Lemnitzer had another objection to the “controlled response” notion: “the advantages to be achieved by limiting our responses, under such conditions, could only be realized by the enforcement upon the Soviets of a degree of tacit ‘cooperation’ which does not now appear realistic.” This had troubled McNamara, too, during the Kaufmann briefing: it takes two to fight a controlled nuclear war, and how would we know that the Soviets, contrary to everything that their leaders said on the subject in public, would play along? In the meantime, however, SIOP-62 had to be replaced with something more controlled, more discriminating, so that McNamara—and the world—could avoid being plagued with its excessive “requirements” for more nuclear weapons and, in the event of “general war,” its monstrously catastrophic consequences.
Project No. 1 of McNamara’s 96 Trombones—“Prepare a draft memorandum revising the basic national security policies and assumptions, including the assumptions relating to ‘counter-force’ strikes….”—elicited a much more sympathetic response. In the International Security Affairs office, the Assistant Secretary of Defense, Paul Nitze, turned the project over to Harry Rowen, who in turn gave it to his very close friend from RAND, now a Pentagon consultant, Daniel Ellsberg. To Rowen, Dan Ellsberg was the ideal man to work on Project No. 1. Just out of Harvard graduate school, before joining RAND full-time, Ellsberg had made something of a splash in the academic world with a series of lectures on “The Art of Coercion,” which used Game Theory to illustrate how nations bluff, coerce and blackmail one another in wars, crises and diplomacy. At RAND, Ellsberg came under the wings of the Wohlstetter clique on SAC vulnerability and firmly attached himself to elements of the Kaufmann philosophy as well.
More than this, Ellsberg was one of the very few civilians anywhere who was intimately familiar with the military’s war plans. In the fall of 1959, RAND had loaned Ellsberg to the Office of Naval Research, under Admiral Harry Felt of the Pacific Command, to do a study on the security and efficiency of command-control-communications systems for nuclear war. Ellsberg felt that to do the study well he had to see the war plans, and got permission from Felt to go into the Top Secret cage at the Command and look at anything he wanted. He read all the war plans that affected the Pacific Command, not just the high-level JCS directives but the operational plans and procedures affecting individual fleets and carriers. No one had ever read the plans beyond one level above and below that of his own set of instructions. Thus no one had any idea of the enormous discrepancies between the basic guidance furnished by the Joint Chiefs of Staff and the orders as they finally emerged in the specific instruction to the ship’s commanding officer in the fleet.
The basic mission of the Pacific Command, or PacCom, in the event of nuclear war, was to bomb cities in Red China. For one thing, those were the easiest targets to hit for attack airplanes flying off carriers based in the Pacific Ocean. One of Ellsberg’s most shocking discoveries was that if the JCS instructed PacCom to hit military targets in Russia instead, the fleet would almost certainly end up hitting cities in China anyway. A particular command’s war plans are only consistent with its geographic deployment, training practices and alert procedures. And all of these were geared to hitting cities in China. In the cascade of messages that would be sent down from the top level to the next level to headquarters to command to carrier, there was no easy way to alter the Emergency Action Message so that the mission could be so radically changed.
In the spring of 1960, Ellsberg spent a good deal of time in Washington, D.C., exploring the war plans even more deeply. While visiting the National Military Command Center, the War Room in the Pentagon, he discovered that the officers transmitting the alert codes did not have copies of the messages that were, in turn, being sent to lower levels of command. When Ellsberg asked the Duty Officer to see the book that decoded the messages, the officer could not find it. At every level of the war plan, then, Ellsberg found a rote set of reactions, confusion, inflexibility, virtually no practical way to modify the all-out, hit-everything-at-once character of the official war plan.
It was also in Washington that Ellsberg met a group of fairly like-minded colonels in the Air Force Staff—especially Glenn Kent, Russell Dougherty, Ernie Cragg and Bob Lukeman—who, after hearing Ellsberg describe the shocking state of affairs in the Pacific, let him in on some of the official JCS and Air Force war plans as well, and talked with him about these plans for several days, six to eight hours at a time. It turned out that these plans were terribly inflexible as well.
SAC had always insisted that there be essentially one war plan—a few “options” to adjust for differences in the weather or various conditions of warning time, but fundamentally a single plan. Especially in the days before very fast computers, devising one war plan was difficult enough. Every detail had to be preplanned so that bombers would hook up with in-flight refueling tankers at just the right time, go on to destroy their targets at just the right time, and return to base. In all of this, there was a phenomenon to be avoided called “interference.” When one airplane dropped its bombs, SAC wanted to make sure that another pilot would not be flying just overhead, only to be killed by the bomb’s blast or blinded by the flash of the fireball. Avoiding this “interference” required very complicated flight paths; the timing had to be precise, down to the minute.
Yet Ellsberg also learned some facts that made SAC’s rationale for having just one war plan appear ludicrous. The communication of Emergency Action Messages often got so scrambled in exercises that different SAC bases would receive the message to “Go” as much as four hours apart from one another. Then the wind differed from place to place and day to day, which also affected flight times. These two factors alone would completely wreck all efforts by the war planners to set off explosions over Russian targets simultaneously and to avoid the “interference” that might prevent a few bombardiers from executing their missions. The finely detailed coordination built into the war plan in fact had little bearing on what would actually happen in a nuclear war. Ellsberg also knew that the SAC planners themselves must realize this, suggesting that all the talk about avoiding “interference” was a ruse, a way of excusing SAC and the Air Force from working so hard on devising alternative war plans, a further barrier justifying the all-out, max-effort, bomb-everything philosophy of nuclear war deeply embedded in SAC’s collective psyche.
The Air Force colonels also showed Ellsberg one of the most tightly held documents in the JCS files—the JSCP, the Joint Strategic Capabilities Plan, referred to as the J-Scap. The JSCP was the war plan; Annex C of the JSCP, the Atomic Annex, specified the guidance for nuclear attacks. No civilian had seen the JSCP before—not the Secretary of Defense, not the President. They were shown something called the JSOP, the Joint Strategic Operational Plan, which was similar but not precisely the same. One feature of the JSCP, but not the JSOP, was a definition of “General War”: “A general war is an armed conflict in which Armed Forces of the U.S.S.R. and those of the United States are overtly and directly engaged.” In 1958, the Army had tried to add to this sentence the phrase, “as principal protagonists with the national survival of both deemed to be at issue,” but the Air Force succeeded in removing the amendment. And in general war, the U.S. Emergency War Plan—the nuclear war plan, which was superseded in late 1960 by the SIOP—would be executed.