from the FAS Project on Government Secrecy
Volume 2004, Issue No. 107
December 3, 2004


The Department of Homeland Security "needs to remember that the homeland whose security it is protecting is one in which democratic debate is supposed to be open and freewheeling," the Washington Post editorialized today.

The Post editorial today casts a much-needed spotlight on two extraordinary aspects of Homeland Security secrecy policy: a requirement that DHS employees sign non-disclosure agreements as a condition of access to certain unclassified information, and the secrecy imposed by the Transportation Security Administration on an entire body of security directives that govern interactions with the public.

See "Homeland Secrecy," Washington Post, December 3 (free reg. req'd):

The problem of secret but legally binding security directives was explored in an article I wrote in Slate on November 18. See "The Secrets of Flight":


The U.S. government is poorly configured to produce, receive and act upon scientifically literate and technically informed policy advice, according to a new report from the Federation of American Scientists.

"The need for effective science and technology advice continues to increase while the infrastructure for providing such help is in a state of crisis," the report begins.

Among other structural problems, "the gap created by the loss of the congressional Office of Technology Assessment in 1996 has not been filled." Meanwhile, science advice to the executive branch has declined with the effective demotion of the president's science advisor in the Bush Administration.

Furthermore, arbitrary and ill-considered controls on public access to certain scientific and technical information have impeded policy formulation and public accountability.

The FAS study proposes a series of policy options for addressing these problems and reversing the decline in science advice to government, should there be a will to do so.

See "Flying Blind: The Rise, Fall, and Possible Resurrection of Science Policy Advice in the United States," by Henry Kelly, Ivan Oelrich, Steven Aftergood, and Benn H. Tannenbaum, December 2004 (117 pages, 355 KB PDF file):


Sometimes the act of classifying scientific or technical information can diminish national security instead of enhancing it.

Last month, a White House panel concluded that the growing classification of government research on computer security is not serving the nation well because it renders such research inaccessible outside of narrow military and intelligence channels.

"Classified cybersecurity R&D is, of course, needed for numerous purposes," observed F. Thomson Leighton, chair of the cybersecurity subcommittee of the President's Information Technology Advisory Committee.

"However, classified work tends not to benefit generic cybersecurity products--which are used throughout society (including the military and intelligence communities)," he said at a meeting last month.

In the future, he said, the government should "favor unclassified basic research" in cybersecurity.

Leighton's speech was first reported in the newsletter Inside the Pentagon on November 25.

See "White House Panel Blasts Pentagon's Cybersecurity R&D Policies" by John T. Bennett, Inside the Pentagon, reposted with permission and with a link to the underlying presentation here:


A wide spectrum of science and technology policy issues treated in the last Congress are surveyed and summarized in a recent Congressional Research Service report.

See "Science and Technology Policy: Issues for the 108th Congress, 2nd session," updated September 8, 2004:


An innovative technology for enriching uranium known as the Separation of Isotopes by Laser Excitation or SILEX process has the distinction of being the first and only privately generated information to be classified by the U.S. Government as Restricted Data under the Atomic Energy Act (Secrecy News, 06/26/01).

A critical assessment of the SILEX technology, which is under development in Australia, was presented in a new report by Greenpeace, which opposes the project.

See "Secrets, Lies, and Uranium Enrichment: The Classified SILEX Project at Lucas Heights," November 2004 (1.7 MB PDF file):


Following months of quiet diplomacy as well as public controversy, the Central Intelligence Agency has yielded to persistent demands from the Republic of Korea that the Agency change the way it spells the name of the South Korean president.

Let no one say that the CIA is incapable of reform.

"The Web site of the Central Intelligence Agency (CIA) corrected the name of the South Korean president to Roh Moo-hyun from previously spelled No Mu-hyun," the Yonhap News Agency in Seoul reported this week, referring to the latest revision of the CIA World Factbook.

(See "South Korea to CIA: It's Roh Not No," Secrecy News, 07/14/04; and "A Lesson in Korean Linguistics," Secrecy News, 07/19/04).

But no matter how many concessions the CIA makes, there are some critics who will never be satisfied.

"Many references [in the CIA World Factbook] still remain wrong," the Yonhap article stated. "The CIA site spells the name of the North Korean leader Kim Jong-il as 'Kim Chong-il' and still applies the McCune-Reischauer system of romanization to spell South Korean provinces, such as 'Cheju' and 'Cholla,' rather than the government's official spelling, 'Jeju' and 'Jeolla'."

See "CIA Factbook Corrects Spelling of S. Korean President," Yonhap News Agency, November 30, 2004:


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

To SUBSCRIBE to Secrecy News, send an email message to with "subscribe" (without quotes) in the body of the message.

To UNSUBSCRIBE, send a blank email message to

OR email your request to

Secrecy News is archived at:

Secrecy News has an RSS feed at: