Skip to content. | Skip to navigation

Central Intelligence Agency
The Work of a Nation. The Center of Intelligence


CIA Home > Library > Center for the Study of Intelligence > Studies Archive Indexes > Documents > Creating a Statutory Inspector General at CIA

Creating a Statutory Inspector General at CIA


A Unique Vantage Point

Creating a Statutory Inspector General at the CIA

L. Britt Snider

The CIA has had an Inspector General (IG) since 1952. But it was only in 1989 that Congress enacted legislation creating an "independent" IG at CIA, appointed by the President with the advice and consent of the Senate.1 Before that, IGs were appointed by the Director of Central Intelligence (DCI).

"Statutory" IGs had been installed at most departments and agencies of the federal government pursuant to the Inspector General Act of 1978, but CIA was exempted from that particular law. The legislation that was enacted in 1989 was modeled after the 1978 Act, but in some respects the CIA IG was more constrained than the IGs of other agencies.2 While providing independence for the IG remained an essential ingredient, Congress did not want an IG at the CIA who was so "independent" that he or she might jeopardize the Agency's operations.

I happened to have been the General Counsel of the Senate Select Committee on Intelligence (SSCI) when this legislation was passed and had a key role in its development and ultimate enactment. Nine years later, after I had left the SSCI, I became the second person appointed under the statute. While Hill lore is rife with stories about congressional staffers who write laws creating jobs for themselves, let me assure the reader that this was the furthest thing from my mind at the time.

Nevertheless, having become the IG by quirk of fate, it seemed incumbent upon me to record what I recall about the peculiar way the legislation creating this position came about. There is an old adage that making law is a lot like making sausage--one does not want to know exactly how it is done. Knowing how this particular sausage was made, however, provides a better understanding of what Congress intended it to be.

Historical Background3

Creating an independent statutory IG at the CIA did not gain political momentum until after the Iran-contra scandal of 1986-87, although all the investigations of the CIA that took place in the mid-1970s had pointed out serious defects in the IG function.

In 1975, the Rockefeller Commission reported that the CIA IG had only five professionals on its staff, and that the scope of its work as well as its access to information was limited. The Commission recommended expanding the staff and improving its caliber, as well as requiring that its reports be provided to executive branch offices outside the CIA, such as the National Security Council.4

A year later, the Church committee similarly found that the IG had been denied access to vital information; that it had otherwise been prevented from doing its job; that its recommendations were being ignored; and that it was not making reports of suspected illegalities to the Attorney General. To address these problems, the committee recommended enactment of a statute that guaranteed the IG's right to Agency information and access to its operations. Congress would be notified if the IG were prohibited from looking into a particular matter. The committee also recommended illegalities be reported to the Attorney General and that the Congress be notified of such referrals. Annual reports from the IG to Congress were also recommended.5

The Pike committee went beyond its Senate counterpart by recommending legislation to create an Inspector General for Intelligence who would have jurisdiction over the entire Intelligence Community, including the CIA.6

None of these proposals was enacted into law at the time. The intelligence "charter" legislation introduced in the Senate in 1978 would have implemented a number of the Church committee recommendations pertaining to the IG, but the Carter administration ultimately withdrew its support for the bill, which then sank of its own weight.

The Iran-Contra Affair

When the CIA's role in the arms sales to Iran came to light in November 1986, a team of investigators from the CIA IG's office swung into action. Two months later, they produced a report based upon their review of Agency records and interviews with knowledgeable employees. While the report was accurate as far as it went, it paled in comparison to the information developed by the congressional and independent counsel investigations that followed.

In its final report, the combined House and Senate investigating committee took note, commenting that the CIA Inspector General "did not appear to have the manpower, resources, or tenacity to acquire key facts uncovered by the other investigations."7 While this criticism may have been unfair given the circumstances, it nonetheless led a majority to recommend that the CIA have an independent statutory IG like other agencies, appointed by the President and confirmed by the Senate.

Specter Takes Up the IG Issue

Following the Iran-contra affair, the SSCI leadership, Chairman David L. Boren and Vice Chairman William S. Cohen, was seized with correcting the systemic problems apparent from the Iran-Contra affair. Most of their attention was devoted to the problem of ensuring that the oversight committees were notified of covert actions "in a timely fashion," as required by existing law. Cohen, in fact, introduced legislation in the fall of 1987 to require that the President notify the committees, without exception, within 48 hours of approving a covert action. While Boren and Cohen were aware of the recommendation of the Iran-Contra committees urging the creation of a statutory IG at CIA, it was well down their list of priorities.

Senator Arlen Specter, on the minority side of the SSCI at the time, also decided to sponsor legislation to deal with the problems evident in the Iran-Contra issue. At the request of Specter's staff assistant, I put together an omnibus package of proposals in the fall of 1987 which Specter introduced as S. 1818. The last proposal in the bill called for the creation of an independent statutory IG at the CIA. The proposal was modeled after the Inspector General Act of 1978, but it contained important differences, which, in the interests of brevity, I will not go into here.

Boren and Cohen recognized that Specter's omnibus bill was substantive, and, having encouraged members to involve themselves in the committee's work, could hardly ignore it. At the same time, it was clear that no legislation would be reported by the committee that did not have the support of Boren and Cohen, and, despite the fact that they had signed the final report of the Iran-Contra committee, both remained cool to the idea of a statutory IG for the CIA. At this point, neither was willing to endorse Specter's IG proposal or co-sponsor his omnibus bill.

Boren, in particular, was concerned that an IG who was too independent could pose a threat to the Agency's operations. He also worried that an independent IG might play too powerful a role in the oversight framework, perhaps even competing with the congressional intelligence committees. He wondered whether the IG was the appropriate place for such power to reside.

The misgivings of the chairman were reinforced by the uncharacteristically hostile stance taken by DCI William Webster towards the Specter proposal. In private conversations with Boren and Cohen, Webster made clear his opposition to installing a political appointee in the IG position, someone who might not understand, or be willing to accommodate, the needs of the Agency and someone whom the DCI did not completely control. In November 1987, in an apparent effort to head off the legislation, Webster advised that he had created a senior steering group within the CIA to find ways of strengthening the IG function.

In the meantime, the SSCI held public hearings in late 1987 on the broader issues raised by the Iran-contra affair. Although Specter's bill ostensibly formed the basis for the hearing, the IG issue took a back seat to the weightier separation-of-powers issues confronting the committee. In light of this, Specter asked for, and was given, a commitment that the committee would hold a hearing on the IG proposal during the next session.

1988: The IG Issue Percolates

When the 100th Congress reconvened in January 1988, the SSCI members found a letter from DCI Webster waiting for them, advising of the "sweeping" changes he had made to the IG's office to correct the deficiencies noted by the Iran-Contra committee.

Boren announced at roughly the same time the creation of an "audit staff" within the existing SSCI staff to bolster the committee's capability to identify financial irregularities at the CIA and other intelligence agencies. Although Boren never said it in so many words, he saw the new committee audit staff as a counterweight to Specter's statutory IG.

Several months later, on 1 March 1988, the committee honored its commitment to Senator Specter by holding a public hearing on the IG provisions of his bill. Testimony was received from the Comptroller General, the State and DoD IGs, and, finally, DCI Webster.8 Few members showed up, however, and Boren signaled at the outset his ambivalence towards the proposal.

Webster tried to make the case that the Agency could have an effective IG without legislation. Indeed, he asked the SSCI for more time to allow the actions he had taken to strengthen the CIA IG bear fruit. More investigators would be hired, he said; more training would be required; and greater professionalism would be brought to the work. Indeed, Webster painted a rather grand vision of the IG, telling the SSCI that he saw the IG as more than a "watchdog," but rather a place where the top leaders and managers of the Agency would be cultivated. He said he intended to make a rotational assignment to the IG part of the career development of the Agency's best and brightest.9

A statutory IG, on the other hand, could only complicate things, said Webster, threatening his ability as DCI to protect intelligence sources and methods.10 He noted that potential sources might be reluctant to work with CIA for fear that an independent IG might disclose their identities. Foreign partners might worry their relationships with the Agency might be exposed. Specter responded by asking whether sensitive information had been compromised by the statutory IGs at State and Defense, but Webster demurred.

By the end of the hearing, it was clear to me that Boren had no intention of moving the Specter bill. Besides being unsure of its merits, he wanted to accommodate the DCI. At the same time, he wanted to give Specter something if only to discourage him from striking out on his own.

The solution we came up with was to include in the SSCI's markup of the FY 1989 intelligence authorization bill a requirement that the DCI make certain reports to the committees regarding the operation of the IG's office. The DCI would be required to file semiannual reports summarizing the activities of the IG for the preceding six-month period as well as one-time reports whenever the DCI selected or removed an IG, or whenever he prohibited the IG from undertaking an investigation, audit, or inspection, or otherwise hindered the IG in the execution of his duties.11

These reporting requirements did not entirely satisfy Senator Specter but kept him at bay. In the meantime, the SSCI staff, and in particular, Senator Specter's staff assistant, began to track the reports being issued by the CIA IG. In the following months, the SSCI obtained two such reports that were, on their face, wholly inadequate, regardless of which side of the IG debate one happened to be on. I recall that one of these reports, which had been specifically requested by the committee, had obvious lines of inquiry that had not been pursued; key witnesses who had not been interviewed; and analysis that was shoddy and unpersuasive. On being shown this particular report by his staff assistant, Senator Specter made a beeline for the Senate floor to express his outrage. I raced over there myself when I heard what was going on but arrived too late to hear the Senator's remarks.

The poor quality of these reports, though, had an impact on the committee beyond Senator Specter. For many on the staff, who had previously viewed the statutory IG debate largely in terms of principled arguments, now began to believe the CIA IG might lack the necessary professionalism.

Whatever change may have occurred in the staff's sentiments over the summer of 1988, the Senate committee's language requiring various DCI reports relating to the IG survived virtually intact when the two committees met in conference on the authorization bill later in the fall.

Senator Glenn Joins the Fray

Until this point, Senator John Glenn, who was a member of the SSCI as well as chairman of the Senate Governmental Affairs Committee (with jurisdiction over the Inspector General Act of 1978), had deferred to Boren's handling of the CIA IG issue.

In the fall of 1988, however, after the committee had acted on the Intelligence Authorization Act for FY 1989, Glenn's Governmental Affairs staff sent to the SSCI draft bill language that would place the CIA IG under the Inspector General Act of 1978. Glenn's staff indicated that he planned to introduce the bill in the next session or propose it as an amendment to the FY 1990 intelligence authorization. Among other things, the bill would have required the CIA IG to report to the governmental affairs committees in each body rather than the intelligence committees. A call to Glenn's staff confirmed that this was indeed what he intended.

This significantly changed the situation as far as Boren was concerned. Glenn was a good friend who Boren sought to accommodate if he could. But he could not agree with the CIA IG's reporting to other than the intelligence committees. Still, it was apparent that if Glenn were to follow through with his plans to introduce the bill or offer a floor amendment to the next authorization bill, it would pass in a nano-second. As far as Congress is concerned, oversight of the executive branch is motherhood. Oversight of the CIA is motherhood, apple pie, and the 4th of July wrapped into one. Senators would not be able to get on the bandwagon fast enough.

Given this development, Boren asked that I draft competing legislation for him to sponsor. The only specific guidance I received was that the bill should provide for the CIA IG to report to the intelligence committees, although I perceived from his comments months earlier that Boren would also want the bill to take into account his misgivings about an "all-powerful" IG.

I developed language for the bill in late fall 1988. While it contained a number of the elements I had put into Senator Specter's bill, such as presidential appointment and Senate confirmation, I took pains to ensure there were differences. At the same time, I knew it would be important to incorporate more features from the 1978 Act than the Specter bill to deflect criticism from the Governmental Affairs Committee.

I shared my draft with Glenn's committee staff, but not surprisingly, they preferred their own.

The 101st Congress

When the new Congress convened in January 1989, Senator Specter introduced a revised bill, S. 199, limited to the IG issue, and made it clear he planned to pursue the legislation in the next session. Given my own recent efforts to develop a competing bill for the chairman, Senator Specter's staff assistant worked with the Office of Senate Legislative Counsel (which provides drafting assistance to members) to come up with a somewhat different proposal than the one I had drafted earlier.

Glenn, on the other hand, chose not to introduce his bill when the new Congress convened, and for several months we waited to see how the issue would play out. Then, one afternoon in early spring 1989, I was summoned to Glenn's office to discuss the Boren draft. When I arrived, I found two chairs placed in front of his desk. I sat in one and his staff director on the Governmental Affairs committee sat in the other. There ensued a mini-debate between the two of us, going over the competing proposals point by point. Senator Glenn asked a lot of questions, and at the end of the discussion, to my astonishment, told the two of us he had decided to support the Boren alternative, provided Boren was serious about moving it. I assured him Boren was.

The SSCI Markup

With Glenn's support, Boren was now in a position to move his version of the IG legislation through the committee, although the possibility that the administration would oppose the bill (given DCI Webster's aversion to the whole idea) still loomed before us.

The committee's markup of the legislative items in the annual authorization bills was usually perfunctory, with the issues (if there were any) having been worked out in advance. At these sessions, as counsel, I would ordinarily describe the proposed legislative items to the committee, and the chairman would move their adoption as a package.

The markup of the IG legislation was different. Indeed, it stands out in my mind as the most unusual in my nine years with the committee. It occurred on a Thursday evening when the Senate had votes stacked up until about 2:00 a.m. The Senators had nothing to do between votes, so that is when the markup was held. The session was held in a secure room in the Capitol so members would not have far to walk when the votes were called. Because it was late in the evening, no one had any other place else to go. Virtually the entire committee was present. Rather than the committee's summarily adopting the proposal, I was asked to walk through each provision, and the members voted on them one by one.

A few of the bill's provisions, such as whether to give the CIA IG subpoena power, were hotly debated. (Ultimately, subpoena authority was voted down.) Some also wondered whether the concept of an independent IG really was appropriate for the CIA. While they were in the minority, their comments did force the proponents to come to grips intellectually with the fundamental issue. We adjourned about 1:00 a.m., and I remember walking back to the office tired but exhilarated. It was the way the founding fathers must have contemplated the Senate would operate, only nowadays there was never enough time.

On the Senate floor a few weeks later a spirited opposition was mounted to the IG proposal, led by Senator Fritz Hollings, but it was clear that with Boren and Cohen providing bipartisan support for the measure, opponents stood little chance of success.12 The IG legislation survived a motion to table by a vote of 64-34.

Facing a Possible Veto

Despite the convincing vote on the floor, Boren knew that, unless he could assure the House committee that the IG provision would not bring on a presidential veto of the bill, it would be difficult to get through conference.13 When the bill was on the Senate floor, President Bush as well as DCI Webster had sent letters to the Senate leadership opposing enactment of the IG provisions. But did opposition to the IG provisions necessarily translate into an administration veto of the authorization bill? The question seemed to hinge upon how seriously the Bush administration regarded the DCI's concerns.

To gauge this issue, Senator Boren reached out to Bob Gates, who was at that point deputy to National Security Adviser Brent Scowcroft. Boren and Gates had become good friends when Boren assumed the chair of the SSCI in January 1987, and one of his first duties had been to preside at Gates's confirmation hearings to be the DCI. Gates withdrew his nomination about a month into the process because he realized he was likely to remain under the scrutiny of the independent counsel for Iran-contra for some time to come. While Boren understood Gates's decision, he regretted not having the opportunity to work with him as DCI.

In August 1989, Gates ultimately provided Boren assurance that the President would not veto the authorization bill if the IG language were included. I do not know precisely how this came about. I was not privy to the discussions between Boren and Gates, nor was I privy to the conversations I later learned Boren had had with DCI Webster on the subject. I suspect General Scowcroft also had to have been involved, but I do not know this for a fact. Boren handled this on his own and told me only that he had assurances the administration would not veto the bill over the IG provision, and that I should work with DCI Webster to take care of certain specific problems he had with the language that had passed the Senate.

Those problems did not prove to be substantial. I was able to accommodate CIA's suggestions without doing damage to the bill, and, working with the HPSCI staff, was able to present an "agreed-on" version of the language to the conference committee. While House members can usually be counted upon to be cynical when it comes to approving additional positions for Senate confirmation (as a rule, they fail to see the "stature" bequeathed by the Senate's anointing lesser officials), they accepted the IG proposal with little discussion. By this point, it was a done deal.

CIA had not helped itself by causing the committees to doubt the efficacy of DCI Webster's actions to strengthen the IG, but, in the end, the statutory IG became law not because the intelligence committees were especially keen to impose this solution on the CIA, but rather to head off what was likely to be less desirable legislation. If this deal had not been struck, Boren would have faced proposals by Senators Glenn or Specter, or both, in the next Congress, weakened by his failure to have gotten his own bill through in the previous session.


In the ensuing weeks, the conference report went back through both houses without a hitch, and, on 30 November 1989, President Bush signed the IG provisions into law as part of the FY 1990 Intelligence Authorization Act. The signing statement issued by the White House was hardly a ringing endorsement, however, with the President stating that he expected the DCI to exercise his statutory authority, where needed, to protect sensitive intelligence information, presumably from the "out-of-control" IG he feared the law permitted.

Not surprisingly, moreover, the Bush administration was slow to appoint someone to fill the position. Almost a year went by before Frederick P. Hitz was sworn in as the first IG under the new law.

As it turned out, Hitz was to serve in that position for the next eight years, and, notwithstanding the Bush administration's earlier trepidation, not once during Hitz's tenure did any of the DCIs he served see the need to invoke the authorities provided by the 1989 law to curtail his activities.



(1) Section 17 of the CIA Act of 1949, 50 U.S.C. 403(q).

(2) I do not try to analyze the legislation in this article. Readers who are interested in comparing the CIA IG with other statutory IGs are referred to a Congressional Research Service Report, "Inspector General in the CIA Compared to Other Statutory Inspectors General," CRS 89-679, 21 December 1989, available at the Library of Congress.

(3) I am indebted in this short section to the research done by Frederick M. Kaiser of the Congressional Research Service, whose article entitled "The Watchers' Watchdog: The CIA Inspector General," published in the Spring 1989 edition of The International Journal of Intelligence and Counterintelligence, provided much of this information.

(4) "U.S. President's Commission on CIA Activities Within the United States," Report, (Washington, DC: GPO, 1975), pp. 88-89, 94.

(5) Church Committee, Book I, pp. 294-304, 394, 460-461; Book II, pp. 333-335.

(6) U.S. Congress, House of Representatives, Select Committee on Intelligence, Recommendations of the Final Report, House Report No. 94-833, 94th Congress, 2d Sess., GPO, Washington, 1976, p. 6.

(7) US Congress, House Select Committee to Investigate Covert Arms Transactions with Iran and the Senate Select Committee on Secret Military Assistance to Iran and the Nicaraguan Opposition, Report on the Iran-Contra Affair, House Report No. 100-433 and Senate Report No. 100-216, 100th Congress, 1st Sess., GPO, Washington, 1987, p. 425.

(8) Select Committee on Intelligence, United States Senate, "Hearings on S. 1818, to Establish an Independent Inspector General," S. Hrg.100-677, 100th Congress, 2d Sess., (Washington, DC: GPO, 1988).

(9) Ibid., pp. 45-47.

(10) Ibid., p. 48.

(11) See Section 504, Intelligence Authorization Act for Fiscal Year 1989.

(12) I recall Senator Fritz Hollings, who was leading the opposition to the bill, calling his staff aide over to his desk, and in a voice audible to most in the chamber, saying "What [have] you got me into, son?"

(13) The oversight committees in those days did everything they could to avoid a presidential veto of the annual authorization bill on the assumption that it would be well nigh impossible to obtain a two-thirds vote to override a veto of an intelligence bill. Without an annual authorization bill, the committees believed they would lose much of their leverage over the Intelligence Community.

L. Britt Snider is CIA's Inspector General.


Posted: May 08, 2007 08:59 AM
Last Updated: May 08, 2007 08:59 AM
Last Reviewed: May 08, 2007 08:59 AM