When an enemy can be anywhere, the state looks everywhere. So how can it infringe on privacy nowhere?
- By Emile SimpsonEmile Simpson is a research fellow at the Harvard Society of Fellows. He was formerly a British Army officer.
The Edward Snowden and NSA dragnet surveillance story that dominated headlines last week was framed in terms of a tension between national security and civil rights: Judge Richard Leon’s ruling that the extent of surveillance may be unconstitutional; new revelations about the extent of the agency’s spying on foreign governments and businesses; and the list of reform recommendations submitted by a blue-ribbon panel to President Barack Obama.
The other angle that’s been a thread throughout is one of technological determinism: the technology is there, says the government, so the NSA has in some ways been "on autopilot," as Secretary of State John Kerry put it in November. The technology should be limited, goes the other side of the argument, if indeed it can be now.
Both these perspectives are plainly legitimate, but also insufficient. While the NSA collects data against foreign intelligence threats generally, the type of data that effectively requires massive collection and storage of individual communications has been presented, for example by Gen. Keith Alexander, in terms of the need to connect the dots to stop another 9/11.
So the extent of NSA surveillance corresponds logically to the extent of the ill-defined successor to the war on terror: When the enemy can be anywhere, the state looks everywhere.
The boundary of the U.S. response to terrorism is the critical frame through which to view the vast impressionistic canvas upon which the NSA attempts to connect the dots: The root issue is not the balance between national security and civil liberties but defining the boundaries of the U.S. response to terrorism.
President Barack Obama’s speech at the National Defense University on May 23, his major national security speech of 2013, seemed to represent a shift in a different direction. "We must define the nature and scope of this struggle, or else it will define us," he intoned, "mindful of James Madison’s warning that ‘No nation could preserve its freedom in the midst of continual warfare.’"
It was accurate and ominous: Just days later, on June 5, the Edward Snowden story broke, revealing the massive footprint of the NSA’s electronic surveillance.
To which freedom was the president referring in his speech? The obvious answer would be an external freedom from attack by an enemy. But the Madison citation comes from his "Political Observations" of 1795. The sentence is the final line of a passage that discusses not freedom from external foes, but looks inwards to domestic freedom from oppressive executive government whose discretionary power is inflated in war: "Of all enemies to public liberty, war is perhaps the most to be dreaded."
Obama identified external and internal risks to freedom; both make sense. Indeed, we tend to think of the two in terms of a proportional relationship in times of war, that one is balanced against the other.
The problem with applying this balancing model in the context of the "struggle" to which the president referred is the fragmentation of the enemy as a clearly definable concept.
The core issue here is specificity. A clear and concise definition of the enemy both aligns the state’s resources against a clear target, and gives a reference point against which to weigh concerns over infringement of civil liberties and keep executive power in check.
Obama’s speech in May voiced an aspiration to specify more narrowly who the enemy is: "We must define our effort not as a boundless ‘global war on terror’ — but rather as a series of persistent, targeted efforts to dismantle specific networks of violent extremists that threaten America."
The speech was, however, inconsistent about how to define this effort. The concept of defeat, with its connotation of war, was rejected: "Neither I, nor any president, can promise the total defeat of terror" and subsequently relied upon: "First, we must finish the work of defeating al Qaeda and its associated forces." Confusing. In this struggle will there remain terrorists who threaten the US after al Qaeda and its franchises are defeated, or not?
The crux of the problem is that the concept of defeating an enemy is unpersuasive when the enemy has no clear bounds: Is the enemy al Qaeda? Al Qaeda franchises? "Terrorists" generally? Or any individual under the sun with jihadist sympathies who is potentially hostile to the United States, including U.S. citizens?
And what about people linked to the al Qaeda franchise to advance local interests but who do not harbor a desire to attack the United States — are they also threats to U.S. national security? Those groups may threaten U.S. allies, engaging national security more broadly.
Once we get to pg. 74 of the "Liberty and Security in a Changing World" report presented to President Obama this week, we read that the conception of national security — against which should be weighed the legislative checks and balances that regulate the NSA’s dragnet collection — goes beyond national security to include "foreign policy issues" too:
"These issues include the threats of international terrorism, the proliferation of weapons of mass destruction, cyber espionage and warfare, the risk of mass atrocities, and the international elements of organized crime and narcotics and human trafficking. They include as well the challenges associated with winding down the war in Afghanistan, profound and revolutionary change in the Middle East, and successfully managing our critically important relationships with China and Russia."
So should the NSA’s dragnet surveillance be premised on a wide concept of national security that extends to foreign policy generally, and the prevention of organized crime? Or should this aspect of NSA surveillance be strictly limited to counterterrorism? And even if limited to terrorism, what are the bounds of the terrorist enemy?
The key boundary is defined by the war, or struggle, which thus identifies the enemy.
Let us expand on Secretary of State Kerry’s defense of the NSA program in November, which does not speak to a clear distinction between war and peace: The agency’s present activities were "in many ways on autopilot," from a trajectory "going back to World War II and to the very difficult years of the Soviet Union and the Cold War and then of course 9/11, the attack on the United States, and the rise of radical extremism in the world… I assure you that innocent people are not being abused in this process, but there is an effort to try to gather information… In some cases, I acknowledge to you, as has the president, that some of these actions have reached too far."
The key association is between the language of innocence and the over-extension of conceptual boundaries that has allowed actions to "reach too far." Innocence indicates that the real targets are guilty; they are criminals. Yet it is also implicit in the secretary of state’s remarks that terrorists are simultaneously enemies to be defeated in war.
That goes not just in Kerry’s historical narrative, but in the formal U.S. legal comprehension of a military confrontation within the paradigm of war: The United States holds itself to be in a non-international armed conf
lict against al Qaeda, which engages the right to self-defense under Article 51 of the U.N. Charter under international law. Moreover, the 2001 Congressional Authorization for Use of Military Force expressly meets the requirements of the U.S. War Powers Resolution in domestic law.
What are the practical consequences of conflating the two concepts?
First, unlike a criminal, who is innocent until proven guilty, there is a presumption of wartime discrimination against the enemy, most obviously in his status as a legitimate target of lethal force. So there is no need to collect evidence against an enemy in the traditional sense. When one does, the intelligence effort expands exponentially.
Second, for the individual to be an enemy is a temporary status dependent on the phenomenon of armed conflict external to that individual. So when the armed conflict ends, the individual loses his enemy status; if he is a prisoner of war he is released. Conversely, criminal status is specific to the individual, and endures until punishment.
In practical terms, conflation of the combatant and criminal concept underpins the expansion of the struggle against terror: When the enemy becomes an identity that individuals can subscribe to, both outside and within the state, and holds a status that is then carried through life with the accompanying criminal designation, the boundaries of the conflict are vastly expanded in time and space.
In summary, there seem to me to be two fundamental questions that will calibrate the future extent of the NSA’s mass collection and storage of individual communications:
First, is this specific aspect of the NSA’s surveillance regime directed against terrorism, or foreign policy issues more broadly? If it is the latter, I do not see an end to mass individual surveillance.
Second, if terrorism is the exclusive justification, then what are the boundaries of the US struggle against terrorism? Until that is established, the ability of the U.S. government and courts to establish a stable balance between the two freedoms is seriously upset, as the state is effectively on a permanent war footing.
Take, for instance, this August 2013 opinion of the Foreign Intelligence Surveillance Court (FISA), which approves an FBI request for the NSA to collect bulk telephone data under Section 215 of the 2001 PATRIOT Act to protect against international terrorism.
The central part of the test, which the court applies to issue orders compelling the production of telephone data, is a requirement for "a statement of facts showing that there are reasonable grounds to believe that the tangible things sought are relevant" to the investigation.
However, the FISA Court makes clear that this test, used for foreign intelligence purposes, has a lower standard of review than comparable legislation used for criminal investigation purposes, for example the Stored Communication Act. The latter requires "specific and articulable facts" showing that the information sought is "relevant and material" to the criminal investigation; the former requires neither. The court expressly identifies that the 2001 PATRIOT Act removed the "specific and articulable" provision from section 215 that was a part of the previous pre-2001 incarnation of that provision, indicating that Congress positively intended to lower the standard of review.
The logical end point of the lack of specificity in America’s struggle against terrorism reveals itself in this deeply unsettling part of the opinion, which potentially includes U.S. citizens: "Accordingly, now the government need not provide specific and articulable facts, demonstrate any connection to a particular suspect, nor show materiality when requesting business records under Section 215."
Instead, the regime enforced by the FISA Court seeks to minimize the scope of NSA surveillance in various ways, such as a requirement for reasonable and articulable suspicion in the terms of the actual query to search the already collected bulk data (a self-monitoring regime), and the possibility of judicial review of the court’s orders. This regime before 2009 was held by a Foreign Intelligence and Security Court judge, Reggie Walton, to have "never functioned effectively." By September 2009, reforms were made.
Did they work? The widespread discomfort in the United States at the extent of NSA surveillance, as suggested in the open letter to the Obama administration sent last week by U.S. tech companies, does not suggest that this approach has encouraged public confidence; secret interpretation of public law rarely does.
What about the Fourth Amendment, brought before Congress by James Madison as part of the Bill of Rights of 1789, and partly informed by resistance to the widespread and indiscriminate search warrants the British used against their American colonists? The Fourth Amendment protects U.S. citizens from unreasonable searches and seizures, and has at its heart the concept of specificity: "No warrants shall issue, but upon probable cause, supported by oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized."
The FISA Court, however, holds that the Fourth Amendment was not engaged. First, because there is no legitimate expectation of privacy in telephone metadata once an individual has made a call, given the phone provider will already have the information. Second, because the collection in question was bulk, and because no one individual has a personal interest.
Doesn’t this mean then, that when an enemy can be anywhere, the state needs to look everywhere, and yet supposedly infringes privacy nowhere?
This would depend on which strand of the argument the FISA Court places more weight on, which is hard to discern from the opinion. The broader point, however, is that these are potentially the kind of logical dead-ends that one reaches when the struggle against terrorism is so ill defined; this is not a logic upon which a stable balance between external and internal freedoms can rest. Indeed, the Fourth Amendment issue is a major point of difference in Judge Richard Leon’s ruling.
It would be naïve to propose that the disaggregation of threats, the specification of enemies, and demarcation of combatant and criminal identity is an easy task. The real issue is how states deal with terrorist networks that configure themselves as franchise movements, to which individuals can subscribe by committing a criminal act which simultaneously triggers enemy status. More broadly, what about the individual who simply declares themselves an admirer, even vicariously through "liking" a pro-al Qaeda post on Facebook?
The political question is really about how far U.S. citizens expect the state to protect them individually from terrorism. Some aspects of the terrorist threat are matters of national security, even arguably an existential threat to the state: Consider the weapon of mass destruction in the city scenario. But most, clearly, are not. Such distinctions between security on the one hand and genuinely na
tional security on the other correlate to how much individual — as opposed to collective — risk the political community is prepared to tolerate. Ultimately, the boundary between individual and collective security can only be established by public debate and the democratic process, not secret courts.
In the final reckoning, this is a strategic as much as a civil rights issue. The contamination of conceptual boundaries between enemy and criminal ultimately subverts the utility of war as a political instrument. War, traditionally seen as the ultimate instrument of political decision, has a problem when it never ends: Military activity increasingly comes to merge with political activity, and politics does not end.
President Obama voiced an aspiration to move to a more bounded concept of the struggle against terrorism in May 2013, but the NSA revelations have exposed a substantial gap between words and reality; the inconsistencies in his speech are laid bare.
At his press conference last Friday, Obama stated an intention to deal with the issue by discussing how to implement his blue ribbon panel’s report. But that approach will not succeed in striking an adequate balance between national security and civil rights, as the report does not deal with the root issue of how to redefine the war on terror. As Madison wrote, no nation can preserve its freedom in the midst of continual warfare.
The U.S. national security strategy will be published in early 2014, the first since 2010. This is the president’s opportunity to follow through on the ostensibly compartmentalizing aspirations of his May speech. Let’s see what happens.