By JAY ACKROYD

     Official secrecy is a recent development in the United States. Before World War I, America took pride in not practicing the skulking ways of European diplomacy and our foreign policy was publicly disclosed. After the war, America’s new prominence in the international arena put a dent in our commitment to open diplomacy. The crushing blow came after WWII, as U.S. fears of a Soviet regime committed to espionage, secret dealings, and the overthrow of the Government led to institutionalization of a regime of secrecy in the United States.

     Increased circumspection in our dealings with other nations may have been a reasonable response to the foreign threat, but there were other options available, even then. Eisenhower, for instance, advocated a doctrine he called "Open Skies." We have nothing to hide, he said. Let anyone fly over our country and see for themselves—and we will fly over theirs. This doctrine was at least somewhat disingenuous; America was a lot better than anyone else at flying over other countries. But the plain fact of the matter was that other countries, particularly the Soviet Union, had much more to hide. Nevertheless, Eisenhower’s policy did not prevail.

     Evaluating the seriousness of the Soviet threat, even in hindsight, is difficult. The Communists certainly intended to overthrow the U.S. They recruited spies and collaborators and were quite successful at uncovering key American secrets. They penetrated Los Alamos twice, once to get the secrets of the A-bomb, once to get the secrets of the H-bomb. But we now know that, certainly by the mid-1970s, the Russians were weak and getting weaker. Today, an enormous national security apparatus, created in response to an enemy which may never have been dangerous enough to warrant it, and which in any case no longer exists, costs American taxpayers more than—nobody knows exactly how much more than—$35 billion a year.

     It can be argued that Soviet ineffectualness has not always been a certainty. But the difficulty of trying to maintain important secrets for any length of time has long been obvious. Many technological secrets, for instance, are revealed, at least in part, as soon as it becomes known that someone has solved a problem. Simply knowing that an H-bomb can be built, given the state of current scientific knowledge, is an enormous leg up to a scientist trying to build one: his goal is then known to be within reach. Moreover, governments have nearly limitless resources to throw at key individuals. They can inevitably find someone who, for a price, will figure out what others have figured out, or discover what others know.

     Intelligence assessments of political situations are also hard to keep secret. If they are to have any value, they must at some point be disclosed to decision-makers. And, intelligence often relies on human beings as sources of information, and human beings can have conflicting interests. Thus, both the sources for and the recipients of assessments represent potential leaks. As time passes, the risk of leaks grows steadily, so that if an assessment is still relevant after any length of time it is generally widely known, at least in summary.

     Still, one might assume that if effective secrecy is hard to come by, partial secrecy is better than nothing. Surely, any measures that slow down the spread of U.S. technological advances to other countries and hide information about U.S. espionage activities are a good thing?

     Unfortunately, there is no free lunch. We pay a price for secrecy, one that is not always obvious and can quickly exceed the dubious benefits it is intended to purchase. The two biggest costs are greater inefficiency in research and development and unreliability of intelligence assessments—both of which have the effect of subverting, not enhancing, national security.

     Recent public revelations that the Chinese acquired critical information about the construction of nuclear warheads from the United States illustrates both the futility of trying to keep key information secret and the inefficiencies secrecy creates. Yet we still hamper our researchers in a vain attempt to lock down information. We try to prevent secrets from leaking out of computer systems by creating classified computer networks unconnected to any other networks. Hackers can't reach these "black" networks because they aren’t hooked into the outside world. But scientists often need access to non-classified information that is resident on public networks—information such as their own non-classified work, email exchanges with colleagues, newsgroup discussions, and the Web. So, scientists working on classified projects must have a pair of computers on their desks, one black and one connected to the world. Both computers have floppy disk drives, which scientists require in order to "sneakernet" information between their two devices. But this makes it easy for a spy to transfer information from the black network to the public network.

     There is, of course, an obvious way to stop this: take the floppy drives out of the black computers. But that solution introduces massive inefficiencies. Now the scientist must retype information in both directions. Since most of the information he really needs isn't classified, removing the floppy drive turns the scientist into a clerk entering public information into the black network. And it could get even worse than this. One way to guard against leaks is to compartmentalize activities, so that scientist A doesn't know what scientist B is doing unless he has a "need to know". But that would mean they would both be typing the same non-classified information into their secured areas.

     You could lose a lot of scientists that way. Recently, John Deutch, former Director of Central Intelligence, lost his security classification for transferring information via floppy disk to his unsecured home computer. If even a former DCI finds the rules so onerous that he felt it necessary to violate them to get his work done, how can routine violations not be expected?

     The bottom line is, you can't keep the secrets and keep the scientists. And you can't manage without the scientists. Simple solution? Give up on keeping secrets. Technology secrets have a short half-life anyway--something like four years, according to a report issued by the Federal Commission on Protecting and Reducing Government Secrecy. So why not declassify everything after five years?

     "Horrors!" cry the General Buck Turgidsons among us. "They'll see everything!" And so they will. But they already see everything that really matters. The way to stay technologically ahead of other nations, as Edward Teller (father of the H-bomb and a proponent of the Strategic Defense Initiative) has advocated throughout his career, is to keep doing cutting edge research—not to try to keep the advances secret.

     Of course, technology isn’t all there is to national security, which depends as well on secret analyses of what we expect foreign governments to do. To all appearances, the U.S. intelligence agencies’ foreign analysis record over the last quarter-century is nothing short of pathetic. They failed to predict the fall of the Shah of Iran, misanalyzed the strength of OPEC, overestimated Soviet power for at least a decade, were caught by surprise by the Indian nuclear test, can't find Saddam Hussein, and lost track of the Chinese embassy in Belgrade.

     Much of this pattern of failure stems directly from trying to keep analyses secret. Secrecy prevents outside scrutiny, which leads to hardening of the intelligence arteries. For instance, we failed to accurately assess the Soviet economy in the 1980s because we blew it in the ’50s, greatly overestimating the U.S.S.R.’s ability to sustain high rates of economic growth. Daniel Patrick Moynihan, in his book Secrecy, tells of a general in charge of nuclear weapons targeting, making his first landing at an airport in the Soviet Union in the early ’80s. The general was shocked at the rutted runway, the rusting equipment, the overall state of disrepair. There was no relationship between what he was seeing and what he had read in his intelligence reports.

     Decades of analysis without external review had preserved lines of previous analyses instead of challenging them, and had led to wholly inaccurate "intelligence." What analyst was going to risk his or her job pointing out that the numbers didn't add up? Who was going to tell William Casey that there was no window of nuclear vulnerability? Had assessments of Soviet power from the 1950s to the 1970s been declassified, experts on the Soviet economy from outside the intelligence community could have provided oversight, and the presidents of at least four administrations would not have been routinely given false information.

     The Cold War was not won by secrecy. It was won by the strength of two American concepts: free enterprise and democracy. If we conduct ourselves in an open, democratic way, living by the principles we espouse, there is no need for an extensive security apparatus. While there are times when some measure of secrecy is necessary, as when diplomatic negotiations are in progress or when a war is being fought, there certainly is no need to keep secrets indefinitely, or for decades. Doing so actually degrades national security because it prevents accountability and accuracy.

     With the Cold War over, the United States needs to downgrade security practices to peacetime levels. With the nation facing no immediate danger from any other state, we can, and should, commit ourselves again to Eisenhower's vision. Openness and accountability are at the heart of a strong and truly secure democracy. The free exchange of information invigorates capitalism. Secrecy undermines the principles we wish the world to follow for the sake of our own interests and those of other nations. Isn't it time for us to have the courage to abide by our principles?

Jay Ackroyd is the president of DBSI, Inc., a custom software development company based in New York City.