Vol. 27 No. 2 (February 2017) pp. 24-27

NATIONAL SECURITY AND DOUBLE GOVERNMENT by Michael J. Glennon. New York: Oxford University Press. 2015. 257pp. Cloth $31.95. ISBN: 978-0-19-020644-4. Paper $14.95. ISBN: 978-0-19-066399-5.

Reviewed by Daniel N. Hoffman, Professor Emeritus, Johnson C. Smith University.

Law professor Michael Glennon offers a fascinating and timely essay on the workings of our actual, as opposed to our official, form of government. His thesis is that, at least within the domain of national security affairs, the classic, Madisonian model of three independent, mutually checking branches has ceased to operate. Instead, we have developed and empowered a “Trumanite” network: the top managers of military, intelligence, diplomatic and law-enforcement agencies—unelected bureaucrats who are substantially impervious to congressional, judicial and even presidential control, and free to continue the same policies without regard to the results of elections or public opinion.

Chapter 1, “Introduction,” remarks that, after campaigning vigorously against many of his predecessor’s national security policies, Barack Obama wound up keeping in place or even expanding most of them. Glennon considers and rejects as inadequate two possible explanations: that the policies were continued because they were the only rational possibilities, or that Obama happened to be personally or situationally incapable of following through with his previous views. Instead, the author draws on Walter Bagehot’s (1966) 1867 view of the British Constitution: that the real work of governing was done by modern, “efficient” institutions, not by the “dignified” ones that, in theory, had the most legitimate authority. Over time, a massive and poorly-understood shift in power from monarch and Lords to Prime Minister and Commons had occurred, creating a system of “double government,” which, if widely understood, would create a crisis of public confidence. In America the institutions differ, but dramatic changes have likewise been driven by the demands of efficiency and enabled by public naiveté.

Chapter 2, “The Trumanite Network,” traces the network to the National Security Act of 1947, which set up the CIA, the National Security Council, and the Joint Chiefs of Staff. These reforms enjoyed enthusiastic liberal support in Congress and were criticized by some conservatives. The new agencies are run by several hundred officials, some politically appointed and some career civil servants. Their culture emphasizes tough-minded realism, team loyalty, and a preference for exaggerating threats rather than underestimating them. “The fundamental driver of Trumanite power has been emergency, the appearance of threats that must be addressed immediately, without bringing in the Madisonian institutions” (p. 21). In this perspective, lessened accountability is a benefit, not a cost. The network is compulsively secretive, deflecting efforts to learn who is responsible for a given policy. Though it is not a monolithic hierarchy, the range of internal disagreement is narrow, and major policy shifts are hard to come by. Crucially, the network is autonomous. The Madisonian institutions appear, as legitimacy requires, to be in charge, but cannot actually be so if the system is to function efficiently.

Chapter 3, “The Sources of Madisonian Illusion,” explores the necessary illusion that the President is the “decider,” despite presidents’ general deference to Trumanite experience and expertise. The dignified Madisonian branches are glorified by civic education and ritual, while the bureaucracy is obscure and complex. Interactions are typically cloaked in mystery, but the media reinforce the illusion of presidential control, and the appearance of harmony is generally maintained. Yet the double government system does have exceptions: “Enough counterexamples must exist to persuade an optimistic public that the reason for [*25] policy continuity is human, not systemic…. Congress, the President, and the courts do sometimes say no to the Trumanites. But they do not do so often enough to endanger double government” (pp. 36-37).

The fourth and longest chapter, “The Reality of Madisonian Weakness,” marshals evidence to flesh out and support Glennon’s central claims. First he discusses the recruitment of federal judges and their often close ties to the national security state. “Whatever the court, judges normally are able to find what appear to the unschooled to be sensible, settled grounds for tossing out challenges to the Trumanites’ projects” (p. 46). Judicial deference is generally extended to sweeping presidential powers that are freely delegated to subordinates. Despite a few modest counterexamples that preserve the façade of judicial independence, the courts are not a serious check on the Trumanite network. Glennon’s findings are consistent with David Rudenstine’s (2016) book-length treatment of the judicial role. It is noteworthy, however, that the deferential case law is far older than 1947: it goes back at least to the Civil War.

Glennon next assesses the weaknesses of Congress that flow from lack of expertise, withholding of classified information, and aversion to political risk. In consequence, “virtually everything important on which national security legislation is based originates with or is shaped by the Trumanite network” (p. 51). Further, an indifferent public tolerates extremely feeble oversight of policy implementation. This analysis is consistent with Scott Horton’s (2015) book-length treatment on the weaknesses of Congress.

Turning to the role of the president, Glennon contrasts his double government thesis with that of the imperial presidency. The key point is that top-down orders for fundamental policy shifts are rare; most presidential “orders” to the Trumanites are actually formulated by the latter. On occasion, they simply proceed without presidential approval or even knowledge. Like the courts and Congress, presidents are intimidated by their expertise and usually go along with their actions or choose among the limited options they offer. “No president has reserves deep enough to support a frontal assault on the Trumanite network” (p. 64). Despite exceptions that sustain the impression of presidential control, the range of presidential choice is tightly hemmed in.

Chapter 4 concludes with a case study of NSA surveillance. Its upshot is that, despite disclosed abuses and public concerns, the Obama administration largely continued the established policies, with a few largely cosmetic reforms. This account is consistent with Charlie Savage’s (2015) book-length treatment on this issue.

Chapter 5, “Plausible Alternative Explanations for Policy Continuity,” elaborates on the alternative explanatory models introduced in Chapter 1. The Rational Actor Model’s defects include its failure to prescribe what should be done in a context of multiple, competing values and multiple, alternative means to a given end. In addition, it fails to acknowledge that rationality is often bounded by cognitive distortions. The Government Politics Model recognizes the pervasiveness of conflict, compromise and confusion in policymaking, but its all-inclusive complexity makes it impossible to identify the but-for causes of specific outcomes. The Organizational Behavior Model, in contrast, highlights key issues of organizational recruitment and culture, standard operating procedures, biased risk assessment, and power allocation and information flow within the network. Each organization, seeking to enhance its influence, develops stable preferences that are little influenced by individual members and leaders. The Trumanite network is not a single organization, but its member organizations share a number of common behavior patterns, with a resulting tendency to “groupthink.” Glennon acknowledges that the alternative explanatory models are not flatly mutually exclusive. Often, a given fact can be taken to support more than one of the overlapping models. “The notion of double government predicts only that national security policy will change little from one administration to the next. It does not predict with the same level of probability whether any [*26] given element of national security policy will remain constant” (p. 90).

Chapter 6, “Is Reform Possible? Checks, Smoke and Mirrors,” argues that prospects for reform are hindered by the weakness of the Madisonian institutions that would have to effectuate it. Behind this lies their lack of solid public approval and the associated weakness of external checks such as media and public opinion. Scholars like Posner and Vermeule (2013) place mistaken confidence in the latter, since political checks cannot succeed without effective legal checks. Stopgap institutional reforms to bolster congressional or judicial power have accomplished little in the past. Glennon argues, “[C]ontinued focus on domestic legalist Band-Aids merely buttresses the illusion that the Madisonian institutions are alive and well” (p. 101). Another approach would insert new, internal legal checks into the Trumanite institutions themselves; this too has been tried with little effect.

In Glennon’s view, the Madisonian model has ultimately failed from lack of the requisite public virtue. “Absent a virtuous electorate, personal ambition and institutional ambition no longer are coextensive. Members’ principal ambition then becomes political survival ….” (p. 105). Given the public’s passivity, all branches share the incentive to defer to the Trumanites. Glennon thus considers the civic-republican argument that government must strive to cultivate civic virtue. Paradoxically, this would entail its own dangerous expansion of governmental power. The author believes that the public is already possessed of sufficient intelligence and judgment, but that it quite rationally refrains from engaging with issues that it cannot control in any event.

In Glennon’s concluding chapter he briefly restates the case that slowly tightening centralized power is a threat to democracy. The Trumanite network is more dangerous than other bureaucracies: “it has the power to kill and arrest and jail, the power to see and hear and read peoples’ every word and action, the power to instill fear and suspicion, the power to quash investigations and quell speech, the power to shape public debate or to curtail it, and the power to hide its deeds and evade its weak-kneed overseers. The Trumanite network holds, in short, the power of “irreversibility” (p. 118).

This brief—the footnotes double its length—and sobering text is well-worth reading, and suitable for courses in national security policy, bureaucracy or constitutional theory. The terms “Trumanite” and “double government” may not be rhetorically ideal, and Glennon might have made more of the roles of the political party system and imperial expansion in undermining the Madisonian model. Yet he makes a substantial contribution to the literature about the threats facing our constitutional order. The book is also extraordinarily timely, given that President Donald Trump campaigned vigorously against many features of the Trumanite consensus, has repeatedly denigrated their competence, and seems bent on shaking up the policymaking process as well as the content of policy. Glennon’s theory is not designed to predict the precise outcome of any specific controversy. Yet, if he is correct, Trump is more likely to create a systemic crisis of confidence than to bend the Trumanites to his will. Unfortunately, the course of such a crisis would likely turn mostly on Trump’s general popularity, less on his specific policy preferences, and scarcely at all on a public grasp of the systemic problems analyzed in this book.


Bagehot, Walter. 1966. THE BRITISH CONSTITUTION. Ithaca, New York: Cornell University Press.


Poser, Eric and Adrian Vermeule. 2013. THE EXECUTIVE UNBOUND. New York: Oxford University Press.


Savage, Charlie. 2015. POWER WARS: INSIDE OBAMA’S POST-9/11 PRESIDENCY. New York: Little, Brown and Co.

© Copyright 2017 by author, Daniel N. Hoffman.