INTRODUCTION I. WHAT Is DEEP SECRECY? PREVIOUS DEFINITIONS A. Academic Understandings B. Popular Understandings II. WHAT IS DEEP STATE SECRECY? A. How Many People Know B. What Sorts of People Know C. How Much They Know D. When They Know E. A New Definition III. THE PROBLEMATICS OF DEEP SECRECY, OR WHY RUMSFELD WAS RIGHT A. Utilitarian Theory 1. Secrecy and utilitarianism 2. Deep secrecy and utilitarianism B. Liberal Democratic Theory 1. Secrecy and liberal democracy 2. Deep secrecy and liberal democracy C. Constitutional Theory 1. Secrecy and the Constitution a. Text b. History c. Structure d. Prudence e. Doctrine 2. Deep secrecy and the Constitution a. Text b. History c. Structure d. Prudence e. Doctrine IV. PRACTICAL SOLUTIONS A. Second-Order Disclosure Requirements B. Congressional-Executive Consultation C. Internal Checks and Balances D. Bureaucratic Culture CONCLUSION INTRODUCTION
[A]s we know, there are known knowns; there are things' we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns--the ones we don't know we don't know. And if one looks throughout the history of our country and other free countries, it is the latter category that tend to be the difficult ones.
--Donald Rumsfeld (1)
In October 2001, the Attorney General reversed the Justice Department's presumption in favor of disclosure for Freedom of Information Act (FOIA) record requests.2 In March 2003, the President issued an executive order permitting the reclassification of previously declassified documents. (3) In fiscal year 2008, executive branch agencies reported 23,421,098 combined classification activities, breaking a record set the previous year. (4) Throughout the presidency of George W. Bush, his administration consistently withheld information from members of Congress and from investigative bodies, (5) subjected FOIA users to long delays, (6) and claimed the authority to refuse to comply with statutes and treaties in cryptic signing statements and in nonpublic opinions by the Office of Legal Counsel.7 Which, if any, of these acts of secrecy posed the greatest threat to democratic values and to good government? Which offended the Constitution? How would one even go about comparing and evaluating these secrets?
In this Article, I argue that Donald Rumsfeld's taxonomy holds an underappreciated answer to these questions. The statement quoted in the epigraph has already drawn a great deal of attention, but not for this reason. Almost from the moment of its utterance, commentators have ridiculed Rumsfeld for his kabbalistic logic and professorial cant. (8) The British Plain English Campaign called it "the most nonsensical remark made by a public figure" in memory. (9) Yet while Rumsfeld's statement was entirely, almost sublimely, unresponsive to the question he was asked--"is there any evidence to indicate that Iraq has attempted to or is willing to supply terrorists with weapons of mass destruction?"--it contained a valid insight. (10) There are things we know we know, things we know we do not know, and things we do not know we do not know. And, in fact, the secrets in the latter category tend to be the most difficult ones for a free society. They are deep secrets.
Their obscurity in the literature is surprising. The proper scope of government secrecy is a classic question in legal and political theory. In recent years, the question has taken on special urgency. New technologies and new adversaries have increased the potential costs of disclosing government-controlled information, (11) at the same time as the rise of online media, changing social norms, and a worldwide "transparency lobby" have increased the demand for it. (12) The Bush administration aggressively challenged the gains of the open-government movement and linked these efforts to an extreme vision of executive power. The use of state secrets appears both more pervasive in practice and more discredited in the public mind than at any point in history. Across many parts of the globe, there has never been such popular and academic interest in separating out legitimate secrecy from illegitimate secrecy.
In the sprawling literature on this topic, a significant aspect has been understudied: what we might call the structure of government secrets, as distinct from their function, their subject matter, or the competing interests in disclosure versus nondisclosure. Sometimes, outside parties are aware that a secret exists even though they are ignorant of its content. They know that Agency X is withholding photographs of detainee abuse, though they do not know what exactly the photographs show. (13) The photographs are a shallow secret. Other times, outside parties are unaware of a secret's existence; they are in the dark about the fact that they are being kept in the dark. They have no clue that Agency X possesses detainee-related evidence of any kind. The photographs are a deep secret. There is no bright line dividing deep from shallow secrets, but rather a continuum running from one extreme to the other. "Deep" and "shallow" are crude, but useful, signifiers for the positioning of a secret across multiple dimensions that together reflect the nature and degree of its concealment.
This Article aims to show that attending to the depth of state secrets can make a variety of conceptual and practical contributions to the debate on their usage. Sociologists and political scientists have distinguished between deep and shallow secrets for more than two decades now, and a number of legal scholars have applied these ideas to topics in private law. (14) Yet, hardly anyone has considered the distinction in detail, and even fewer have applied it to secrets of state. When we move from the private to the public sphere, the categories of "deep" and "shallow" become more problematic, and many borderline cases arise. The notion of depth nonetheless enriches our ability to describe, assess, and compare state secrets, both in the abstract and, once specific secrets have emerged, on a case-by-case basis. It helps us to think and speak more clearly about things we do not know.
Part I reviews the literature on deep and shallow secrecy. The scholarship is provocative and insightful, but there is not much of it. Part II identifies problems with the existing understandings of these concepts, and develops a new, more rigorous definition for secrets that are kept in a government context.
Part III moves from the positive to the normative and considers a wide range of arguments for and against deep state secrecy. While many substantial arguments can be made in defense of governments' keeping secrets, it turns out to be much harder to justify their use of deep secrets. Depth can be instrumentally valuable for the secret-keeper, but it is liable to generate perverse consequences, to exacerbate the costs of shallow secrecy, and to create a host of distinct harms. Rumsfeld was right: it is the unknown unknowns we have most reason to fear.
To reach this conclusion, Part III explores the role of state secrecy generally (as well as deep secrecy specifically) in utilitarian, liberal democratic, and constitutional theory. Some of this discussion synthesizes arguments many others have made; a good deal of it is new to the literature. My hope is that this Part can provide a useful framework for a debate that has often concentrated on discrete secrecy practices or discrete normative concerns to the exclusion of all others. The analysis of secrecy and the Constitution, in particular, marks an original contribution. Outside of a few areas--executive privilege, the First Amendment and the "right to know," access to judicial and administrative proceedings--very little has been written about secrecy's place in our constitutional order. The Article aims to spark conversation on this subject as well.
Part IV offers a series of practical solutions aimed at minimizing deep secrets in government while preserving adequate space for necessary shallow secrets. The Bush administration serves as a foil. One of the Article's key lessons is that the space between full public disclosure and maximal opacity is much larger than is commonly realized, and there are many ways to reduce the depth of state secrets without spilling their contents to the wider world. Even reforms that are purely internal to the executive branch can go a long way toward moderating depth, and thereby toward enhancing the deliberative quality and democratic legitimacy of our government's massive secrecy system.
WHAT IS DEEP SECRECY? PREVIOUS DEFINITIONS
The distinction between deep and shallow secrets has a distinguished but abbreviated pedigree in American legal theory. Beginning with Kim Lane Scheppele, scholars have applied the distinction in a variety of private law contexts, such as contracts and torts. Very rarely has anyone sought to refine Scheppele's framework--no one has challenged her original definitions--or to consider its relationship to public law. The concepts of deep and shallow secrecy are almost completely absent from the literature on government transparency.
Subpart A summarizes this line of scholarship. Although the phrase "deep secrecy" has several popular idiomatic meanings, as explained in Subpart B, it is the formal definition articulated by Scheppele that I will be considering and developing in this Article.
In a few pages from her 1988 book Legal Secrets: Equality and Efficiency in the Common Law, sociologist Kim Lane Scheppele introduced the concept of a deep secret. (15) Secrets, according to Scheppele, are items of information that one party, the "secret-keeper," intentionally conceals from another party, the "target." (16) Sometimes the target of a secret knows or suspects that information is being concealed from her, even though she does not know the content of the information. She has learned...