National Security State

It seems ages ago now, but there really was a time when some civil libertarians held out hope for Barack Obama's presidency. If elected, this former constitutional law professor might be "our first president who is a civil libertarian," Jeffrey Rosen enthused in The New York Times in March 2008. On inauguration night in 2009, defense lawyers at Guantanamo Bay actually formed a celebratory conga line, chanting "rule of law, baby!"

They and many other Obama optimists woke up to a hell of a hangover, one that's lasted six years. The president has launched more than six times as many drone strikes as George W. Bush; ordered the remote-control execution of an American citizen; continued and expanded dragnet domestic surveillance programs based on a secret interpretation of the PATRIOT Act; and launched two undeclared wars.

The question Michael Glennon asks at the outset of his important new book, National Security and Double Government, is: "Why does national security policy remain constant even when one President is replaced by another, who as a candidate, repeatedly, forcefully, and eloquently promised fundamental changes in that policy?"

His answer is altogether darker and more radical than you'd reasonably expect from a former Senate Foreign Relations Committee legal counsel and current international law professor at Tufts. Glennon argues, in essence, that the national security state has become a runaway train and that presidential elections are contests that determine who gets to pretend he's driving.

Glennon takes the book's central metaphor of "double government" from the 19th century British essayist Walter Bagehot, longtime editor of The Economist. In 1867's The English Constitution, Bagehot described how real power in the British government had quietly shifted from one set of institutions, the monarchy and House of Lords, to another: the prime minister, the cabinet, and the House of Commons. By the late 19th century, Britain had become a "concealed republic" with only the outward trappings of a monarchy.

The United States is moving in the opposite direction, Glennon argues. As power has shifted toward the permanent national security and intelligence bureaucracies, we face an "emergent autocracy" in the guise of a democratic republic. We've "moved beyond a mere imperial presidency," he writes, "to a structure of double government in which even the President exercises little substantive control over the overall direction of US national security policy."

We're used to the idea that Congress has ceded most of its formal powers over national security policy to an aggrandizing chief executive. But it's counterintuitive, to say the least, to suggest that government's chief executive isn't really in charge. That's likely why the two standard explanations for otherwise inexplicable policy continuity focus on the president as the main protagonist.

The first such explanation is what Glennon terms "the rational actor model," the idea that we get the national security policies we do because these are the national security policies we need, given the threats we face.

This is the account preferred by Harvard Law School's Jack Goldsmith, a former head of the Office of Legal Counsel during the Bush administration, who serves as something of a foil for Glennon. As Goldsmith once put it, "The presidency invariably gives its occupants a sober outlook on problems of national security."

Sitting in the Oval Office and getting a faceful of President's Daily Briefs tends to concentrate the mind, the story goes, so it's no surprise that when confronted with new information about the dangerous world we live in, Obama changed his mind about the substantive correctness of his predecessor's counterterrorism policies. Goldsmith quotes Jack Kennedy's observation that it's "much easier to make the speeches than it is to finally make the judgments."

The second conventional explanation for policy continuity is what Glennon calls the "government politics model." This emphasizes factors like the content of the president's character and the political pressures brought to bear on the presidency.

Maybe, for example, what the Daily Brief really concentrates the president's mind on is political self-preservation. He becomes ever more aware that he's going to be held personally responsible if a bomb goes off anywhere in the country, particularly if he's a liberal Democrat whose foes are eager to paint him as a soft-on-terror McGovernite.

Of course, it's insane to hold any elected official personally responsible for providing seamless protection to a country of over 300 million people; all the surveillance and drone strikes in the world can't begin to meet that boundless responsibility. But as Obama's onetime national security adviser James L. Jones put it: "Who wants to be the guy that says we don't need [these powers] anymore and then three weeks later something happens?"

Glennon offers another explanation, one that has been overshadowed by the conventional wisdom's focus on the president as decider in chief. Policy continuity is better understood through the "organizational behavior model," he says, which looks to a "Trumanite network" of managers in the military, intelligence bureaucracies, and law enforcement "who are responsible for protecting the nation and who have come to operate largely immune from constitutional and electoral restraints." Glennon calls them "Trumanites" because of our 33rd president's role in founding the CIA, the modern Defense Department, the Joint Chiefs of Staff, and the National Security Agency (NSA).

Though Glennon doesn't describe his thesis in terms of public choice theory, it echoes that discipline's insight that institutions are run for the benefit of the people who run the institutions. For the Trumanites, Glennon explains, "benefits take the form of enlarged budgets, personnel, missions; costs take the form of retrenchments in each." Witness the vast archipelago of intelligence facilities-nearly three Pentagons' worth of office space-that have been erected in greater Washington, D.C., since 9/11.

Security bureaucracies may sometimes resist new missions "seen as undercutting their culture or efficiency," but they will reliably err on the side of overprotection and threat inflation. "The fundamental driver of Trumanite power has been emergency," Glennon writes, and so "the network thus has little incentive to identify or eliminate the ultimate source of threats (e.g., unwanted intervention in the internal affairs of other nations)."

The national security state is becoming an autonomous, self-perpetuating entity, Glennon warns. It sets the table for elected officials' choices and increasingly dictates terms to them. The permanent bureaucracy basks in the "glow" of Madisonian institutions, drawing legitimacy from the illusion that elected officials are in charge. But while the buck may stop with the president, the real power resides with the Trumanites.

This explanation is strongest in the realm of state surveillance, which serves as Glennon's central case study. Recall the embarrassing revelation, in the summer of 2013, that the NSA was tapping German Chancellor Angela Merkel's cellphone. What did the president know, and when did he know it? If you believe top administration officials, Obama was almost as surprised as Merkel. Glennon quotes Secretary of State John Kerry to the effect that the Merkel wiretap, like a lot of NSA programs, occurred "on autopilot."

On one hand, that's what you'd expect them to say. On the other hand, the claim is entirely plausible, and it is consistent with the earlier history of NSA abuses uncovered by the Church Committee in the 1970s. Under Project SHAMROCK, for example, the NSA collected the content of virtually all cable traffic entering or leaving the United States for three decades-150,000 messages a month at its height. It was, the committee's final report concluded, "probably the largest governmental interception program affecting Americans ever undertaken." And yet it's not clear that any president ordered, approved, or was even aware of SHAMROCK. When the program's existence was exposed in the mid-'70s, Louis Tordella, longtime deputy director of the NSA, admitted that he didn't know whether any president or attorney general had ever been briefed on it.

The picture grows somewhat more complicated when we look at the modern practice of presidential war making. From the Truman administration onward, the president has accumulated enormous unchecked authority, despite James Madison's conviction that, since the executive department was "most distinguished by its propensity to war," it is "the practice of all states, in proportion as they are free, to disarm this propensity of its influence."

When it comes to picking the wars we wage, it's not clear that the Trumanites are fully in charge. Take four major war-powers decisions during the Obama administration: the Afghan surge, the escalation of drone attacks, the Libya intervention, and the current war against ISIS. I put the Trumanite win-loss record at roughly .500 here. The military and national security bureaucracy fought hard for the surge and the drone escalation, and got them. They generally opposed the Libyan action, and some prominent Trumanites-such as the chairman of the Joint Chiefs -appear to have been reluctant to endorse our latest war in the Middle East.

In the case of this most recent war, domestic politics seems a better explanation: The president yielded to the near-irresistible demand that he "do something" about the beheading of Americans and the implosion of the Iraqi state. Bombing ISIS is something, so we're doing it.

The Obama experience suggests we get the wars the Trumanites want-and also some they don't. But this is hardly fatal to Glennon's thesis. He stresses that "a good theory of institutional behavior can predict, at best, only tendency over time"; his "predicts only that national security policy will change little from one administration to the next." So far, that theory is holding up rather well.

Even so, I've always been partial to one version of the "government politics" explanation. A few years ago, I wrote a book arguing that "Americans' unconfined conception of presidential responsibility is the source of much of our political woe and some of the gravest threats to our liberties." If the political reality is such that the president will be held personally accountable for any domestic terror attack, don't be surprised when he seeks powers nearly as vast as the expectations put upon him.

Glennon acknowledges it's not either-or; "explanations overlap," he writes. Dumb wars and security-state overreach are the result of political choices and the bureaucratic imperative. Policy continuity is depressingly overdetermined.

Real-time histories of key national security decisions in the Obama years tend to underscore this point. In Kill or Capture, reporter Daniel Klaidman describes the enormous political pressure the Obama administration was under after the failed "underwear bomber" attack on December 25, 2009. "For the White House," Klaidman writes, "the psychic toll of Christmas Day was profound. Obama realized that if a failed terror attempt could suck up so much political oxygen, a successful attack would absolutely devastate his presidency. And much as he liked to talk about returning to first principles, Obama also had a powerful instinct for self-correction-as well as self-preservation."

The psychic aftershock of Christmas 2009 helped shape a lot of what followed: from body scanners at airports to ramped-up drone strikes to the lethal targeting of an American citizen.

But to Glennon's point, the administration was under pressure from the Trumanites well before that. In the 2012 book, The Obamians: The Struggle Inside the White House to Redefine American Power, James Mann describes a concerted effort by then-CIA director Michael Hayden and other senior intelligence officials to preserve business as usual by scaring the hell out of the incoming Obama team. Their private name for this scheme was the "Aw, Shit! Campaign."

The scare tactics worked. Klaidman reports that both Harold Koh, legal advisor at the State Department, and Jeh Johnson, the Pentagon's general counsel, used the same metaphor to describe the military pressure for more targeted killings: a runaway train. It was like "a massive freight train hurling down the tracks" Koh said. "You would have to throw yourself on the tracks to try to stop it," said Johnson.

All this helps shed light on Obama's strange and disorienting May 2013 "drone speech" at the National Defense University in Washington, D.C., in which the president seemed to be speaking not as commander in chief, but as his own loyal opposition.

In the speech, Obama said things like "Unless we discipline our thinking, our definitions, our actions, we may be drawn into more wars we don't need to fight, or continue to grant Presidents unbound powers." And: "The very precision of drone strikes…can also lead a president and his team to view [them] as a cure-all for terrorism." I remember thinking: "A president"? Which one? Anyone in particular? Who's in charge here, anyway?

National Security and Double Government suggests that the answer to that last question isn't quite so obvious, that the "most powerful man in the world" isn't nearly as powerful as he might appear.

It remains the case that Obama had the formal authority to say no to mass surveillance and perpetual war. But saying no would require resisting enormous bureaucratic and political pressure. And anybody willing to do what it takes to become president is unlikely to transform himself into a self-denying Cincinnatus once in office. Political survivors don't jump in front of trains.

Still, hope is a hard habit to break. Today, Rand Paul is the leading candidate for the role of "first civil libertarian president." Glennon's book gives us good reason to consider just how audacious such a hope might be.