In two recent posts I argued that it is useful to think of the national security establishment as a thoughtless organism prone to certain predictable behaviors such as self-preservation, expansion, and secrecy. But what are the policy implications, exactly, of that way of conceptualizing things?
To begin with, if policymakers and the public have a sophisticated understanding of what I called the “Sorcerer’s Apprentice” nature of any agencies we create or maintain, we’re more likely to put in place the right kinds of tight controls and limits to ensure that the agency doesn’t run amok. We’ll be more likely to anticipate the areas in which the agency will likely seek to expand its authorities, to exaggerate problems, to try poaching the domains of adjoining agencies, to escape oversight through secrecy, and so forth—and to craft laws and policies to counteract those tendencies.
Many people with long experience in Washington already understand these dynamics at one level or another—but many other people are very naïve about them. If an agency says there’s a “missile gap” with the Soviets, or warns about imminent “cyber-terrorism” leading to blood on the streets, or pushes for judges to be cut out of the equation when surveillance orders are generated, it does not occur to many people that these claims may reflect the natural tendency of government agencies to seek expanded powers, domains, and budgets. Many people, when they feel warm and fuzzy about a president, therefore feel warm and fuzzy about all the agencies that, they think, are under that president’s control—and when they hate and distrust a president, they likewise hate and distrust the agencies.
So a good understanding of the nature of the beast will help introduce a healthy skepticism towards many of the claims that government agencies make, and an understanding that they are bigger than the personalities of any of their leaders.
The dynamics I talked about probably apply to any bureaucracy, and as a result it is important to put in place realistic oversight and checks and balances whether we’re talking about the NSA or the Department of Housing and Urban Development. But when we’re dealing with the national security agencies, it’s doubly important that we put in place such controls. First, when mistakes are made in the national security arena, the stakes are higher: people die, human rights are violated, and Americans’ privacy and civil liberties are destroyed.
Second, the wall of secrecy behind which our national security agencies operate has the effect of magnifying all the problems that I have talked about. If bureaucracies are prone to do whatever it takes to increase their own scope and power, how much worse the problem is when they can selectively hide and release information in order to promote that agenda, manipulating the very broad secrecy powers they’ve been given in the name of protecting us all.
So oversight over national security-related agencies has to better, not worse than that imposed on the average bureaucracy. Yet we have seen just the opposite. Our security agencies have been granted vast deference, their self-serving claims about various threats and dangers all too often taken at face value—and they have been allowed to hide behind secrecy powers far more sweeping than necessary to serve our true national interest.
Finally, an understanding of the agencies’ natural impulse toward growth suggests that we ought to take a hard look at how much we are spending on these agencies, compare it to spending levels of the past—including during the period when we faced the threat of the Soviet Union and nuclear war—and ask ourselves whether their natural tendencies to constantly seek expansion have been given too much leeway.
A more sophisticated understanding of the nature of our national security agencies doesn’t mean merely that the NSA needs programs to combat “groupthink” or the diffusion of responsibility. It means that a fundamental restructuring of our security establishment is in order to impose the checks and balances that it needs but currently lacks.
(Third in a series)
Part four in the series is here.