressive counter-reactions should serve as a cautionary precedent in its own right.

Sagar’s history also ignores the fact that, even though the Federalists in the 1790s were often outraged by leaks that they thought endangered their “infant empire,” none of the known miscreants was prosecuted. Their Sedition Law did not criminalize leaking, but only defamatory comments (Hoffman 1981a). Sagar’s more recent historical materials are extensive, but equally selective. He does not closely examine the long record of troublesome wars and treaties that we have entered under the influence of secrets and lies. For example, the roles played by secrecy and deception in launching and prolonging our Vietnam and Iraq wars are not assessed, because these are prime examples of politically controversial actions about which the president necessarily knows best. Sagar [*58] simply wonders how anyone can be so irresponsible as to question the “talents and integrity” of virtuous leaders from George Washington to Dick Cheney (p. 187). (The term “war crimes,” by the way, does not appear in the book.) He does show the great difficulty we have had in combatting excesses of secrecy, but his materials do not support his claim that “state secrecy is approved in principle and censured in practice” (p. 49); rather, they suggest the opposite.

Sagar’s constitutional theory is similarly unbalanced. Though our Framers explicitly attempted to avoid the tyranny of centralized power and to instill the rule of law, Sagar has them tacitly enshrining in the presidency a sweeping principle of secrecy. His claim that the Framers allowed the president a free hand when it comes to secrecy vitiates the separation of powers and the First Amendment. To capably perform their assigned roles, Congress, the courts, the press and the public often need access to information held by the executive branch; yet Sagar rejects the logical inference that each must have a corresponding power or right to obtain it.

In particular, First Amendment protection for leaking is fully consistent with the Amendment’s text, the Founders’ scruples, and the novelty of prosecutions for leaking. Yet, despite having shown that courts are unavoidably prone to defer to executive power claims, Sagar confidently relies on a line of court decisions, mostly quite recent, holding in effect that officials are properly bound to secrecy: they are presidential servants, not free citizens. Those decisions are vulnerable to strong criticism (Hoffman 1997, ch.4).

Only Daniel Ellsberg’s unauthorized disclosures enabled us to realize that our government had systematically used secrets and lies to mobilize and sustain public support for an effort of highly questionable value. Yet Sagar sees no constitutional bar to prosecutions for leaks under the 1917 Espionage Act--save perhaps for those of journalists, which he acknowledges might violate the freedom of the press (pp. 171-180). The notion that Ellsberg (or Edward Snowden, whose story post-dates the book) deserves to be honored and constitutionally protected from prosecution seems unthinkable.

Even further from Sagar’s grasp is the point that governmental secrecy, both literally and in every practical sense, is a prior restraint on speech. Its entire purpose is to prevent anyone, including foreign enemies, but also members of congressional oversight committees, courts, the press, and voters, from learning of and responding to what the government is doing or contemplating. By design, secrecy systematically undermines official accountability, checks and balances, and the rule of law. It accordingly deserves the same, highest level of scrutiny applied to other forms of prior restraint. There is no basis in the Constitution or democratic theory for carving out a categorical national security exception to the First Amendment (compare NEW YORK TIMES CO. V. UNITED STATES). In this domain too, the wisdom and justice of policy depend on accurate information and a free marketplace of ideas.

The broad discretion Sagar accords to the executive in the national security domain can easily extend into other domains and has in fact done so, just as domestic executive powers have expanded in tandem with international ones. As early as MARBURY, the Court sought to distinguish between official actions pertaining to foreign or defense affairs and disputes affecting individual rights, with the former being matters of discretion and the latter questions [*59] of law. Plainly, however, many actions fit into both categories. That an action may serve national security does not logically imply that no right is violated.
Moving beyond constitutional law, the DEMOCRATIC theory Sagar invokes must envision a far more active, influential role for the public than did classical republican theory, necessitating even broader access to information. Yet he asserts, without extended argument, that leaking of secrets is very seldom warranted, simply because our election system adequately ensures that presidents will generally be virtuous. The pronounced variations of presidential ability, as well as virtue, do not figure in Sagar’s argument. His strongly patriotic sentiments, reflected in the book’s dedication “To these great and glorious United States,” somehow translate into veneration of the presidency as an institution, almost regardless of the conduct of its occupant--Richard Nixon’s Watergate being the sole acknowledged exception. Nixon, of course, believed that “when the President does it, that means it can’t be illegal.” What if he had played the national security card, as he considered doing, and destroyed his tapes? Sagar’s “democratic” theory is tantamount to making the president sovereign, not the People.

Moreover, to argue as if the president personally made all decisions on secrecy wrongly conflates executive power with presidential power. Over one million officials currently have power to classify documents wholesale, subject only to vague and scarcely enforced standards, and presidents are easily misled by classified briefings. Sagar acknowledges that these unelected bureaucrats systematically favor secrecy over disclosure, and that, indubitably, way too much information is classified; yet these facts seem not to affect his weighing of risks and benefits and do not figure in his “democratic” theory. Secrets and lies have clearly played great roles in the decline of public trust in government in recent decades. To blame this decline primarily on irresponsible leakers and publishers seems perverse.

Sagar does endorse Jack Goldsmith’s advice that an administration “should be as open as possible, and when secrecy is truly necessary it must organize and conduct itself in a way that is beyond reproach, even in a time of danger” (p. 188). Yet he cannot prove that this has generally been the case—or even that it has ever been the case. Because it is hard to know what officials have done, let alone why they did it, Sagar advises that lawmakers, judges, and the press must “refrain from picking sides” (p. 202). Yet picking sides is central to both democratic politics and the judicial process. By not picking sides about the recent, unprecedented burst of prosecutions for leaking, and this under a president who had promised unparalleled transparency, we effectively side with rampant executive secrecy.



Fisher, Louis. 2004. THE POLITICS OF EXECUTIVE PRIVILEGE. Durham, NC: Carolina Academic Press.

Goldsmith, Jack L. 2008. “Secrecy and Safety.” NEW REPUBLIC, August 13.


Hoffman, Daniel N. 1984. “A Republic, if You Can Keep it.” MICHIGAN LAW REVIEW 82: 997-1006


Hoffman, Daniel N. 1981a. “Contempt of the United States: The Political Crime that Wasn’t.” AMERICAN JOURNAL OF LEGAL HISTORY 25:343


Kitrosser, Heidi. 2008. “Classified Information Leaks and Free Speech.” UNIVERSITY OF ILLINOIS LAW REVIEW 881.

Machiavelli, Niccolo. 2003. THE PRINCE. New York: Penguin Classics.

Rossiter, Clinton (ed.) 1961. THE FEDERALIST. New American Library.


MARBURY V. MADISON, 5 U.S. 137 (1803).


UNITED STATES V. NIXON, 418 U.S. 683 (1974).
Copyright 2015 by the author, Daniel Hoffman.