from the FAS Project on Government Secrecy
Volume 2002, Issue No. 110
November 4, 2002


The National Security Agency observed its fiftieth anniversary last weekend in a characteristically low key manner.

("How you can tell an extrovert from an introvert at NSA? In the elevators, the extroverts look at the OTHER guy's shoes." Or rather, the NSA extroverts are the ones that were telling that joke last weekend.)

NSA, the nation's codemaking, codebreaking and signals intelligence organization, was established on October 24, 1952 by President Harry S. Truman in a top secret, 8-page presidential memorandum. Formal announcement of the new agency was delayed until November 4, 1952 -- Election Day -- in order to keep the creation of the Agency out of the news, according to NSA.

Speaking at a November 1 anniversary ceremony at NSA headquarters at Fort Meade, Maryland, historian David Kahn offered his thoughts on "the death of cryptanalysis."

Kahn, author of The Codebreakers and other pioneering histories of cryptography, noted the technological challenges confronting NSA and observed that it is far from the omniscient, omnipotent entity that outsiders sometimes imagine.

"NSA doesn't know or control everything, as shown by public-key cryptography and the beating NSA took on key escrow and the fact that U.S. Navy submarines use Microsoft Windows," he said.

See David Kahn's invited remarks here:

President Truman's 1952 memorandum establishing the NSA is available on the website of the National Security Archive here:

A January 2001 Congressional Research Service report entitled "The National Security Agency: Issues for Congress" by Richard A. Best Jr. may be found here:


"It is not the intention of the [Bush] administration to create a new category of sensitive but unclassified [information]," said White House science advisor John H. Marburger III last week.

"But the fact is that even at the present time, federal agencies do have information that they don't generally release to the public, like law enforcement information and lists of vulnerabilities of certain types that are simply not made public, but they're not classified," Dr. Marburger said. He spoke on the National Public Radio program "Talk of the Nation" on November 1.

However, Dr. Marburger's assurances did not immediately relieve concerns among other scientists about the impact of the "sensitive but unclassified" designation.

"Despite what's been said, my sense is the administration has gone well beyond the kinds of controls that are fairly innocuous," said Dr. Steven M. Block of Stanford University.

The archived audio of their conversation can be heard here:

Since there will always be some disagreement about what official information should and should not be disclosed, the process for making that determination should be as open and as equitable as possible.

"Crafting a new policy that responds to sometimes competing interests in security and public access should not be an extraordinarily difficult task."

See my article "Making Sense of Government Information Restrictions" in the Summer 2002 edition of "Issues in Science and Technology" here:


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: