Where legislatures have created freedom of information laws, an expansive reading of government privacy privileges inevitably leads to a conflict in values.
In last year's Virginia General Assembly session, a commission was created to review
Virginia's Freedom of Information Act and to make recommendations to change it if necessary. The commission, made up of a handful of legislators, a media representative, and a representative of local-government interests, started holding hearings in June but is nowhere near its goal of making comprehensive recommendations for amending the statute. The commission will, in all likelihood, be renewed for another year in the next legislative session.
An issue before the commission was what many access advocates view as an unnecessarily heavy-handed approach by governments to various privileges related to legal advice. The two most prevalent classes are the attorney-client privilege, protecting discussions between client and counsel; and the attorney work product privilege, protecting the analytical, and sometimes factual, material generated during or in anticipation of litigation.
Virginia's Freedom of Information Act, which covers records and meetings, contains an exception to the rule of open meetings when a public body discusses pending or probable litigation with its attorney. Last summer, a Bedford County school board member indicated that she would like to discuss the possibility of having a random drug-testing policy at the high school. Without any indication that such a policy was going to be instituted, the board's attorney recommended a closed session discussion because he felt that due to the controversial nature of drug-testing, such a policy would inevitably lead to litigation. The local newspapers took the board to court, arguing that such a discussion had no real connection with probable litigation. The judge admonished the board for its decision, but, saying it was a close call, ruled that the board did not violate the law by holding the closed session.
Unfortunately, stories like this are common. Perhaps the most egregious example occurred several years ago when British Columbia's information and privacy commissioner, David Flaherty, ruled that a provincial board's discussions in the presence of its attorney were covered by attorney-client privilege only to the extent that they dealt with legal matters on which the board sought the attorney's advice. Interpreting a relatively new open government law, Flaherty correctly pointed out that the Legislature, in passing the law, had established a public policy that openness be the rule and secrecy the exception. To carry out the Legislature's will, Flaherty pointed out, the attorney-client privilege had to be read narrowly to protect only those kinds of legal matters for which the privilege was originally created.
The provincial Supreme Court saw it differently. Acting as if some sacred legal prerogative was in danger, the court ruled that Flaherty had significantly overstepped his bounds and that the attorney-client privilege covered any discussion between the board and the attorney, no matter the subject. The court's underlying message was that any board proceedings could be privileged as long as its attorney was present.
The ability to invoke the attorney-client privilege on the federal level has always been somewhat constrained by the lack of a consensus as to when an attorney is advising a client. In large institutions that employ many attorneys, there would be no records disclosed if the mere fact of their creation by an attorney was sufficient to invoke the privilege. As a result, attorney-client privilege is reasonably infrequent on the federal level, but in part that may be because the deliberative process privilege is so expansive that it can be invoked to protect much of what attorney-client privilege might be expected to cover.
Another federal problem is identifying exactly who the client is. While forcing presidential confidante Bruce Lindsey to testify before the grand jury may have wreaked havoc with the confidential relationship the president must have with his closest advisers, Lindsey was clearly not Clinton's personal attorney but employed by the government.
The reach of the attorney work product privilege is broader in the federal system. The privilege applies only to records created during or in anticipation of litigation. The Washington, D.C., Circuit previously had a local rule indicating that the privilege expired once all proceedings related to it were completed. As a result, the U.S. Court of Appeals for the District of Columbia Circuit told the Federal Trade Commission that it had to disclose closed litigation files in response to an FOIA request. The government appealed the decision to the Supreme Court, and in the 1983 decision FTC v. Grolier, the court ruled that the work product privilege protected the records in perpetuity. Ironically, although the government fought hard to establish this rule, it later changed its policy and presently encourages agencies to disclose such records unless an actual harm would occur as the result of disclosure.
There was a time when the work product privilege was distinguished by whether the records contained opinions or facts. Typically, opinions were protected since, in establishing the privilege in its landmark decision in Hickman v. Taylor, the Supreme Court had noted that it was intended to protect the "mental impressions" of the attorney. But facts were not privileged and were required to be disclosed. However, about a decade ago, the D.C. Circuit ruled that if the privilege applied, it applied to any records compiled by the attorney, regardless of whether they were opinions or facts. That ruling has not been followed uniformly by federal courts and there are later decisions of courts outside Washington which preserve the opinion-fact distinction.
What's troubling about the growth of these privileges is that courts have been willing to interpret them expansively. Such a willingness on the part of the courts is an invitation for agencies and other public bodies to apply the privileges more broadly with a certain confidence that the courts will give the privileges their broadest interpretation.
Where legislatures have created freedom of information laws, an expansive reading of the privileges inevitably leads to a conflict in values. Access statutes usually are built around a sliding scale of exemptions in the sense that a certain core of exceptions are virtually mandatory, such as, on the federal level, national security, trade secrets, and now, generally speaking, privacy. Others exemptions are discretionary, applied according to the judgment of the agency. Exemptions that encompass attorney-client and attorney work product privileges are generally discretionary, meaning the agency does not have to invoke them. The privileges should be invoked where there is an identifiable harm to agency functions from disclosure. But where the privilege may apply and there is no other reason to withhold except the application of the privilege, agencies should be strongly encouraged to disclose information.
Freedom of information laws create rights to government information. They do not set up requesters as supplicants at the hands of bureaucrats, begging for a stray morsel of information generously parceled out by government. Where right meets discretion, discretion should yield. Privileges are just that -- privileges. And where those privileges are abused, they may be taken away. In the context of open-government laws, the privileges have clearly been abused.
Harry Hammitt is editor/publisher of Access Reports, a newsletter published in Lynchburg, Va., covering open-government laws and information-policy issues.
Email | Web