According to this story, Vicki Divoll, former counsel to the Senate Select Committee on Intelligence, has been barred by SSCI from discussing in the media (specifically Talking Points Memo) certain non-classified information relating to the committee’s oversight of intelligence programs. Divoll gave an interview to TPM regarding the congressional role in intelligence oversight and submitted it to SSCI for review prior to publication, apparently not expecting that there would be any significant concerns. To her surprise: “[F]or the first time in her career, the committee took the extraordinary step, on a bipartisan basis, of declaring the interview’s entire contents a violation of her non-disclosure agreement and effectively forbade her from putting any of it on the record.”
Divoll and TPM present this as an arbitrary decision by SSCI to block public discussion of intelligence oversight. TPM says that the interview did not involve “classified sources and methods of intelligence gathering” but “general information about how the committee functions– and how it should function.” It says that “[a]mong the insights Divoll shared with us was the important role that staff can and should play in oversight of the executive branch’s intelligence activities.” Moreover, Divoll’s statements “tracked closely with information gleaned from other sources, and the public record.”
No doubt the committee has a different perspective on the matter. Still, given that Divoll left the employ of the committee 10 years ago and has frequently discussed matters related to her tenure at SSCI in the media since then, apparently without objection by the committee, this is a somewhat curious development. It raises the questions of what legal authority the committee has to block a former staffer from discussing matters of public interest, how broad that authority might be, and what arguments Divoll might have to challenge that authority. We will turn to those issues now.
Senate Rule XXIX(5). This rule provides:
Any Senator, officer or employee of the Senate who shall disclose the secret or confidential business or proceedings of the Senate, including the business and proceedings of the committees, subcommittees and offices of the Senate, shall be liable, if a Senator, to suffer expulsion from the body; and, if an officer or employee, to dismissal from the service of the Senate, and to punishment for contempt.
At the outset, it is unclear if this provision applies to former Senate employees at all. If it were not for the last clause, referring to “punishment for contempt,” it would seem the paragraph applies only to current staff because the only other remedy, “dismissal,” cannot be used for former employees. The reference to contempt, however, raises the possibility that a former employee could be punished by the Senate for a violation of the rule. Contempt might be used, for example, as an additional punishment for an employee dismissed for violating Rule XXIX(5), or as a means of punishing an employee who resigned after making an improper disclosure.
But while Rule XXIX(5) might be interpreted to authorize punishment for former employees who violated its strictures while still employed by the Senate, it does not, on its face, apply to disclosures made after Senate employment ends. The rule applies to disclosure by an “employee of the Senate,” not a “former employee of the Senate.” Nor is there any indication that the rule was intended to cover disclosures by former employees. Although the original rule dates back more than a century, it was not until 1992 that it was amended to include “employees” specifically. The impetus for this change, along with several others made to clarify and broaden the rule, was a report by a special counsel hired by the Senate to investigate leaking in connection with the Keating Five ethics case and the Clarence Thomas nomination hearings. The relevant sections of the report, which focuses on anonymous leaks related to matters pending before the Senate, do not address the issue of disclosures by former employees nor indicate any need to broaden the rule to cover former employees. Similarly, the explanatory floor statement by Majority Leader George Mitchell gives no indication that former employees are to be covered.
As illustrated by S. Res. 400 (discussed next), the Senate is capable of explicitly barring disclosures by former employees when it wishes to do so. In light of this fact and the potential for infringement on First Amendment rights, there is a strong argument against broadly construing Rule XXIX(5) to prohibit disclosures by former employees. (Even if the rule were construed to cover former employees, it would only prohibit Divoll’s interview to the extent that she disclosed “secret or confidential business or proceedings,” which is possible but not self-evident based on what has been publicly disclosed about the interview).
S. Res. 400. This resolution, which established SSCI in 1976, explicitly addresses the non-disclosure obligation of former SSCI staffers. Section 6 provides in part:
No employee of the Select Committee or any person engaged by contract or otherwise to perform services for or at the request of such committee shall be given access to any classified information by such committee unless such employee or person has (1) agreed in writing and under oath to be bound by the rules of the Senate (including the jurisdiction of the Select Committee on Ethics) and of such committee as to the security of such information during and after the period of his employment or contractual agreement with such committee; and (2) received an appropriate security clearance as determined by such committee in consultation with the Director of National Intelligence.
(emphasis added). Section 6 makes clear that a SSCI staff member must agree in writing to be bound by Senate and committee disclosure rules even after the staffer leaves the committee’s employ. However, this section is limited to the security of “classified information;” it says nothing about disclosures of non-classified information. Again, therefore, the applicability to Divoll’s situation is doubtful.
Even if Divoll’s disclosures do not violate Senate rules, they might still be prohibited under committee rules. SSCI has promulgated additional rules governing the disclosure of information. SSCI’s general rule-making authority, like that of other Senate committees, comes Senate Rule XXVI(2), which provides that “[e]ach committee shall adopt rules (not inconsistent with the Rules of the Senate) governing the procedure of such committee.”
SSCI Rule 10.5. This committee rule provides in part:
The members of the Committee staff shall not discuss either the substance or procedure of the work of the Committee with any person not a member of the Committee or the Committee staff for any purpose . . . either during their tenure as a member of the Committee staff or at any time thereafter except as directed by the Committee in accordance with Section 8 of S. Res. 400 of the 94th Congress and the provisions of these rules . . .
(emphasis added) This provision, applied literally, would certainly appear to bar Divoll’s discussions with TPM, in which she provided “general information about how the committee functions– and how it should function.” Yet this provision is so broad that it is difficult to imagine that it could be enforced. There are other SSCI rules which authorize the disclosure of certain classified or confidential information under certain circumstances, but even assuming that these rules override the general prohibition of SSCI Rule 10.5, which is by no means clear from the way the rules are written, they do not authorize the disclosure of non-confidential or public information. Therefore, SSCI Rule 10.5 would seem to bar any current or former staffer from discussing public information regarding the committee’s work, essentially a lifetime gag order.
Any attempt to enforce SSCI Rule 10.5, particularly against a former staffer, would raise serious constitutional issues. Perhaps this is why the committee did not rely on this rule in rejecting Divoll’s interview. Indeed, if SSCI Rule 10.5 applied, it would appear that there would be no way for the committee to approve Divoll’s request even if it wanted to, except perhaps by a formal committee vote under Section 8 of S. Res. 400.
Former SSCI Rule 9.6. During Divoll’s tenure, this rule prohibited both members and staff from disclosing classified and certain other information, such as executive session material, except as specifically authorized by that rule. But it is doubtful that this provision, which now exists in a different form in SSCI Rule 9.7, could apply to Divoll’s current situation. For one thing, unlike SSCI Rule 10.5, it does not expressly purport to apply to employees after they leave the committee. For another, it would appear the type of information Divoll disclosed to TPM may not fall within any of the categories covered by former SSCI Rule 9.6, although both ambiguity in the rule and uncertainty about the information disclosed make it impossible to say this with assurance. Finally, it should be noted that if former SSCI Rule 9.6 did apply, there would be no way for Divoll to have received permission to speak with TPM because authorizations for disclosure under that rule were limited to recipients with a need to know information for an official government purpose.
It is worth mentioning that Divoll was apparently fired from SSCI for violating SSCI Rule 9.6. This 2011 Politico article reports that Divoll disclosed to a Senate Judiciary Committee staffer the outcome of a vote on an amendment to the intelligence authorization bill. Because the vote was taken at an executive session of SSCI, such disclosure was prohibited, under SSCI Rule 9.6, unless it was authorized by the chair and vice-chair of the committee.
The article also raises questions about how consistently these non-disclosure rules are applied. According to a “former senior Congressional aide” (who might himself have been violating SSCI’s non-disclosure rules), the rules were not normally enforced in such a draconian fashion: “‘Nobody else . . . has ever taken the hit on that,’ said the aide, who spoke on condition of anonymity. ‘Particularly when it came to discuss[ing] this kind of unclassified information, there has been a longstanding issue [of how to talk] about these kinds of things or are these rules a gag rule? . . . Members have been challenged on this too.’”
SSCI Rules 10.6 and 9.3. The committee’s decision to reject Divoll’s interview rested in large part on the nondisclosure agreement Divoll had signed, pursuant to Section 6 of S. Res. 400 and SSCI Rule 10.6. The latter provides:
No member of the Committee staff shall be employed by the Committee unless and until a member of the Committee staff agrees in writing, as a condition of employment to abide by the conditions of the nondisclosure agreement promulgated by the Senate Select Committee on Intelligence, pursuant to Section 6 of S. Res. 400 of the 94th Congress, 2d Session, and to abide by the Committee’s code of conduct.
As noted previously, Section 6 of S. Res. 400 only applies to classified information. It appears, however, that SSCI’s nondisclosure agreement covers, or is interpreted by SSCI to cover, non-classified information as well. According to TPM, the committee contended that the information Divoll provided TPM violated the nondisclosure agreement because it was both “out of date” and “committee sensitive.” Exactly how the nondisclosure agreement bars disclosure of information that is “out of date” is a mystery, but it appears that the “committee sensitive” objection stems from current SSCI rules, which define and prohibit the disclosure of “committee sensitive” information.
Under SSCI Rule 9.3, “committee sensitive” information is defined as “information or material that pertains to the confidential business or proceedings of the Select Committee on Intelligence, within the meaning of paragraph 5 of Rule XXIX of the Standing Rules of the Senate, and is: (1) in the possession or under the control of the Committee; (2) discussed or presented in an executive session of the Committee; (3) the work product of a Committee member or staff member; (4) properly identified or marked by a Committee member or staff member who authored the document; or (5) designated as such by the Chairman and Vice Chairman (or by the Staff Director and Minority Staff Director acting on their behalf).”
While I have not seen the nondisclosure agreement, my surmise would be that the employee agrees to be bound by the Committee code of conduct, and this is interpreted by SSCI to mean that the employee must continue to comply, even after the end of her employment, with the restrictions on disclosing committee sensitive information. Assuming that Divoll’s interview focused on “general information about how the committee functions– and how it should function,” it is questionable whether she disclosed “committee sensitive” information within the meaning of SSCI Rule 9.3. For example, if she merely discussed publicly available information, it would be a stretch to say that she violated the rules, even though this information arguably “pertains” to the confidential business or proceedings of the committee. Her own views regarding the efficacy of the committee’s oversight of intelligence, even if rooted in her experience at SSCI, also would seem to be outside the scope of SSCI Rule 9.3 so long as she did not reveal any confidential information about the committee’s business or proceedings.
A key element of “committee sensitive” information in SSCI Rule 9.3 is that it may be “designated as such” by the chair and vice-chair of the committee. This does not supplant the requirement that the information must “pertain” to confidential business or proceedings of the committee, but if this requirement is read broadly enough, it could give the chair and vice-chair virtual carte blanche to prohibit any former SSCI staffer from speaking about any matter remotely related to the business of the committee. Perhaps such a provision would be upheld in court, but at least there are legal questions worth pondering. Do Senate rules give SSCI the authority to bar former staffers from disclosing non-classified information? If so, can SSCI expand the non-disclosure obligation beyond the disclosure of “confidential business or proceedings” within the meaning of Senate Rule XXIX(5)? Is the definition of “committee sensitive” information too broad and vague to survive constitutional muster? Does the First Amendment permit what amounts to a lifetime gag order preventing former SSCI staff speaking about the business of the committee? Can Divoll be required to adhere to a definition of “committee sensitive” information that did not exist when she served on the committee?
Even if SSCI’s position with regard to its own rules and Divoll’s nondisclosure agreement lacks merit, Divoll would have another reason not to proceed with her TPM interview. As a lawyer and member of the D.C. Bar, Divoll is subject to Rule 1.6 of the D.C. Bar Rules of Professional Conduct. D.C. Bar Rule 1.6(a)(1) prohibits a lawyer from “reveal[ing] a confidence or secret of the lawyer’s client.” A “secret” is not limited to confidential information, but includes any “information gained in the professional relationship that the client has requested by held inviolate, or the disclosure of which would be embarrassing, or would be likely to be detrimental, to the client.” Since it appears that Divoll’s interview would reveal information she gained in the course of representing SSCI (even if it were all public information), D.C. Bar Rule 1.6 would seem to prohibit her from proceeding with the interview without the client’s informed consent.
However, a recent decision of the Virginia Supreme Court, Hunter v. Virginia State Bar (Feb. 28, 2013) (hat tip: Jack Marshall), held that the Virginia Bar could not use this rule to punish an attorney who blogged about cases he had handled in court. The bar argued that the attorney violated Va. Rule 1.6 by discussing information, even though public, which might be detrimental or embarrassing to his client. The lawyer, represented by Rodney Smolla, a leading First Amendment expert (and former Clark law clerk!), argued that the bar’s “interpretation of Rule 1.6 is unconstitutional because the matters discussed in his blogs had previously been revealed in public judicial proceedings and, therefore, as concluded matters, were protected by the First Amendment.” Op. at 18. The court agreed with the lawyer and held that the blogging was constitutionally protected. Id. at 19-21.
Thus, to the extent that Divoll were prohibited by the legal ethics rules from speaking about information already in the public record, those rules may violate her First Amendment rights.
It would be rash to draw any firm conclusions about Divoll’s legal obligations without knowing the contents of her nondisclosure agreement and what she disclosed during her interview with TPM. However, to the extent that Divoll has been prohibited from discussing public information about the quality of, and potential improvements to, congressional oversight of intelligence, her First Amendment rights may have been violated. There are also questions about SSCI’s authority to impose, in effect, a broad lifetime gag order against former staff speaking about the business of the committee without permission from the chair and vice-chair.
Although the chances of SSCI actually taking legal action against Divoll for disclosing non-classified information are fairly remote, both ethics and prudence dictate that she follow her client’s interpretation of her nondisclosure obligations at this juncture. If the committee’s decision is really as arbitrary as Divoll seems to believe, however, she could consider challenging it in court.
Vicki Divoll, meet Rodney Smolla.
Even if SSCI is acting within its legal rights, this episode raises questions about whether “it is the fundamental policy of the Senate to favor openness and public access to information,” as the Senate declared in adopting S. Res. 363, which enacted the amendments to Senate Rule XXIX(5) in the 102d Congress. This would seem like a good time to enhance, rather than undermine, public confidence in congressional oversight of intelligence activities.
Maybe that’s just me. (Update: Maybe not).