The question surrounding whether e-mail is an agency record subject to access laws and any record-keeping statutes was first addressed the day before George Bush's inauguration, when the Reagan administration was dragged to court to prevent it from deleting the contents of the e-mail system at the National Security Council. That NCS' e-mail system contained copies of the surreptitious messages between Oliver North and John Poindexter made the question that much more sexy.
But the result of the protracted litigation established that e-mail was a record and, on the federal level, was subject to the same kind of treatment paper records receive under the Federal Records Act. The National Archives issued e-mail policy, placing the federal government in the forefront in dealing with records management issues concerning such records. But, while the Archives' policy was a good model, it was not flawless, and a U.S. district court judge has now told the Archives that parts of the policy are contrary to established administrative law.
In his decision in Public Citizen v. John Carlin, U.S. District Court Judge Paul Friedman agreed with Public Citizen, which represented writer/
researcher Scott Armstrong and other similarly situated plaintiffs, that the Archives had taken an inappropriate view of the retention of e-mail records.
The case involved General Records Schedule 20, promulgated by the Archives in 1995. Under a General Records Schedule, categories of records possessing characteristics that are typical across agencies can be destroyed without getting permission from the Archives. Traditionally, records described in a GRS are administrative in nature that, as a rule, have been determined beforehand to have no historical or long-term value.
Friedman found the Archives had strayed from the congressional purpose for using General Records Schedules, noting that "the legislative history shows that both houses of Congress contemplated that General Records Schedules would be used only for 'routine "housekeeping records," such as those relating to the hiring of personnel, procurement of supplies, and fiscal management that are common to many agencies.'" He indicated that "given the legislative history of Section 3303a(d) of the Federal Records Act regarding the scope of General Records Schedules, it is not surprising that every general records schedule promulgated to date, except GRS 20, covers administrative housekeeping records common to all agencies."
But the Archives argued that the term "specified form or character" in section 3303a(d) should be construed to mean record of a common medium. Friedman rejected that reading, noting that "the government's interpretation of Section 3303a(d) not only contradicts the clear intent of Congress, but is irrational on its face." He continued, "The general schedules were designed to handle records that document housekeeping functions, procedures and transactions, such as personnel, maintenance or procurement, not unique aspects of a given agency. The routine nature of these kinds of records makes them both of a character common to multiple agencies, and also of lesser value. Thus, there is a relationship between the commonality of records covered by a general schedule and their diminished value."
This commonality was not present in GRS 20, Friedman noted. He observed that "the common feature of records scheduled under GRS 20 -- the fact that they have been generated by electronic technology -- has no relation to each record's value. No one would argue, for example, that a cable from the Secretary of State to an ambassador at the U.S. Embassy about an impending crisis or an electronic mail message written by the Secretary of State regarding the President's decision to declare war on another country has the same value as a GSA word-processing file regarding procurement of desks, simply because both records were created by electronic technology. Unlike the GSA record, the Secretary of State's cable or electronic mail message documents unique and important operations of government that may have historic value. The archivist should not use the same disposal criteria applied to the GSA procurement document on the secretary's message just because both records are found on computers."
Friedman explained that "the government surely would not argue that Section 3303a(d) gives the archivist authority to schedule the disposal of all handwritten records for disposal once their contents have been preserved in electronic or microfiche record-keeping systems, simply because handwritten documents are created by a method common to all agencies. A schedule that would include all handwritten documents would be inconsistent with the limited purpose of general schedules to provide an efficient method of disposing of records common to several or all agencies that do not possess long-term value. For the same reasons, it is inconsistent with the limited purpose of General Records Schedules to schedule all electronic records simply because they were created or stored on government computers. Simply put, Congress did not intend that records of such disparate value be lumped together under one disposition schedule. Such a method for disposing of records is inconsistent with the responsibility placed on the archivist to insure the protection and preservation of valuable government records."
The D.C. Circuit previously ruled in Armstrong v. Executive Office of the President that electronic records did not lose their value when printed off in paper form, since, at that time, there were elements of e-mail messages that did not appear when transformed to paper copies. While the archives had addressed these problems, Friedman noted that "electronic records often have a number of other unique and valuable features not found in the paper print-outs of the records." Some of these unique features were directly tied to the medium -- increased indexing and search capabilities, and the ability to disseminate the information through phone lines -- that could not be duplicated in paper form.
Friedman concluded that the archivist failed in his statutory duty to oversee the appropriate retention and destruction of federal records. He noted that "[the archivist] implicitly determined that no electronic records have sufficient administrative, legal, research or other value to warrant continued preservation once copies of such records are placed in paper, microfiche or electronic record-keeping systems. This is an irrational determination, and one that is necessarily premised on the illogical notion that a paper copy adequately preserves the value of an
electronic record. While, in some cases, paper copies may in fact adequately preserve the administrative, legal, research or historical value of an electronic record, there is no rational basis for the archivist's conclusion that a paper copy invariably adequately preserves such value in all cases and that electronic records never retain any administrative, legal, research or other value once such records have been copied to paper." He indicated that "the archivist has absolved both himself and the federal agencies he is supposed to oversee of their statutory duties to evaluate specific electronic records as to their value. The archivist has also given agencies carte blanche to destroy electronic versions without the archivist's approval when the agency believes they are no longer needed by the agency."
He added that "leaving to the agency the decision of `when' to discard records turns Section 3303a(d) on its head. Under GRS 20, an agency's decision to dispose of records determines the records' retention period. But Congress intended just the opposite -- that the lapse of a specified retention period, along with a determination by the archivist, in advance, that the records will not possess value in the future, would govern when records may be destroyed. The [Records Disposal Act] unequivocally requires the archivist's approval over any schedule submitted by, or imposed upon, a federal agency. The archivist's failure to set a specified retention period in GRS 20 essentially removes the archivist as a check on the destruction of electronic records."
Friedman was sympathetic to the government's concern that many agencies did not have the ability to store electronic records for long periods. But he noted that "the solution to this problem is not to absolve all agencies of using electronic record-keeping systems because some agencies do not now have the capability to do so. Rather, it makes much more sense to require each agency to put forward its best records management strategy, taking into account its particular assets and deficiencies." He pointed out that "for many years, Congress has recognized the importance of maintaining certain copies of government records and has established safeguards to insure that records with sufficient administrative, legal, research or other value are not destroyed. It should come as no surprise that electronic records, which take less storage space than paper records, should be subject to the same safeguards. While Congress has given the archivist an enormous task in managing the federal government's records, the court is confident that the archivist will be able to efficiently dispose of electronic records within the structure provided by existing law."
The Archives apparently took a wrong turn in developing GRS 20 and Friedman told the agency to get back on the right road. But what Friedman's decision emphasizes is that e-mail and other electronic records are different in kind than their paper counterparts and will need to be treated differently. The computer revolution has always been sold on the concept that computers would make our lives easier by allowing us to work more efficiently. The task set out for the National Archives is to figure out how computer records can best be retained in both the short- and long-term, and how those records that can be destroyed can be responsibly identified.
Harry Hammitt is editor/publisher of Access Reports, a newsletter published in Lynchburg, Va., covering open government laws and information policy issues. E-mail: <email@example.com>.
January Table of Contents