from the FAS Project on Government Secrecy
Volume 2012, Issue No. 127
December 19, 2012

Secrecy News Blog:


It is plainly true that executive branch officials will sometimes disclose classified information to reporters and other uncleared individuals. But this practice is not explicitly authorized in any official statement of classification policy. In fact, with an exception for life-threatening emergencies, it is usually understood to be prohibited.

How can the obviously flexible practice and the seemingly prohibitive policy be reconciled? A newly updated report from the Congressional Research Service presents a close reading of the relevant rules and regulations in search of some wiggle room for authorized disclosures of classified information.

"Nothing in the Executive Order addresses an informal procedure for releasing classified information [to reporters]. E.O. 13526 section 1.1 provides that '[c]lassified information shall not be declassified automatically as a result of any unauthorized disclosure of identical or similar information,' but does not address what happens in the event of a disclosure that was in fact authorized," the CRS report observes.

"By definition, classified information is designated as such based on whether its unauthorized disclosure can reasonably be expected to cause a certain level of damage to the national security. This may be read to suggest that disclosures may be authorized under such circumstances when no damage to national security is reasonably expected." (But under those circumstances, it might be noted, the information should be promptly declassified.)

The CRS report, written by legislative attorney Jennifer K. Elsea, continues: "Nothing in the order provides explicit authority to release classified information that exists apart from the authority to declassify, but it is possible that such discretionary authority is recognized to release information outside the community of authorized holders without formally declassifying it." Indeed, this appears to be an accurate characterization of actual practice.

In any case, "there is little to stop agency heads and other high-ranking officials from releasing classified information to persons without a security clearance when it is seen as suiting government needs." Again, an accurate description-- particularly since "the Attorney General has prosecutorial discretion to choose which leaks to prosecute."

See "The Protection of Classified Information: The Legal Framework," updated December 17, 2012:

Overall, "Executive Branch policy appears to treat an official disclosure as a declassifying event, while non-attributed disclosures [to reporters or others] have no effect on the classification status of the information," the author writes.

"For example, the Department of Defense instructs agency officials, in the event that classified information appears in the media, to neither confirm nor deny the accuracy of the information. The Under Secretary of Defense for Intelligence is then advised to 'consult with the Assistant Secretary of Defense for Public Affairs and other officials having a primary interest in the information to determine if the information was officially released under proper authority.'"

But, the CRS report astutely notes, the relevant DoD regulation "does not clarify what happens in the event the disclosure turns out to have been properly authorized."

And so it seems that the DoD regulation offers the conceptual space for an authorized disclosure of classified information.

(As if to provide an ironic illustration of the point, the Under Secretary of Defense for Intelligence himself -- Michael Vickers -- was reportedly cited in a referral to the Department of Justice for disclosing potentially restricted information concerning the pursuit of Osama bin Laden to filmmakers. See "Bin Laden film leak was referred to Justice; leaker top Obama official" by Marisa Taylor and Jonathan S. Landay, McClatchy Newspapers, December 17, 2012. In a statement last night, the Department of Defense confirmed that Mr. Vickers is a subject of a pending Inspector General investigation. But it said the information in question was unclassified in its entirety.)

The CRS report naturally does not constitute an authoritative interpretation of the executive order, and in some respects it may be in error. The report mistakenly states (at footnote 51) that the DOJ Media Leak Questionnaire that agencies must complete when a referring a leak for investigation is "apparently... part of a Memorandum of Understanding concluded between the Department of Justice and elements of the Intelligence Community." But a review of the Memorandum, described in Secrecy News earlier this week, shows that that supposition is incorrect. The two are separate documents. See "Crimes Reports and the Leak Referral Process," Secrecy News, December 17, 2012.

Anti-leak legislation that is pending in the Senate would require executive branch officials to record all authorized disclosures of classified intelligence to the press, and to notify Congress when they occur (cf. sections 501 and 502 of Title V of the FY 2013 intelligence authorization bill).

These provisions, which may prove unworkable in practice, are presumably intended to enable Congress to publicly comment on classified intelligence matters with the same freedom that agency officials already do. But the public interest concern raised by the notification provisions is that if they are strictly imposed, they may discourage all authorized disclosures of classified intelligence, yielding a net reduction in public access to government information.


New and updated reports from the Congressional Research Service that Congress has directed CRS not to release to the public include the following.

Rising Economic Powers and U.S. Trade Policy, December 3, 2012:

Unauthorized Aliens Residing in the United States: Estimates Since 1986, December 13, 2012:

DOD Alternative Fuels: Policy, Initiatives and Legislative Activity, December 14, 2012:

Federal Land Ownership: Current Acquisition and Disposal Authorities, December 13, 2012:

The Controlled Substances Act: Regulatory Requirements, December 13, 2012:


Intelligence community officials have been meeting with representatives of the National Archives to discuss the anticipated declassification and release of intelligence imagery from the KH-9 satellite dating between 1971 and 1984.

Officials have been negotiating the transfer of the original negatives from the KH-9 system and the provision of finding aids, according to a newly released but heavily redacted report from the National Geospatial Intelligence Agency, dated June 2012.

Multiple releases of declassified imagery are planned over the coming year "with final delivery of imagery scheduled for September 2013." See "Intelligence Imagery Set to be Disclosed in 2013," Secrecy News, October 22, 2012.


The latest report from the elite JASON science advisory panel is devoted to the subject of "compressive sensing." This term generally refers to the use of sensors for imaging (or other sensing) of an object in a manner that uses a limited subset of the available data in order to improve efficiency or conserve resources.

"Compressive sensing involves intentionally under-sampling an object or image, typically in a random manner, and then using a companion process known as sparse reconstruction to recover the complete object or image information...," the JASON report says.

"Compressed sensing can conceivably lead to reductions in data link requirements, reductions in radar resources needed for radar image formation (thereby providing the radar more resources for its other functions such as target detection, target tracking, and fire control), increased angular resolution without commensurate increases in array costs, and increased fields of view without degradation in resolution..."

"Compressive sensing is not a 'free lunch'," the report cautions, "but always involves a tradeoff; reduced data may save measurement resources, but it also means a lower signal-to-noise ratio and possibly other artifacts, such as side lobes or false alarms."

A copy of the new JASON report was obtained by Secrecy News. See "Compressive Sensing for DoD Sensor Systems," November 2012:


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: