As I gathered from talking with CINCPAC nuclear planners, there was a strong incentive for them to assume—and they did assume—that under any circumstances in which we were fighting Russia, we would also want to annihilate its Communist partners, the Chinese. Because of range limitations, almost no Russian targets lay within CINCPAC reach, except for a few in the area of Vladivostok and Siberia. Thus, if the president gave an order to attack only Soviet targets, CINCPAC forces, having destroyed Vladivostok and a few other minor targets in eastern Russia, would essentially have to sit out the war as observers—“on the sidelines,” as they thought of it—during the big game.
That this thought was intolerable to officers in the Pacific at levels near the very highest was confirmed for our whole study group on the afternoon we made an official visit to the flagship of the Seventh Fleet, the St. Paul, steaming in western Pacific waters. After landing by helicopter from a carrier, we held a meeting with Vice Admirals Kivette and Ekstrom, Commander of Naval Air in the Pacific.
I have previously described their comments on the problem of delegation. However, by far the strongest reaction to any question we raised with the two admirals in our two-hour meeting came when I mentioned as a possibility a decision by the president to go to war against the Soviet Union alone, not against China. Both admirals drew back and seemed genuinely to go into shock. Admiral Kivette said, “I would hope that’s out of the question!”
I repeated the question: “But suppose that an order did come from the JCS to execute war plans against the Soviet Union only. How would you respond to it, and how long would it take you?”
There was a long silence in which it appeared that Admiral Ekstrom was almost holding back an urge to vomit. Then he said, enunciating each phrase separately, almost gasping, as if in pained incredulity, “You have … to assume … some … modicum … of rationality … in higher authority … that they would not do something … so insane … as to go to war … against one Communist power … while letting the other one off … scot-free.”
Faced with such a visceral response (the elisions are in my notes, handwritten just afterwards), I chose not to pursue that line of discussion, although it was already becoming evident in intelligence available at RAND that a split between the Chinese and the Soviets had developed. (It turned out to have arisen in particular out of the Russian refusal to provide nuclear weapons to the mainland Chinese during the Taiwan Strait crisis of 1958, and their subsequent withdrawal of Soviet nuclear technicians from China.)
I thought I was discovering a parochial bias in the Pacific Command that should be brought to the attention of planners and decision makers at the national level. I was wrong. It was the next year that I learned, in the Pentagon, that President Eisenhower and the Joint Chiefs of Staff shared the admirals’ views entirely. They had no intention, under any circumstances, to shock Admiral Kivette with an order to spare the Chinese—even initially or provisionally—in any war with the Soviet Union.
But by the time I learned that, it had long been clear to me that if the highest authorities did give such an order—if they had changed their minds in a crisis and did, after all, wish operations to exclude China at least initially—it would be virtually impossible to implement that order quickly in the Pacific. That was true for technical as well as bureaucratic reasons. CINCPAC planners were working extremely hard, around the clock each year, just to produce one single plan for nuclear war against the Sino-Soviet bloc, and they simply didn’t have the ability to produce a second plan for war with the Soviet Union alone.
Out of a list of tens of thousands of targets throughout the Sino-Soviet bloc that intelligence had identified as important, the Pentagon ordered lower levels to make plans to hit around a thousand of the most valuable of these in the case of general war with the Communist bloc. The major challenge lower-level planners faced was that many of these targets were co-located: that is, two targets to be hit by two different planes were close enough that the blast from one could knock the second plane out of the air, or, at a much greater distance, blind its pilots.
To avoid this problem—referred to as “interference”—the planners were doing extremely intricate calculations, mostly by hand, to devise routes into the target timed so that the planes wouldn’t be blown out of the air by nearby explosions. (In this era, nuclear warheads were to be delivered primarily by planes. When, decades later, delivery systems mostly switched over to missiles, planners discovered the same problem: the blast from one missile would knock other missiles off course or destroy them.) They were dealing with thousands of targets with multiple weapons on most of them, so they had plans for the planes to weave through a virtual minefield of detonations, timing it perfectly to miss an explosion on this side of the plane, and then one on the other, on and on.
The key to all this was knowing exactly when each explosion was going to go off. Thus, everything had to be timed perfectly, based on how long it took for a crew to get off the ground after receiving an Execute order, how long it took to get up to altitude, the cruise speed of the plane once it reached altitude, and the distance to target. Plans specified that a particular explosion would go off at time-over-target, or TOT (for example, 117 minutes and 32 seconds after the Execute order), and then a nearby explosion would go off 2 minutes and 12 seconds later, and so forth. If everything went according to plan, no plane would be struck down by the explosion from a bomb dropped by another plane; no “fratricide” would occur.
As I read these plans and discussed them with the planners, I quickly noticed several glaring problems with this entire endeavor: obvious and predictable reasons why everything would not go according to plan.
To begin with, I read reports from launch drills all over the Pacific and saw that the difference in times, for different bases, was often hours between sending the Execute order and the actual launches on the various bases, in a plan in which seconds mattered for planes to miss nearby explosions. The problem was not with crews on alert; they practiced getting the planes off the ground a great deal, and could do it within ten minutes of receiving their orders. (Of course, that, too, varied in reality, but this element of the chain was comparatively dependable.) However, that was ten minutes from receiving the order.
The orders were supposed to get to the hundreds of different aircraft carriers and bases throughout the Pacific at the same time, and all the plans were based on the assumption that they would. Yet, as I read the reports of command post exercises—lists of when drill orders went out and when the bases actually received these orders during the exercises—I saw that the actual time when the various bases received their orders often varied by one, two, or as much as four hours. Some bases never received the orders. There were always problems in atmospheric disturbances or in messages getting misdirected or held up in some relay point.
Furthermore, the ability to meet the times in the plan depended very heavily on wind. If the planes were all coming from the same direction, then wind would have little effect; it would either slow all the planes down or speed all of them up at the same rate. However, the planes hitting each target came from different bases; this was deliberate cross-targeting (in case one base had been destroyed). The planes often came at the target from 90- or even 180-degree differences in angle. Thus, whatever the wind was doing, it would affect the two sets of planes totally differently—slowing one down and speeding the other up.
How did the planners deal with the fact that you wouldn’t even know which way the wind would be blowing on the various paths to the targets at the actual time of the real Execute order? There was no way to make arrangements for each possible variation in wind direction and intensity, so their way of dealing with the problem was not to allow for wind at all. They simply assumed there was no wind. This made the plans worthless for avoiding interference.
I pointed these two problems out to a planner once. “Yes, I’ve thought of these problems before,” he said.
“Well, doesn’t that make you question th
e value of making all these calculations and plans?”
“These men are risking their lives flying out there. We’ve got to do what we can to save their lives.”
“But it doesn’t seem that this plan has any chance to save any lives at all. It would save lives only if the execution followed the plan down to the second, and there’s not even the remotest possibility of that happening.”
“Well, we’re ordered to make these calculations, so that’s what we do.”
The complexity of the calculations involved in this (illusory) effort meant that the planners couldn’t make many alternate plans. It took them all year to produce the single yearly-updated plan, and while they paid lip service to the need for flexibility, in practice they were extremely resistant to the idea of allowing for more than one plan: their plan, which included targets throughout China as well as Russia. The thought of even tinkering with their target list—let alone omitting a whole nation from it—sent shudders down planners’ spines whenever I raised it.
Many operations and plan centers and command posts in Okinawa, Formosa, Guam, and Tokyo, and on several carriers and command ships in the Pacific that I visited, had a large map showing nuclear targets. It was their most secret map, usually covered by a screen or curtain when people who lacked authorization (unlike me) were being briefed in the room. These maps, typically, did not demarcate at all between China and Russia. The Sino-Soviet bloc appeared as one giant landmass, with arrows and pins indicating the various targets. You could not tell simply by inspecting the pins whether they were in China or Russia. On some maps, local planners had pinned a piece of colored string indicating roughly the boundary between Russia and China.
This meant that a high-level planner in that division, faced with orders to strike one country but not the other, could not, just by inspecting those targets, decide reliably which ones to pull. In fact, as I learned, doing so would be an extremely laborious process. The computer programs listed tail numbers (which was the way airplanes were designated) assigned to particular coordinates, but they did not list the countries along with the coordinates. Sorting out what coordinate was in which country could not be done in minutes or hours; it would take days or weeks.
Furthermore, on actual runways I visited in Guam, Okinawa, and Korea and on carriers, planes were targeted in a particular fashion on the alert runway—ready to take off on a ten-minute alert. One plane with a 1.1-megaton bomb slung under it was targeted for the Vladivostok area, while a plane next to it on the runway, which would be taking off at a few seconds’ interval, was targeted and briefed and rehearsed for a target in China. Thus, planes within an alert section on a given runway would be entirely scrambled in terms of their national targets. They practiced their drills—which involved complex timing between the different takeoffs—to launch in the same routine sequence. There was no routine for only the China-targeted planes to launch, or only the Russia-targeted planes.
The pilots themselves generally did not know which country they were targeting; the targeting was handled by the crew of an IBM computer system, which did not identify whether the target would hit China or Russia, providing only coordinates instead. Thus, there was no way, either manually or in the computer programs, to quickly unscramble the targets and assure that, for example, only planes 7, 6, and 11, which were targeted for Russia, should take off from that alert force.
All these factors combined to create a situation in which, if we were under attack, it would be simply physically impossible to retaliate against Russian or Chinese targets alone, even if the president ordered his forces to do so.
High-level authorities, wishing to target Russia alone, could, in principle, cut out CINCPAC forces entirely, since all but a few of their bases were focused on China. But they would do that only if it occurred to them that there might be a problem. I never found anyone in Washington who had any idea that there was this kind of problem. No one, as far as I could tell, was aware of it. Few people had access to more than a couple of levels of nuclear war planning, and those who did didn’t take the time to review far lower-level plans or to observe actual implementation provisions in the field. They left these tasks to lower-level commanders (who generally didn’t have access to the higher-level plans).
But as I was to learn, the reason they hadn’t confronted this as a possible problem was not due, as I’d come to imagine, to the idiosyncrasies of the PACOM commanders or their geographic position. It had come from above. As I came to see the highest-level planning in the Pentagon—not available at PACOM and not supposed to be seen by civilian officials—the president and CINCSAC were no more inclined than the Pacific Command to contemplate a war with Russia alone that spared the Chinese.
CHAPTER 6
The War Plan
Reading the JSCP
In the course of my work in the Pacific, I had several discussions with Dr. Ruth M. Davis, who was in charge of computer development for CINCPAC. She was one of the highest-ranking civilians working directly for the military anywhere. When I described some of the puzzles and startling characteristics of the plans I was reading, she told me, in great confidence, of a plan she said I should see if I wanted to understand the nature of U.S. nuclear war planning. It was called the JSCP (pronounced “J-SCAP”), for Joint Strategic Capabilities Plan, and it was on this that CINCPAC’s GEOP was based. She said that the secretary of defense and the president did not know of the nature or even the existence of the JSCP, nor did any other civilian authority. That was confirmed for me by an officer in the war plans division of the Air Force, Lieutenant Colonel Bob Lukeman, who eventually lent me a copy to read in the Pentagon.
To understand how there could be a top-level nuclear war plan of which the secretary of defense had no awareness, it’s necessary to know something of the history of the relationship between the secretary of defense and the military. Prior to 1947, when the National Military Establishment, renamed in 1949 the Department of Defense, was created—combining the Departments of War (Army) and Navy, with the Air Force emerging from the Army as an independent service—there was no secretary of defense. The responsibilities of the secretary of defense gradually evolved over the next decade. Before 1958, the secretary of defense and his assistant secretaries in the Office of the Secretary of Defense (OSD) were seen as functioning essentially in nonoperational areas such as procurement, research and development, personnel, and budget, and not as having responsibilities or command powers in direct areas of combat operations or planning.
Thus, a secretary of defense like Charles Wilson might or might not be in on high-level crisis discussions and decisions, such as the Quemoy crisis of 1954–55. The record of this early period shows that the secretaries of defense were sometimes present in crucial meetings and sometimes not. It depended on their personalities and their relationship with the president. During the whole early era of the institution of this office, the JCS had a basis for saying that the secretary and his subordinate staff had no “need to know” operational war plans, since he was not involved in the operational command.
In 1958, however, the Reorganization Act put the secretary of defense directly in the chain of command, second to the president as a link to the unified and specified commanders and their subordinate commands. (A unified commander was, essentially, a theater commander who, as in the Pacific or Europe, had elements from different services under his command. A specified commander—there was only one, the Strategic Air Command—had units from just one service.) This act cut the JCS out from the chain of command. It was President Eisenhower’s specific intent to do that. He had no respect for the JCS as a body, having dealt with them as Army chief of staff and later as the supreme commander in Europe. He was particularly disillusioned with their postwar performance and wanted to abolish them entirely. However, they were preserved mainly by Congress, which wrote into the Reorganization Act that, without being in the chain of command, the JCS should serve as the “principal military advisors” to the president.
&nbs
p; The secretary of defense at the time of the 1958 act was Neil McElroy, who had been CEO at Procter & Gamble. He was said to be a very intelligent man, but he had no background in military matters and he put in an unusually short workday because he tended to his sick wife. Thus, as I was later told by Air Staff officers, it was relatively easy for the JCS to manipulate him. They got McElroy to sign a Department of Defense directive that reinterpreted the legislated act: “The chain-of-command is from the President as Commander-in-Chief, to the Secretary of Defense, to the Unified and Specified Commanders, through the Joint Chiefs of Staff” (emphasis added). This implied that the Joint Chiefs would be, in some sense, a channel for his directives. They further got him to agree, as a practical matter, to delegate all operational responsibilities to them. In effect, the act and President Eisenhower’s intentions were circumvented. Although it was still on the books, it resulted in no real change of operating responsibilities in 1958 or 1959.
Secretary Thomas Gates, who succeeded McElroy under Eisenhower, had much stronger instincts to exercise control. Yet, with respect to the Office of the Secretary of Defense—comprising the secretary and his staff but also all the deputy and assistant secretaries and their staffs—the JSCP was and remained what later secretary of defense Donald Rumsfeld would have called an “unknown unknown”: something they didn’t know they didn’t know about.
In fact, I was to learn, the JCS had formally adopted, in writing, a set of practices designed to keep the secretary of defense from ever asking any questions directly about the general war plan. The first protective practice was to call the annual war plan the Joint Strategic Capabilities Plan, which did not betray to a layman that it had to do with current operations or, more specifically, with current nuclear war targeting. It was usually referred to by its initials JSCP, but the JCS had issued a directive in writing, which I read, that the words “Joint Strategic Capabilities Plan, or the capital letters JSCP, should never appear in correspondence between the JCS and any agency of the Office of the Secretary of Defense.”
The Doomsday Machine Page 11
