What is to be learned from the space shuttle disaster is that the entire human race is in a spaceship and the decision-making time for correcting mistakes has been cut to seconds. The men at the controls have become so preoccupied with the momentum of the arms race that they are neglecting the painstaking joint attention required to keep the spaceship on course.
The tragedy should remind the world that, despite the most systematic precautions, accidents are possible when dealing with high technology. No machines are impervious to accident or malfunction, whether bicycles or automobiles or computers or spaceships or intercontinental ballistic missiles or the firing devices on nuclear explosives.
The space shuttle accident, therefore, forces us to fix attention on the intricate technology basic to the war capability of the United States and the Soviet Union.
Begin with the computerized system for detecting attacking missiles. During World War II, civilian spotters stood on hilltops of American coastal cities to report the sight or sound of airplanes by telephone to a control station. These reports would be checked against information at hand about posted flights. Unidentified aircraft would be monitored on a grid from station to station. At the beginning of the war, military aircraft flew by propellers at speeds under 400 miles per hour; many of them under 200 m.p.h. There was ample time for the tracking station to check and cross-check all movements in the sky, with time to alert our own combat aircraft as well as to activate anti-aircraft equipment.
By the end of the war, however, speeds had increased to the point where civilian spotters were as useless as Indian smoke signals against an attacker with cannons. Radar replaced human surveillance. Within a few years, military jets were flying faster than the speed of sound. Then came ballistic missiles, with speeds measured not in terms of hundreds of miles per hour but in thousands. Even enemies on opposite sides of the world were only two or three hours apart.
As the speed of missiles increased, so did reliance on technology. By the mid-1970s, computers were given the job of analyzing data appearing on radar screens from tracking stations around the world. The presence of unauthorized or strange objects would be picked up by radar and fed into computers capable of determining whether such objects were possible attacking missiles or planes. We would then have an hour or two to activate defenses or retaliatory missiles. The rate of malfunction was remarkably low.
Even so, from 1981 to 1985, more than 100 possible missile attacks on the United States were flashed on our military computer screens. Fortunately, we had enough time to check out these alerts and determine that they were the result of computer error. It might take as many as 30 minutes to discover that a signal was false. But at least we had that margin of time to prevent buttons from being pressed--to prevent the loosing of a nuclear attack on the enemy that would result, in all probability, in a counternuclear attack on the United States.
Since there is no reason to believe that Soviet computer technology is superior to our own, it becomes necessary to recognize that erroneous blips have turned up on Soviet computer screens. But America's very success in placing missile launching platforms close to Soviet borders has reduced the time available for Soviet experts to check for possible computer errors. For example, American Pershing 2 missiles are less than 10 minutes away from major Soviet targets. Since this may not provide enough time to rule out the possibility of computer error, Soviet decision-makers may have to bet the life of their nation on guesswork. This puts not just Soviet and American citizens but all the world in jeopardy because of computer error or malfunction.
The fact of Soviet submarines with missile launchers not far off the coasts of the United States has a similar effect on American defense strategy--so much so, in fact, that serious discussions have been held inside the government as to whether we should turn loose our nuclear retaliatory capability on alert rather than on verification, since there may not be enough time for the latter.
Perhaps the most basic flaw of all in the computer alert system is that it doesn't allow for third-party complicity. The computer can, of course, tell something from the shape and a great deal from the early location of a radar blip, but not all blips are picked up at site of origin. After only a few minutes, the precise source can be something of a guessing game. A missile launched by a submarine, for example, provides no automatic or certain information about the identity of the sender. A third party that thinks it is in a position to profit from a war between two other nations could conceivably launch its missile at one or the other, thus setting off a nuclear chain reaction.