Here is an unsung and perhaps unspectacular part of intelligence work, but I have often seen spectacular results emerge from it when our intelligence analysts are called upon to produce the estimate that the policymaker requires.
Some estimates are requested by senior policy officers of government to guide them in dealing with problems before them or to get an idea of how others may react to a particular line of action we may be considering. Others are prepared on a regularly scheduled basis, as, for example, the periodic reports on Soviet military and technical preparations. Before some estimates are prepared, a hurry-up call is sent to those who collect the intelligence to try to fill certain gaps in the information required for a complete analysis of a problem. Such gaps might be in the military or economic information available, or in our knowledge of the intentions of a particular government at a particular time.
Finally, estimates are often prepared because some member of the intelligence community feels that a certain situation requires attention. The cloud in the sky may be no bigger than a man’s hand, but it may portend the storm; and it is the duty of intelligence to sound an alarm before a situation reaches crisis proportions. While the charge is sometimes made that intelligence has failed to warn of some crises, the press and outsiders do not know the number of times that it has given the necessary warning because this, again, is one of the sides of intelligence that is not advertised.
One general range of subjects that receives constant attention and very frequent, regular estimates is the development of what we call military hardware, particularly by the Soviet Union. This means Soviet programs and progress in missiles, nuclear warheads, nuclear submarines, advanced type of aircraft and anything that might approach a breakthrough in any of the sectors of this field, as well as in the field of space. This is one of the most difficult tasks which faces the intelligence estimator.
Here one has to deal with Soviet capabilities to produce a given system, the role assigned to the system by the military and its true priority in the whole military field. It is always difficult to predict how much emphasis will be given to any particular system until the research and development stage has been completed, the tests of effectiveness have been carried out and the factories have been given the order to proceed with actual production. As long as a Soviet system is still in its early stages, our estimates will stress capabilities and probable intentions; as hard facts become available, it is possible to give an estimate of the actual programming of the system.
In 1954, for example, there was evidence that the Soviet Union was producing long-range intercontinental heavy bombers comparable to our B-52s. At first, every indication, including the 1955 fly-by I have described, pointed to the conclusion that the Russians were adopting this weapon as a major element of their offensive strength and planned to produce heavy bombers as fast as their economy and technology permitted. An estimate of the build-up of this bomber force over the next few years was called for by the Defense Department and supplied by the intelligence community. It was based on knowledge of the Soviet aircraft-manufacturing industry and the types of aircraft under construction, and included projections concerning the future rate of build-up on the basis of existing production rates and expected expansion of industrial capacity. There was hard evidence of Soviet capability to produce bombers at a certain rate if they so desired. At the time of the estimate, the available evidence indicated that they did so desire, and intended to translate this capability into an actual program. All this led to speculation in this country as to a “bomber gap.”
Naturally, intelligence kept a close watch on events. Production did not rise as rapidly as had seemed likely; evidence accumulated that the performance of the heavy bomber was less than satisfactory. At some point, probably about 1957, the Soviet leaders apparently decided to limit heavy bomber production drastically. The bomber gap never materialized. This became quite understandable, as evidence of progress in the Russian intercontinental missile program was then appearing and beginning to cause concern. Thus, while previous estimates of capability in bomber production remained valid, policy changes in the Soviet Union necessitated a new estimate on our part as to future development of the heavy bomber.
Intentions can be modified or policies reversed, and intelligence estimates dealing with them can rarely by unqualified. Witness how, just recently, our own intentions concerning the Skybolt missile have changed and how this must affect the calculations of Soviet intelligence.
The Soviet missile program, like that of the heavy bomber, had various vicissitudes. The Soviets saw early, probably earlier than we did, the significance of the missile as the weapon of the future and the potential psychological impact of space achievements. They saw this even before it was clear that a nuclear warhead could be so reduced in weight and size as to be deliverable over great distances by the big boosters which they correctly judged to be within the range of possibility. Given their geographical situation—their strategic requirements differ from ours—they soon realized that even a short- or medium-range missile would have great value in their program to dominate Europe.
The origins of the program go back to the end of World War II, when the Soviet Union, having carefully followed the progress made by the Germans with their V-1 and V-2 missiles, made every effort to gather together as much of the German developmental hardware and as many German rocket experts as they could get their hands on while they were conquering Eastern Germany. The Soviets also hired a considerable number of German experts in addition to those they seized and forcibly deported.
It would be a mistake, however, to credit their missile proficiency today largely to the Germans. The Russians themselves have a long history in this field and developed high competence quickly. They never took the Germans fully into their confidence but pumped them dry of knowledge, kept them a few years at the drawing boards and away from the testing areas, and then sent most of them back home. While these people proved to be a useful source of intelligence to the West, they had never been brought into contact with the actual Soviet development and could tell little beyond what they had themselves contributed.
In the early postwar years there was a good deal of skepticism in the United States about the future of guided missiles. One of the skeptics was Dr. Vannevar Bush, the outstanding head of our wartime Office of Scientific Research and Development, which coordinated the work of some 30,000 scientists, engineers and technicians. As late as 1949 he raised serious questions whether the guided missile could be “made to hit anything at the end of its flight”; he also felt its cost would be “astronomical.” He added that as a means of carrying high explosives, “it is a fantastic proposal.” He felt that in view of the cost of atomic bombs, we would not “trust them to a highly complex and possibly erratic carrier of inherent low precision.”2
2 Modern Arms and Free Men (New York: Simon & Schuster Inc., 1949).
While there were some eminent men of science who differed from this view, it nevertheless was widely held. In the postwar years, before we had developed the thermonuclear bomb and the small but relatively powerful nuclear weapons, we failed to give the attention to the guided missile which, in the light of hindsight, we should have given it.
Another reason for this failure, and here intelligence enters into it, was the fact that in the first decade after the end of the war, we had inadequate information with regard to the Soviet missile program.
Drawing boards are silent, and short-range missiles make little commotion. As the techniques of science were put to work and the U-2 photographs became available after 1956, “hard” intelligence began to flow into the hands of the impatient estimators. Their impatience was understandable, for great pressure had been put on them by those in the Department of Defense concerned with our own missile programs and missile defenses. Planning in such a field takes years, and the Defense Department felt that this was a case in which it was justified in asking th
e intelligence community to project several years in advance the probable attainments of the Soviet program.
As in the earlier case of Soviet bomber production, the intelligence community, I am safe in saying, would be quite content if it were not called upon for such crystal-ball gazing. But since military planning requires estimates of this nature, the planners say to the intelligence officers: “If you won’t give us some estimate as to the future, we will have to prepare it ourselves. You intelligence officers should really be in a better position to make the predictions than we are.” For the intelligence service to deny this would be tantamount to saying it was not up to its job.
Thus early figures of Soviet missile production had to be developed on the basis of estimated production and development capabilities over a period in the future. Once again it was necessary to determine how the Soviet Union would allocate its total military effort. How much of it would go into missiles? How much into developing the nuclear potential? How much into the heavy bomber, as well as the fighter planes and ground-to-air defense to meet hostile bombers? How much into submarines? And, in general, how much into elements of attack and how much into those of defense?
It was due to this measure of incertitude during the late 1950s that the national debate over the so-called missile gap developed. Then, based on certain proven capabilities of the Soviets and on our view of their intentions and overall strategy, estimates were made as to the number of missiles and nuclear warheads which could be available and on launchers several years in the future.
There is no doubt that tests of Soviet missiles in 1957 and afterward showed a high competence in the ICBM field. Soviet shots of seven to eight thousand miles into the far Pacific were well advertised, as, of course, was the orbiting of the first Sputnik. Their testing in the intermediate fields must also have been gratifying to them. But would they use their bulky and somewhat awkward “first generation” ICBM, effective though it was, as the missile to deploy, or would they wait for a second or third generation? Were they in such a hurry to capitalize on a moment of possible missile superiority that they would sacrifice this to a more orderly program? The answer, in retrospect, seems to be that they chose the orderly program. As soon as this evidence appeared, the ICBM estimates, as in the case of the bombers, were revised downward.
Today, after the Cuba incident of October 1962, when Khrushchev did install “offensive” missiles in Cuba, one may well ask whether their recent actions do not indicate that they are in more of a hurry with their missile program. They were willing to take great risks to get some IRBM and MRBM bases in Cuba to create the equivalent, as a threat to us, of a considerable additional number of ICBM bases in the heartland of Russia.
In any event, the intelligence collected on Soviet missiles has been excellent as to the nature and quality of the potential threat. Our intelligence was also both good and timely as to Soviet production of high-thrust engines and the work on Sputnik. And all this intelligence spurred us to press forward with our own missile and space programs.
There is an area of intelligence estimating involving military hardware that is confusing to the uninitiated. On innumerable occasions during my period of work with the CIA, I was asked how the United States stood as compared to the Soviet Union in various respects. Were our bombers better? Did we have more missiles? How did we stand in the race for nuclear weapons? Here I had to explain that, as intelligence officers, we were not experts on American military weapons development. The job of the intelligence officer is to appraise the military strength of other countries, not that of the United States.
It is important, however, for our own policymakers to have the answers to such questions about comparative strengths. To meet this need, procedures were set up during the Eisenhower administrations to form net estimative groups. Intelligence officers were always members of the groups; the other members included experts having full knowledge of United States programs in the particular family of weapons where comparisons were sought—missiles, bombers, nuclear bombs and the like. Then net estimates would be produced indicating the relative position of the two countries and, wherever possible, where we would stand in a few years given our own existing programs and our estimate of those of the Soviet. This proved to be a most useful exercise.
When one turns from the military to the political field, the problems for the estimators are often even more complex. Analysis of human behavior and anticipation of human reactions can never be assigned to a computer, and they baffle the most clever analyst.
More than a decade ago, in the autumn of 1950, this country had to face in North Korea the difficult decision of whether or not to push forward to the Yalu River and reunite Korea. If we did so, would the Chinese Communists answer with a direct attack? Or would they stay quiescent—if, for example, Korean rather than U.S. and UN troops formed the bulk of the advance, or if we did not disturb the Chinese sources of electric power in North Korea?
At that time, we had good intelligence as to the location and strength of the Chinese Communist forces on the far side of the Yalu. We had to estimate the intentions of Moscow and Peking. We were not in on their secret councils and decisions. In such cases, it is arrogant, as well as dangerous, for the intelligence officer to venture a firm opinion in the absence of telltale information on the positioning and moving of troops, the bringing up of strategic supplies and the like. I can speak with detachment about the 1950 Yalu estimates, for they were made just before I joined the CIA. The conclusions of the estimators were that it was a toss-up, but they leaned to the side that under certain circumstances the Chinese probably would not intervene. In fact, we just did not know what the Chinese Communists would do, and we did not know how far the Soviet Union would press them or agree to support them if they moved.
One cannot assume that a Communist leader will act or react as we would or that he will always be right in his estimates of our reactions. In Cuba, in October of 1962, Khrushchev presumably “estimated” that he could sneak his missiles into the island, plant them and camouflage them, and then, at a time of his own choosing, face the United States with a fait accompli which we would accept rather than risk war. Certainly here he misestimated—just as some on our side had misestimated that Khrushchev would not attempt to place offensive weapons in Cuba, right under our nose.
The role of intelligence in the early phases of the Cuban crisis of October, 1962, was the subject of a public report by the Preparedness Subcommittee of the Armed Services Committee of the Senate, under the chairmanship of Senator John Stennis of Mississippi. The subcommittee’s main conclusion reads as follows: “Faulty evaluation and the predisposition of the Intelligence Community to the philosophical conviction that it would be incompatible with Soviet policy to introduce strategic missiles into Cuba resulted in intelligence judgments and evaluations which later proved to be erroneous.”
This criticism of intelligence was directed to the period in September and early October, prior to the obtaining of adequate photography. Then there had been certain intelligence estimates to the general effect that it was unlikely intermediate-range missiles, i.e., missiles which could reach far into the United States, would be introduced into Cuba by the Soviets. There were some people, however, notably Mr. McCone, the Director of Central Intelligence, who had expressed at the time serious premonitions, but the intelligence community generally felt that Khrushchev would not risk a course of action so directly threatening to the United States and one which subsequent activities showed he was prepared to abandon abruptly in the face of strong American reaction. Cuba is yet another instance to warn us that one must be prepared for the Soviets to do the unexpected, the unusual, the shocking, confident in his own ability to retreat, as well as to advance, when the opposition gets too hot and also confident that he can make these retreats without seriously affecting his own domestic position. With complete control of the media of communication within his own country, he can explain away
a retreat in Cuba as just another example of the “peaceful” posture of the Soviet Union.
In the preparation of estimates with regard to Soviet policy, their actions and reactions, it is always well to have among the estimators one or two persons who are designated to play the roles of the devil’s advocate, who can advance all the reasons why a Khrushchev could take an unusual, dramatic or, as viewed from our own vantage point, even an unwise and unremunerative course of action. Of course, one would reach rather ridiculous conclusions, and certainly wrong conclusions in most cases, if one always came up with an estimate that the abnormal is what the Soviet Union will probably do. It is well, however, that the policymakers should be reminded from time to time that such abnormalities in Soviet action are not to be excluded.
If some of our own estimators went wrong in the Cuban affair, Khrushchev and his advisers committed an even more serious misestimate in apparently concluding that he could get away with this crude maneuver without a stern American rejoinder. Intelligence officers have to face the fact that whenever a dramatic event occurs in the foreign relations field—an event for which the pubic may not have been prepared—one can usually count on the cry going up, “Intelligence has failed again.” The charge may at times be correct. But there are also many occasions when an event has been foreseen and correctly estimated but intelligence has been unable to advertise its success.
This was true of the Suez invasion of 1956. Here intelligence was well alerted as to what Israel and then Britain and France were likely to do. The public received the impression, however, that there had been an intelligence failure; statements were issued by U.S. officials to the effect that the country had not been given advance warning of the action. Our officials, of course, intended to imply only that the British and French and Israelis had failed to tell us what they were doing. In fact, United States intelligence had kept the government informed but, as usual, did not advertise its achievement.
The Craft of Intelligence Page 21