from the FAS Project on Government Secrecy
Volume 2010, Issue No. 55
July 6, 2010

Secrecy News Blog:


The Obama White House has threatened to veto a pending intelligence bill if it includes a provision that would authorize the Government Accountability Office (GAO) to perform audits of intelligence programs at the request of Congress. But a Department of Defense Directive issued last week explicitly allows for GAO access to highly classified special access programs, including intelligence programs, under certain conditions.

The newly revised DoD Directive 5205.07 on special access programs (SAPs) states that: "General [sic] Accountability Office (GAO) personnel shall be granted SAP access if: a. The Director, DoD SAPCO [SAP Central Office], concurs after consultation with the chair and ranking minority member of a defense or intelligence committee [and] b. The GAO nominee has the appropriate security clearance level."

In other words, the Pentagon's new directive permits what the Obama Administration is stubbornly striving to prevent, namely a role for the GAO in intelligence oversight.

DoD special access programs are the most tightly secured of all the Pentagon's classified programs. They include activities within three classified domains: intelligence, acquisition, and operations.

The previous version of the same DoD Directive on special access programs, which was issued in 2006 and revised in 2008, made no mention of the GAO. However, a 2009 DoD Instruction stated that classified DoD information on intelligence and counterintelligence "may be furnished to GAO representatives having a legitimate need to know." ("DoD Should Not 'Categorically" Deny GAO Access to Intelligence," Secrecy News, February 4, 2009.)

As an historical matter, GAO has long had access to classified DoD programs of the highest sensitivity, and has produced numerous reports on special access programs, including many in unclassified form. But the CIA and other non-DoD intelligence agencies have resisted GAO oversight.

"In practice, defense [intelligence] agencies do not adopt the 'hard line' CIA approach but generally seek to cooperate with GAO representatives," the late Stanley Moskowitz of the CIA wrote in a 1994 memorandum for the Director of Central Intelligence.

Most recently, the Senate Intelligence Committee reportedly said it would yield to the White House and would renounce any right to use the GAO in its oversight activities. But House Speaker Nancy Pelosi rejected that concession, and she has been insisting that a role for GAO in intelligence oversight must be recognized by the Administration. To a significant extent, considering the dominance of defense intelligence agencies within the intelligence community, one could say that it now has been so recognized. Only the details remain to be negotiated.


The intersection of science and national security in the 20th century produced many peculiar phenomena, some of which are illuminated in a new biography of physicist Hugh Everett III (1930-1982).

Everett is best known, if at all, as the originator of the "many worlds" interpretation of quantum mechanics. Roughly, this theory holds that whenever a quantum state is measured, every possible outcome of the measurement is realized in a different universe, which somehow means that the world is constantly branching out into a multiplicity of separate, parallel worlds.

"Many prominent physicists, including [Richard] Feynman, thought many worlds was a ludicrous idea," notes Peter Byrne, the author of the new Everett biography. But others were startled and impressed by its mathematical and conceptual coherence. (The notion of parallel universes also became a staple of science fiction writing.)

Having made his controversial mark in physics, Everett went on to perform nuclear war planning for the top secret Weapons Systems Evaluation Group, a sort of Pentagon think tank which generated analysis for the Joint Chiefs of Staff.

"When Everett joined WSEG in 1956, it was analyzing the cost-benefits of global and limited nuclear warfare," writes Mr. Byrne. "Ongoing studies examined nuclear blast and fallout kill ratios; the impact of jamming the electronics of guided missiles and airplanes; and the disturbing problem of the 'nuclear blackout,' i.e. massive electrical disturbances unleashed by nuclear bombs exploded high in the atmosphere." The WSEG's "Report 50," portions of which have recently been declassified along with other WSEG records, "became a basic source document" for nuclear war planning.

Everett apparently wrote the software for the first Single Integrated Operating Plan (SIOP), the U.S. nuclear warfighting plan. He also designed the first "relational database" software, an early word-processing program, and more.

The new biography of Hugh Everett never veers off into hagiography, a temptation that might have been easy to resist since Everett had more than his fair share of shortcomings. But investigative reporter Peter Byrne has produced a thoughtful account of an original figure and his diverse contributions to a momentous period in the history of science and national security. One imagines that there are still many other such stories waiting to be told.

See "The Many Worlds of Hugh Everett III" by Peter Byrne, Oxford University Press, 2010.


The Journal of Defense Research (JDR) was a classified publication sponsored by the Defense Advanced Research Projects Agency to encourage dissemination of classified research on topics of military or national security interest. It began publication in 1969, replacing the former Journal of Missile Defense Research.

Many years later, most of the Journal's contents still seem to be classified, but the table of contents of the Journal's first decade has been declassified and is now available on the Federation of American Scientists website here:

Perhaps unsurprisingly, most of the names of the authors whose work was published in JDR are unfamiliar, with a few exceptions (e.g., Garwin and Augustine; Hugh Everett's name does not appear). The topics of the papers provide a snapshot of the technologies and the strategic concerns of the time, and give an indication of the scale of classified government research that was devoted to addressing them.

"The JDR is 'mission essential' as a classified research tool," a Defense Science Board Task Force stated in 1985. "Being the only classified journal of its type, the JDR is used to communicate ideas amongst the defense community and is a basic tool for researchers."


Congress has forbidden the Congressional Research Service to make its publications directly available to the public, so it is left to others to do so. New CRS reports obtained by Secrecy News include the following.

"Intelligence Reform After Five Years: The Role of the Director of National Intelligence (DNI)," June 22, 2010:

"Questioning Supreme Court Nominees About Their Views on Legal or Constitutional Issues: A Recurring Issue," June 23, 2010:

"The 'Volcker Rule': Proposals to Limit 'Speculative' Proprietary Trading by Banks," June 22, 2010:

"The Nunn-McCurdy Act: Background, Analysis, and Issues for Congress," June 21, 2010:

"Environmental Considerations in Federal Procurement: An Overview of the Legal Authorities and Their Implementation," June 21, 2010:

"EPA Regulation of Greenhouse Gases: Congressional Responses and Options," June 8, 2010:


Secrecy News is written by Steven Aftergood and published by the Federation of American Scientists.

The Secrecy News blog is at:

To SUBSCRIBE to Secrecy News, go to:


OR email your request to

Secrecy News is archived at:

SUPPORT the FAS Project on Government Secrecy with a donation here: