OLC’s Evolving Position on Testimonial Immunity

In this post I will look at OLC’s claim that its advice on testimonial immunity of senior presidential advisers has been consistent “for nearly five decades.” See 5-20-19 OLC Opinion at 1. As we saw in my first post, since the 1940s the executive branch has generally resisted congressional demands for testimony from such advisers, but on a number of occasions it has permitted these advisers to testify in open congressional hearings and on other occasions it has agreed or offered to provide information from these advisers in alternative ways. Until the mid to late-1990s, the executive branch’s position on this subject was not presented to Congress as an assertion of absolute constitutional immunity, but more like the prophylactic rule described in my last post. Moreover, when OLC’s internal memoranda from this time period are scrutinized (to the extent they are available), they are compatible with this more modest interpretation of its position.

It was not until the Clinton administration that OLC articulated a formal and definitive defense of the proposition that senior presidential advisers are constitutionally immune from compelled congressional testimony. Even then, OLC seems to have accepted this proposition without any serious legal analysis and, in particular, without any consideration of important developments in the case law since Assistant Attorney General William Rehnquist first casually suggested it in 1971.

Communications to Congress. In 1996, the House Committee on Government Reform and Oversight issued a report on the findings of its investigation into the operations of the White House Communications Agency. See H.R. Rep. 104-748 (Aug. 2, 1996). During the course of its review, the committee requested the testimony of two senior White House officials. They declined to appear, and the White House counsel (Jack Quinn) sent a letter to the committee explaining that “it is a longstanding principle, rooted in the Constitutional separation of powers and the authority vested in the President by Article II of the Constitution, that White House officials generally do not testify before Congress, except in extraordinary circumstances not present here.” See Louis Fisher, The Politics of Executive Privilege 199 & n.1 (2004).

The committee majority’s response was dismissive, noting that “White House officials do testify before Congress on many different issues, and on a fairly frequent basis.” H.R. Rep. 104-748, at 13. This statement was correct, but it did not directly address the White House’s claim that such officials “generally” do not testify except in “extraordinary circumstances.”

The committee minority, on the other hand, supported the White House position and elaborated as follows:

The White House argued that under the Separation of Powers Doctrine, there is long-standing precedent of declining invitations to testify before congressional panels for White House officials who are involved in policymaking. However, exceptions have been made under extraordinary circumstances, involving issues of fact which only the requested witness can address. Conversely, objections are not raised where invitations to testify are issued to White House employees involved in “operations.”

This policy has evolved to the generally accepted principle that the testimony of White House employees involved in “operations” is open to formal congressional inquiry, while the testimony of an employee involved in “policy” is not subject to formal congressional inquiry. This principle has been recognized and applied by all previous administrations.

H.R. Rep. 104-748, at 23. To support its contention, the minority reproduced in full a 1992 letter sent by the George H. W. Bush White House responding to a request for testimony of senior White House officials on the administration’s policy toward Iraq prior to the first Gulf War. See id. at 23-24. The 1992 letter cited the “longstanding practice of the Executive Branch to decline requests for testimony by members of the President’s personal staff,” but stated that in light of “the unusual circumstances presented here,” the Bush administration would be “prepared to work with you to develop an alternative, mutually acceptable mechanism for making available to the Members of the Committee the White House officials whose testimony you have sought.” Letter for Jack Brooks, Chairman, Committee on the Judiciary, U.S. House of Representatives, from Nicholas E. Calio, Assistant to the President for Legislative Affairs (June 16, 1992).

It should be noted that the OLC opinion cites the 1992 Calio letter, but it omits its offer to provide information in light of the “unusual circumstances” and it does not mention the 1996 Quinn letter at all. See 5-20-19 OLC Opinion at 3 n.1 & 10. It should be further noted that neither the 1992 Calio letter nor the 1996 Quinn letter use the word “immunity” or claim in substance that “Congress may not constitutionally compel the testimony” of certain White House officials. Neither do any of the prior congressional communications cited by OLC contain such language. Instead, these communications refer to the executive branch “policy” or “practice” with regard to congressional requests for testimony of White House officials. They also acknowledge, explicitly or implicitly, that this policy or practice admits of exceptions for “extraordinary” or “unusual” circumstances.

The minority views expressed in H.R. Rep. 104-748 accurately summarize the practice and understanding between the branches up to that time. Congress tacitly accepted the executive branch’s policy with respect to testimony of White House officials by rarely requesting or insisting upon the testimony of such officials with regard to policymaking. On the other hand, the executive branch rarely objected or pressed its objection to testimony by White House officials who were involved in operational matters or who were fact witnesses with unique knowledge regarding the subject of the congressional inquiry. H.R. Rep. 104-748 at 23. This was particularly true in cases of alleged criminal misconduct or other serious wrongdoing. Id. at 23 & n.2.

In short, up until 1996 there is nothing in the public record or in the executive branch’s communications with Congress to indicate that OLC or the Justice Department generally claimed that senior presidential advisers were entitled to absolute testimonial immunity under the Constitution. Instead, the record reflects a prophylactic executive policy, “rooted in” constitutional principles and designed to protect confidential policymaking discussions between the president and his immediate advisers.

Rehnquist Memorandum. Although OLC’s confidential communications with executive branch clients prior to 1996 hint at a more forceful legal position than was communicated to Congress, they are still far less definitive than OLC portrays in its later opinions. OLC places its primary reliance on the aforementioned 1971 memorandum written by Assistant Attorney General Rehnquist (this is also where the clock starts for the “nearly five decades” of advice). See Memorandum for John D. Ehrlichman, Assistant to the President for Domestic Affairs, from William H. Rehnquist, Assistant Attorney General, Office of Legal Counsel, Re: Power of Congressional Committee to Compel Appearance or Testimony of “White House Staff” (Feb. 5, 1971) (“Rehnquist Memorandum”).

Specifically, OLC relies on the following language:

The President and his immediate advisers—that is, those who customarily meet with the President on a regular or frequent basis—should be deemed absolutely immune from testimonial compulsion by a congressional committee. They not only may not be examined with respect to their official duties, but they may not even be compelled to appear before a congressional committee. They are presumptively available to the President 24 hours a day, and the necessity of either accommodating a congressional committee or persuading a court to arrange a more convenient time, could impair that availability.

Rehnquist Memorandum at 7.

Although this language, quoted out of context, sounds like Rehnquist is offering a firm legal opinion, a reading of the full memorandum suggests otherwise. Most obviously, Rehnquist immediately prefaces the language in question by noting he is offering “generalizations” which are “necessarily tentative and sketchy.” OLC conveniently omits these qualifications, which show that at most Rehnquist was offering a preliminary view that required further analysis.

There is reason to doubt, moreover, that Rehnquist was stating even a preliminary legal opinion. His nine-page memorandum is rather conversational and informal in tone (perhaps explained by the fact that Ehrlichman was his law school classmate) and focuses as much on political considerations as legal ones. It contains only a very cursory review of legal principles and makes no claim that those principles support a conclusion of absolute immunity.

Rehnquist starts his legal discussion with this observation:

Everyone associated with the Executive Branch . . . has taken the position that the President himself is absolutely immune from subpoena by anyone, at the Seat of Government or away from it. This, of course, does not answer the question as to whether his immediate advisers are likewise exempt.

Rehnquist Memorandum at 3. Here Rehnquist acknowledges two fairly self-evident but important points: (1) any immunity to which the president’s advisers are entitled cannot be broader than the president’s own immunity and (2) the fact that the president enjoys an immunity by no means establishes that his advisers do as well.

Rehnquist then goes on to discuss some legal principles that might inform an immunity analysis. In judicial proceedings, he notes that the courts make a distinction between “a claim of absolute immunity from even being sworn from even being sworn as a witness,” on the one hand, and “a right to claim privilege in answering certain questions,” on the other. Rehnquist Memorandum at 4. The former is very rare, being limited to criminal defendants (and, he adds possibly in jest, newspaper reporters under a recent Ninth Circuit decision). Other privilege-holders are required to appear and assert the privilege on a question by question basis. Id.

Rehnquist also notes some considerations that might support the refusal of an executive branch witness to appear at a hearing. He explains that merely requiring the witness to appear places some burden on the executive “depending on the length of time the witness is expected to be present, the place the hearing is to be held, and the closeness of the relationship between the witness and the President.” Rehnquist Memorandum at 4. These considerations do not seem to be of the type that would give rise to a categorical doctrine of absolute immunity, and Rehnquist gives no indication that they do. Instead, he moves on to a discussion of historical practice (which he describes as “erratic” and “inconclusive”).

It is possible to read Rehnquist’s memorandum as tentatively suggesting it would be legally defensible to assert absolute immunity for senior presidential advisers, though he presents no argument or analysis to indicate this was either the legally correct view or the one that would prevail if challenged in court. A more persuasive interpretation, though, is that Rehnquist was proposing a “coherent set of defensible principles” the executive branch could advance in its negotiations with Congress. Rehnquist Memorandum at 8. This interpretation is consistent with his focus on the “tactical” aspects of executive-congressional information disputes. See id. at 6-7 (“[T]he Executive Branch has a headstart in any controversy with the Legislative Branch, since the Legislative Branch wants something the Executive Branch has, and therefore the initiative lies with the former. All the Executive has to do is maintain the status quo, and he prevails.”).

Furthermore, Rehnquist makes clear that the president must decide on a case-by-case basis whether a particular adviser should testify, and he warns against making ill-advised objections that are not sustainable. See Rehnquist Memorandum at 7-8 & n.4. As an example, he points to the 1944 case involving Jonathan Daniels (discussed in my prior post) where Congress’s threat to exercise its contempt power forced the administration to back down. Again, he is focused more on practical advice than legal analysis.

The more pragmatic interpretation of Rehnquist’s advice is further bolstered by congressional testimony he gave later that same year. In one hearing, Rehnquist acknowledged that from a legal standpoint a subpoenaed executive branch official would “be at least a compellable witness in the sense he would have to respond to a subpoena the same way a witness does who seeks to invoke the privilege against self-incrimination . . . .” U.S. Government Information Policies and Practices—The Pentagon Papers: Hearings Before a Subcomm. of the H.R. Comm. on Gov’t Operations 92d Cong. 385 (1971) (statement of William H. Rehnquist, Assistant Attorney General) (quoted in Emily Berman, Executive Privilege: A Legislative Remedy 46 & n.263 (Brennan Center for Justice 2009)). This testimony is consistent with an understanding of Rehnquist’s memorandum as setting forth a general set of principles for use in the political accommodation process between Congress and the executive branch. On the other hand, if the memorandum is read as claiming a constitutional immunity from appearing before Congress, it is fair to say Rehnquist “apparently recanted” this position in his congressional testimony. See Comm. on the Judiciary, U.S. House of Representatives v. Miers, 558 F.Supp.2d 53, __ (D.D.C. 2008).

Further reason to question OLC’s interpretation of the Rehnquist memorandum comes from another congressional hearing in 1971. At this hearing, Rehnquist directly addressed congressional concerns about senior White House officials (in particular, National Security Adviser Henry Kissinger) who had declined requests to testify before Congress. See Executive Privilege: The Withholding of Information by the Executive, Hearings Before the Subcomm. on Separation of Powers of the Senate Comm. on the Judiciary, 92d Cong. (Aug. 4, 1971) (Rehnquist Prepared Statement). Rehnquist did not claim that these advisers were constitutionally immune from testifying before Congress. Instead, he explained that “[t]here have been a number of instances in which Presidential advisers have failed to appear before Congress on the ground that the only information they could furnish resulted from conversations with, or advice given to, the President.” Rehnquist Prepared Statement at 26 (emphasis added). He gave as examples the refusals to appear of Steelman, Adams and Pierson discussed in my first post. Thus, he did not characterize these as cases where the presidential advisers were immune from appearing, but rather as cases in which they declined to appear because all of their potential testimony was privileged.

Contrary to any theory that presidential advisers enjoy absolute testimonial immunity, moreover, Rehnquist acknowledged that senior presidential advisers had properly testified in some instances, giving as specific examples Donald Dawson in 1951 and Sherman Adams in 1958. Rehnquist Prepared Statement at 27. Although he described these as cases in which the advisers testified about “their private affairs,” they are more accurately characterized as matters involving alleged misconduct in office.

Regardless of Rehnquist’s intent, his 1971 memorandum certainly did not constitute the formal articulation of an executive branch legal position. This is apparent not only from the memorandum itself, but from the fact that it was never communicated to Congress or anyone outside the executive branch. Indeed, to this day the memorandum has not been published by OLC.

Subsequent Internal OLC Memoranda. OLC cites 8 subsequent memoranda issued during the Nixon, Ford, Carter and Reagan administrations that supposedly endorse the “legal principle” of testimonial immunity for senior presidential advisers. See 5-20-19 OLC Opinion at 3 & n.1. These memoranda do not appear to have been shared with Congress or made public contemporaneously. Indeed, as far as I can tell, most of them are not available to this day. I have been able to locate only one of them, thanks to Professor Tuan Samahon, who obtained it in 2011 by FOIA request to OLC. See Memorandum for Margaret McKenna, Deputy Counsel to the President, from John M. Harmon, Assistant Attorney General, Office of Legal Counsel, Re: Dual-Purpose Presidential Advisers (Aug. 11, 1977). I will refer to this as the “McKenna-Harmon memorandum” to distinguish it from another memorandum (which I have not seen) written by Harmon in 1977 that OLC refers to as the “Harmon memorandum.”

An appendix to the McKenna-Harmon memorandum states “there is a general recognition that a Presidential assistant need not appear in response to an invitation or subpoena from a Congressional committee provided that the President directs him not to.” McKenna-Harmon Memorandum app. 1. It then goes on to explain:

The primary underpinning of the doctrine of executive privilege in this area is that frank and candid discussions between the President and his personal staff are essential to the effective discharge of the President’s executive responsibilities and that such Presidential discussions can take place only if their contents are kept confidential. While justifying the refusal of a Presidential assistant to answer Congressional requests for information relating to the discharge of the President’s executive responsibilities, this rationale, standing alone, does not appear to excuse Presidential assistants from appearing at all. However, a corollary of this rationale, based more on policy than on constitutional law, is that Presidential assistants need not appear, if so directed by the President, because all of their responsibilities are subject to a claim of privilege.

McKenna-Harmon Memorandum app. 2 (emphasis added).

This statement seems most consistent with the explanations offered by administrations to Congress during the decades following the Rehnquist memorandum. The assertion of a right to direct presidential advisers not to appear before Congress was in essence a prophylactic policy designed to protect executive privilege, not a claim that the Constitution provided such advisers absolute immunity from testimonial compulsion.

To be sure, the McKenna-Harmon memorandum refers to a “further rationale” founded on separation of powers that presidential aides may be considered the president’s “alter ego” and therefore entitled to the same testimonial immunity as the president himself. But it is not clear from the memorandum whether OLC was actually endorsing this rationale as a ground to refuse to provide presidential advisers as witnesses in cases where executive privilege would not protect their testimony. The McKenna-Harmon memorandum does not directly address this question and it certainly does not engage in any serious analysis of the question (e.g., it does not consider arguments that might be made against testimonial immunity).

Perhaps some of the other unpublished OLC memoranda contain such analysis but I tend to doubt it. What seems to have happened is that the executive branch developed a policy against allowing White House staff to testify before Congress (at least with respect to ordinary policy oversight), which Rehnquist rather casually and preliminarily endorsed (with respect to senior White House staff) in his 1971 memorandum. Subsequent OLC memoranda treated the combination of the executive branch practice and Rehnquist’s memorandum as if they represented a generally accepted legal principle. Because the executive branch’s public position was merely to assert a policy that was subject to negotiations with Congress, OLC apparently did not feel the need to revisit Rehnquist’s original “tentative and sketchy” view. The fact that its author was a supreme court justice and then chief justice of the United States may have played a role in that decision.

Conclusion. Beginning in the Clinton administration, OLC for the first time began to issue formal opinions that senior presidential advisers were constitutionally immune from testifying before Congress. See Assertion of Executive Privilege With Respect to Clemency Decision, 23 OLC 1 (Sept. 26, 1999); see also Immunity of the Counsel to the President from Compelled Congressional Testimony, 20 OLC 308 (Sept. 3, 1996). During the George W. Bush administration, OLC then opined that a former senior presidential adviser was likewise absolutely immune. See Immunity of the Former Counsel to the President from Compelled Congressional Testimony, 31 OLC 191 (July 10, 2007).

Unfortunately, these opinions consisted of little more than quoting the language from Rehnquist memorandum (without the “tentative and sketchy” qualification, of course), and citing later OLC memoranda as some sort of supporting authority. Somewhat astonishingly, these opinions failed to consider, or even acknowledge, important developments in the case law that occurred subsequent to 1971, including the Supreme Court’s decisions in United States v. Nixon, Harlow v. Fitzgerald and Clinton v. Jones.

This, it would turn out, was a mistake. See Miers, , 558 F. Supp.2d at __ (Court “not at all persuaded” by these OLC opinions, which “are for the most part conclusory and recursive” and fail to “cite[] to a single judicial opinion recognizing the asserted absolute immunity”). As we shall see in my final post, the unusual development of OLC’s legal position may explain why it is so surprisingly weak.