| 摘要 | The Freedom of Information Act (FOIA) is part of the basic weaponry of modern regulatory war, deployable against regulators and regulated alike. It differs, however, from other weaponry in the conflict, in that it is largely immune from arms limitation debate. Public discussion of the act displays a range of opinion extending from constructively-critical-but-respectful through admiring to enthralled. The media, of course, praise it lavishly, since they understandably like the “free information” it promises and provides. The Congress tends to agree with the media. The executive branch generally limits its criticism to relatively narrow or technical aspects—lest it seem to be committing the governmental equivalent of “taking the Fifth.” The regulated sector also wishes to demonstrate that it has nothing to hide, and is in any case torn between aversion to those features of the act that unreasonably compromise its interests and affection for those that unreasonably compromise the government’s. Through the mutually reinforcing praise of many who should know better, the act is paraded about with the veneration normally reserved for the First Amendment itself.
Little should be expected, then, of efforts now under way in both houses of Congress to revise the act. But however dim the prospect for fundamental change, the FOIA is worth examining, if only as an academic exercise. It is the Taj Mahal of the Doctrine of Unanticipated Consequences, the Sistine Chapel of Cost-Benefit Analysis Ignored.
Almost all of the Freedom of Information Act’s current problems are attributable not to the original legislation enacted in 1966, but to the 1974 amendments. The 1966 version was a relatively toothless beast, sometimes kicked about shamelessly by the agencies. They delayed responses to requests for documents, replied with arbitrary denials, and overclassified documents to take advantage of the “national security” exemption. The ‘74 amendments were meant to remedy these defects—but they went much further. They can in fact only be understood as the product of the extraordinary era that produced them—when “public interest law,” “consumerism,” and “investigative journalism” were at their zenith, public trust in the government at its nadir, and the executive branch and Congress functioning more like two separate governments than two branches of the same. The amendments were drawn and debated in committee while President Nixon was in the final agony of Watergate, and were passed when President Ford was in the precarious early days of his unelected term. The executive branch managed to make a bad situation worse, by adamantly resisting virtually all changes in the act, even those that Congress was obviously bent on achieving. By the time it realized the error of its obstinacy, it was too late: the changes had been drafted and negotiated among congressmen and committees without the degree of agency participation and advice that might have made the final product—while still unpalatable—at least more realistic. The extent of the disaster may be gauged by the fact that, barely two months after taking office as a result of the Watergate coverup, President Ford felt he had to veto a bill that proclaimed “Freedom of Information” in its title. It passed easily over his veto.
The Freedom of Information Act is the Taj Mahal of the Doctrine of Unanticipated Consequences, the Sistine Chapel of Cost-Benefit Analysis Ignored.
When one compares what the Freedom of Information Act was in contemplation with what it has turned out to be in reality, it is apparent that something went wrong. The act and its amendments were promoted as a means of finding out about the operations of government; they have been used largely as a means of obtaining data in the government’s hands concerning private institutions. They were promoted as a boon to the press, the public interest group, the little guy; they have been used most frequently by corporate lawyers. They were promoted as a minimal imposition on the operations of government; they have greatly burdened investigative agencies and the courts. The House Committee Report estimated that the 1974 amendments would cost only $100,000 a year; a single request under them has cost more than $400,000. There has grown up, since 1974, an entire industry and profession based upon the Freedom of Information Act. An organization has been formed, the American Society of Access Professionals, composed of men and women (mostly government employees) who have made their careers in this field. A two-volume FOIA loose-leaf service, updated monthly, retails at $438 a year; another one, supplemented only semiannually, is somewhat cheaper. Every week the Legal Times of Washington runs a page or more of notable new FOIA filings—mostly to enable corporate lawyers to find out what it is that other corporate lawyers are trying to find out. The necessary training for any big-time litigating lawyer now includes not only the cross-examination of witnesses, but use of the Freedom of Information Act. In short, it is a far cry from John Q. Public finding out how his government works.
The act and its amendments … were promoted as a boon to the press, the public interest group, the little guy; they have been used most frequently by corporate lawyers. … It is a far cry from John Q. Public finding out how his government works.
What happened in the 1974 amendments to the Freedom of Information Act is similar to what happened in much of the regulatory legislation and rulemaking of that era: an entirely desirable objective was pursued singlemindedly to the exclusion of equally valid competing interests. In the currently favored terminology, a lack of cost-benefit analysis; in more commonsensical terms, a loss of all sense of proportion.
Take, for example, the matter of costs. As noted above, the 1974 amendments were estimated by Congress to cost $100,000 a year. They have in fact cost many millions of dollars—no one knows precisely how much. The main reason is that the amendments forbid the government from charging the requester for the so-called processing costs. Responding to a request generally requires three steps: (1) searching for the requested documents; (2) reviewing or “processing” them to determine whether any of the material they contain is exempt from disclosure, to decide whether the exemption should be asserted, and, if so, to make the line-by-line deletions; and (3) duplicating them. Before 1974, the cost for all of this work was chargeable to the requester; since 1974, step two has been at the government’s expense. In many cases, it is the most costly part of the process, often requiring the personal attention of high-level personnel for long periods of time. If, for example, material in an investigative file is requested, someone familiar with the investigation must go through the material line by line to delete those portions, and only those portions, that would disclose a confidential source or come within one of the other specific exceptions to the requirement of disclosure. More-ever, even steps one and three are at the government’s expense “where the agency determines that waiver or reduction of the fee is in the public interest because furnishing the information can be considered as primarily benefiting the general public.” Even where the agency parsimoniously refuses to grant this waiver, the more generous judiciary sometimes mandates it—which happened, for example, in the case of the FOIA request by the Rosenberg children.
The question, of course, is whether this public expense is worth it, bearing in mind that the FOIA requester is not required to have any particular “need to know.” The inquiry that creates this expense—perhaps for hundreds of thousands of documents—may be motivated by no more than idle curiosity. The “free lunch” aspect of the FOIA is significant not only because it takes money from the Treasury that could be better spent elsewhere, but also because it brings into the system requests that are not really important enough to be there, crowding the genuinely desirable ones to the end of the line. In the absence of any “need to know” requirement, price is the only device available for rationing these governmental services—and in many cases a price based on search and re-production costs is simply not adequate.
Other features of the amendments reflect the same unthinking extravagance and disregard of competing priorities. Although federal agencies carry out a great many important activities, rarely does the law impose a specific deadline for agency action. Yet the FOIA requester is entitled by law to get an answer to his request within ten working days—and, if it is denied, to get a ruling on his appeal within another twenty. (There is a provision for an additional ten days “in unusual circumstances.”) So the investigative agent who is needed to review a file must lay aside his other work and undertake that task as his top priority.
It is also rare that a federal official must be subjected to a disciplinary investigation—even for malicious baby-snatching under color of law, much less mere negligence. But if he should happen to trifle with an FOIA request, stand back! In a provision unique in the United States Code, the 1974 amendments specify that whenever a court considering an appeal from an FOIA denial
issues a written finding that the circumstances … raise questions whether agency personnel acted arbitrarily or capriciously with respect to the withholding, the Civil Service Commission [now the Special Counsel of the Merit Systems Protection Board] shall promptly initiate a proceeding to determine whether disciplinary action is warranted against the officer or employee who was primarily responsible for the withholding.
In the courts, the statute provides that FOIA appeals shall “take precedence on the docket over all cases and shall be assigned for hearing and trial or for argument at the earliest practicable date and expedited in every way.” (There is an exception to this preferential treatment for “cases the court considers of greater importance.”) And if the requester taking the agency to court “has substantially prevailed,” the court is authorized to make the government pay his attorney fees and litigation costs. One would have thought it infinitely more important to pay the attorney fees and litigation costs of persons who are erroneously or even frivolously prosecuted by the government—but of course the law makes no provision for such payment.
The preferred status of the FOIA requester in the courts is also evident in the standard of review. If a federal agency assesses a penalty against you or revokes a certificate that is necessary for your livelihood, it will do you no good to persuade a judge that the agency is probably wrong. The courts cannot reverse the agency merely because they disagree with its assessment of the facts. They can do so only when there is a lack of “substantial evidence” to support its finding. If, however, an agency denies a freedom of information request, shazam!—the full force of the Third Branch of government is summoned to the wronged party’s assistance. The denial is subject to de novo review—which means that the court will examine the records on its own and come to its own independent decision. And whereas the general rule is that the citizen appealing to the courts must show that the agency acted improperly, in the case of an FOIA denial “the burden is on the agency to sustain its action.”
The foregoing defects (and others could be added) might not be defects in the best of all possible worlds. They are foolish extravagances only because we do not have an unlimited amount of federal money to spend, an unlimited number of agency employees to assign, an unlimited number of judges to hear and decide cases. We must, alas, set some priorities—and unless the world is mad the usual Freedom of Information Act request should not be high on the list.
Some other effects of the 1974 amendments, however, would be malignant even in a world without shortages. Prominent in this category is the provision which requires the courts to determine (again de novo) the propriety of classification of documents on the grounds of national security or foreign affairs. What is needful for our national defense and what will impair the conduct of our foreign affairs are questions of the sort that the courts will avoid—on the basis of the “political question” doctrine—even when they arise in the context of the most significant civil and criminal litigation. Imagine pushing the courts into such inquiries for the purpose of ruling on an FOIA request! This disposition appears even more incredible if one compares it with the provisions of the Foreign Intelligence Surveillance Act. There, for the much more compelling purpose of determining whether secret electronic surveillance will be allowed, the court must accept the certification of a high-level executive official that the information sought is necessary to the national defense or the conduct of foreign affairs unless, on the basis of the accompanying data, that certification is “clearly erroneous.”
But the most ironic absolute defect of the ‘74 amendments was perhaps unintended at the time and seems to have gone virtually unnoticed since. The amendments have significantly reduced the privacy, and hence the autonomy, of all our nongovernmental institutions—corporations, labor unions, universities, churches, political and social clubs—all those private associations that form, as Tocqueville observed, diverse centers of power apart from what would otherwise be the all-powerful democratic state. Some of the activities of these associations should be open to public scrutiny, and prior to 1974 Congress made that judgment on a relatively specific basis, in enacting such disclosure statutes as the Securities Exchange Act, the Public Utility Holding Company Act, and the Labor-Management Reporting and Disclosure Act. Of course, in addition to those particular activities of private institutions that require publication, virtually all activities of private institutions may be subjected to governmental investigation—and increasingly are, to ensure compliance with the innumerable requirements of federal laws and regulations. By and large, it has been left to the agencies to determine when investigation is appropriate, and the courts have been most liberal in sustaining investigative authority.
The effect of the 1974 amendments to the Freedom of Information Act was to eliminate the distinction between investigation and publication. The “investigative files” exemption in the original act was narrowed so as to permit withholding of documents acquired or produced in a law enforcement investigation only if disclosure would cause specific damage to the investigative process or to particular private interests (for example, reveal the identity of a confidential source). The way things now work, the government may obtain almost anything in the course of an investigation; and once the investigation is completed the public (or, more specifically, the opponents or competitors of the investigated institution) may obtain all that the investigative file contains, unless one of a few narrow exemptions applies. There is an exemption (though the agency has discretion not to invoke it) for confidential commercial information. But there is none that protects an institution’s consultative and deliberative processes—the minutes of a university’s faculty meetings, for example. It is noteworthy that internal consultation and advice within the government itself is exempted from disclosure since, as the 1966 House Committee Report explained, “a full and frank exchange of opinions would be impossible if all internal communications were made public.” But no such exemption exists for the internal communications of private organizations that come into the government’s hands. In other words, legislation that was supposed to lay bare the workings of government is in fact more protective of the privacy needs of government than of private institutions.
In other words, legislation that was supposed to lay bare the workings of government is in fact more protective of the privacy needs of government than of private institutions.
There seems little hope, however, that these absolute defects of the Freedom of Information Act, much less its mere extravagances, will be corrected. And once the fundamentally flawed premises of the ‘74 amendments are accepted, as they have been, all efforts at even minor reform take on an Alice-in-Wonderland air. For example: The government is concerned about use of the Freedom of Information Act as a weapon in litigation. Requests by a litigant for judicially compelled production of documents from the opposing party’s files (so-called discovery requests) can be kept within reasonable bounds by the court itself. But when the government is the adversary, there no longer is any need to use the judicial discovery mechanism. An FOIA request can be as wide as the great outdoors; and the government must produce the information within ten working days—or, as a practical matter, within such longer period as the requester is willing to negotiate. It is not only a good way to get scads of useful information; it is also a means of keeping the government’s litigation team busy reviewing carloads of documents instead of tending to the trial of the case. The story is told of a criminal defense lawyer who negotiated a favorable plea for his client by filing an onerous FOIA request that would have taken weeks of the U.S. attorney’s time. And why not? Anyone can file such a request, and surely the attorney is obliged to use all lawful means to serve the interest of his client.
… once the fundamentally flawed premises of the ‘74 amendments are accepted, as they have been, all efforts at even minor reform take on an Alice-in-Wonderland air.
Well, the government’s proposed solution for this problem is to forbid FOIA requests by litigants once litigation has commenced. Apart from the practical difficulty of enforcing such a ban, consider the Mad Hatter result it would produce: Absolutely anybody in the world (the FOIA requester does not, by the way, have to be a U.S. citizen) would be able to put the government through the inordinate trouble and expense of the FOIA process except—you guessed it—the person most legitimately interested in the requested information.
The defects of the Freedom of Information Act cannot be cured as long as we are dominated by the obsession that gave them birth—that the first line of defense against an arbitrary executive is do-it-yourself oversight by the public and its surrogate, the press. On that assumption, the FOIA’s excesses are not defects at all, but merely the necessary price for our freedoms. It is a romantic notion, but the facts simply do not bear it out. The major exposés of recent times, from CIA mail openings to Watergate to the FBI COINTELPRO operations, owe virtually nothing to the FOIA but are primarily the product of the institutionalized checks and balances within our system of representative democracy. This is not to say that public access to government information has no useful role—only that it is not the ultimate guarantee of responsible government, justifying the sweeping aside of all other public and private interests at the mere invocation of the magical words “freedom of information.”
The act’s defects cannot be cured as long as we are dominated by the obsession that gave them birth—that the first line of defense against an arbitrary executive is do-it-yourself oversight by the public and its surrogate, the press.
It is possible to save the desirable features of the FOIA—and even to give it teeth it did not have before 1974—without going to absurd extremes. But don’t hold your breath. As the legislative debate is now shaping up, a few minor though worthwhile changes may be made, such as exemption of CIA case files. But the basically unsound judgments of the ‘74 amendments are probably part of the permanent legacy of Watergate. We need not, however, admire the emperor’s clothes. ■
Antonin Scalia, the editor of Regulation, is professor of law at the University of Chicago.
|
