One way to find out the nature of critical tasks is to look at ways in which the defeated forces actually came close to achieving their objective, in this case the vigorous defense of Oahu against air attack. As we argued above, failure is never homogeneous, and two aspects of the Pearl Harbor attack bring this point out particularly clearly. First, two of the twenty or thirty army air force pilots who got off the ground that day attacked a swarm of Japanese airplanes over Haleiwa airfield: they had little success, but Haleiwa became the only field to escape serious Japanese attack. Even relatively limited air action by fighter aircraft showed that the assault could be disrupted. Second, and more striking, is the discrepancy between the performance between navy and army antiaircraft fire. Whereas the batteries on all navy warships were in action between five and seven minutes after the first attack had begun, only four out of the army’s thirty-one batteries engaged the Japanese at all, many of the remainder not coming on line until noon, several hours after the raid had ended.51 These discrepancies are not only puzzles that call for a solution, they indicate something of what might have been achieved by way of a successful defense of Pearl Harbor.
Once guns were manned and airplanes aloft, the Americans managed to disrupt or at least sharply diminish the force of the second Japanese attack. Neither the quality nor the quantity of American weaponry or skills was grossly deficient: A vigorous defense was entirely possible. One critical failure, then, was that of alertness of the active defenses of Pearl Harbor, and by alertness in this case we mean the ability to come into play within minutes, or perhaps as much as a half hour of warning. A second critical failure was passive in nature—the absence of certain measures that might not have increased Japanese losses but that would have limited the damage that the raiders could have done. This includes the dispersal of aircraft (which were lined up wing tip to wing tip in order to enable protection against sabotage) and the burial of the vulnerable oil tanks, which luckily were never hit. Two measures in particular were missing: barrage balloons (sausage-shaped balloons tethered to long wires) and antitorpedo nets to shield important warships. The former would have complicated immensely the task of low-flying Japanese airplanes; the latter would have rendered useless one of their most lethal weapons, the torpedo. Interestingly enough, to the end Japanese intelligence agents in Hawaii received insistent queries concerning the presence of either of these defensive measures—and for good reason.
The success of the Japanese attack depended on their ability to use torpedoes launched by aircraft in shallow waters and accurately to drop bombs on “point” (as opposed to “area”) targets. Given the nature of air-launched torpedoes, the limitations of contemporary aiming devices, and the relatively small bomb loads of Japanese aircraft, this could only be achieved by low-level flying, with all its many hazards. The chief vulnerability of the Japanese plan lay not in the danger of discovery, which the Japanese understood and accepted, Admiral Isoruku Yamamoto warning task force officers that they might have to fight their way into Pearl Harbor.52 Rather, the danger lay in the possibility that American defenses would force them to operate in such a way that the combined carrier task force would do little harm to its chosen targets and lose its scarcest resource of all—the crack pilots of the Japanese Naval Air Force.
The critical American failures—to maintain an adequate level of alertness and to have in place an appropriate air defense—were intimately linked. It was not a matter of simply having patrols out, although it certainly was a major failing of Admiral Kimmel’s not to have ordered limited reconnaissance to the north and northwest of the islands—an area long recognized as one vulnerable to an enemy approach. Rather, Pearl Harbor lacked an effective air defense system, even if most of the component pieces of machinery (antiaircraft batteries, fighter planes, radar, and so on) were in place. No central operations room controlled the air space over Oahu, for example. Thus, when the radar operators at Opana Point detected the approaching Japanese at 0702 on the morning of December 7, their report to the virtually unmanned information center was interpreted as a flight of B-17s coming in for refueling on the way to the Philippines. Had the radar plot been appropriately interpreted, however, only the fighter planes of the Fourteenth Pursuit Wing would probably have been alerted—and they were on a mere four-hour alert. The information center had no responsibility for alerting the navy afloat or ashore.
The subsequent Pearl Harbor investigations revealed that the army depended for its information primarily on navy long-range reconnaissance for warning, radar being regarded as a new and unreliable device. Yet the army had no idea what kind of reconnaissance the navy had or would implement. At the same time, the navy, whose installations were to be protected by army fighter planes and antiaircraft guns, did not control the army’s level of alert or even understand that General Short had chosen the lowest level of alert, which only covered antisabotage precautions. The navy was supposed to control air operations against enemy ships heading toward Pearl Harbor, the army air force to coordinate air operations overland—using airplanes allocated by separate commanders more or less as they saw fit.53 This lack of communication and coordination within Hawaii has not received the same attention as the lack of communication between Washington and Hawaii, yet in the end it proved the more dangerous.
LAYERED ANALYSIS
Having identified the critical failures we can begin analyzing the behavior of different layers of organization and command. We look for the interactions between these organizations, as well as assess how well they performed their proper tasks and missions. By so doing we help avoid the pitfall of “horizontal history” and lay the groundwork for the concluding pan of our study. There is no formula for selecting the right levels to examine; that depends on the case at hand. Echelons of command or organization have varying importance depending on their military mission and chiefly on the nature of the war in question. A guerrilla conflict, for example, may give no scope to the middle echelons (battalion or brigade): Instead, the actions of small-unit leaders and regional commanders may be the most important. In more conventional forms of warfare, however, these middle echelons may prove the critical ones.
In the case of Pearl Harbor four levels stand out, the first two of which can be subdivided in turn. The first of these is the high command in Washington, which can be broken down into the civilian echelon (the president and the secretaries of navy and war) on the one hand and the military echelon (the chief of naval operations and the chief of staff of the army) on the other. The chief responsibilities of these echelons were to provide adequate resources to the Hawaiian commands, which they did. The chief question raised over their performance has to do with their effort to warn the local commanders of the impending Japanese attack.
The question of warning at Pearl Harbor is an extremely tangled one, about which many books could be and have been written. The following points, however, are central. First, Washington had no timely and unambiguous warning of a Japanese attack on Pearl Harbor, although, like some of the local commanders, it had worried about just such a surprise attack for over a year and had communicated those concerns to Kimmel and Short. Chief of Staff George C. Marshall, for example, had written the following to Short on the day the latter took over his command:
My impression of the Hawaiian problem has been that if no serious harm is done us during the first six hours of known hostilities, thereafter the existing defenses would discourage an enemy against the hazard of an attack. The risk of sabotage and the risk involved in a surprise raid by Air and by submarine, constitute the real perils of the situation.54
On the other hand, in the weeks before December 7 both Stark’s and Marshall’s offices withheld, for a variety of reasons, certain important intelligence items from Hawaii (most notably a Japanese request to its local agents for detailed descriptions of where American warships were moored in Pearl Harbor). However, on November 27 Kimmel was alerted with a warning that began, “this dispatch is to be considered a war warning,” and Sh
ort received a similar message from General Marshall. In addition, both commanders were generally apprised of the collapse of negotiations with the Japanese and the official view that a major Japanese aggressive move was in the works. The evidence suggests that thereafter neither Marshall nor Stark monitored closely the defensive measures undertaken by local commanders.
The second echelon is that of the major commands in Hawaii, which may again be subdivided. In one category is the commander in chief, Pacific Fleet, who was also the commander in chief, U.S. Fleet, who had overall control of navy forces in the vicinity of Pearl Harbor. In the second category was Rear Admiral Claude C. Bloch, Commandant of the Fourteenth Naval District (COM 14 in navy jargon), who had responsibility for all shore installations in Hawaii, and the commanding general of the Hawaiian department, Lieutenant General Walter Short. This subdivision indicates already some of the problems under which American forces in Hawaii labored: although Short’s nominal counterpart was Admiral Bloch, he worked more closely with Admiral Kimmel, who had the real authority over the deployment of naval forces. It is at this level that decisions were made concerning the nature of military alerts to be adopted in peacetime—both the structure of such alerts and their actual implementation. The critical decision here was Short’s decision to order a level one alert (against sabotage only) rather than a level two (all measures in one, plus precautions against enemy air, surface, and submarine action) or three (all-out attack). Kimmel had placed the navy on what was known as a “modified level 3 alert,” which required partial manning of antiaircraft weapons and was the lowest of the navy’s three alert levels.55 It should be noted that not only did the navy and army have different alert procedures: Kimmel did not even know that the army had more than one kind of alert.
Below the level of the higher commanders in Hawaii were the component commands there, and in particular those of the Army: the Hawaiian army air force and the Hawaiian coast artillery, which controlled most of the fighter planes and antiaircraft artillery respectively devoted to the defense of the base. These commands, which might have been expected to provide the bulk of the coordinated defense of the island—the vectoring in of fighters to intercept the Japanese, the activation and supply of antiaircraft batteries, and on the navy side, the sustained reconnaissance by long-range patrol aircraft of the commander, Naval Base Defense Air Force—exhibited the most spectacular failures in the attack, as we have already discussed.
Fourth, and finally, we have the units at the sharp end, the battleships and fighter squadrons, the antiaircraft batteries and support units that most keenly felt the effects of the air raids. Here it should be noted that the critical failure was that of alertness. Within these units, particularly on board ship, the response to the attack was remarkably fast and, within the limits of practicality, effective.
THE ANALYTICAL MATRIX
Having described the critical failures, and selected the chief layers of command, we can now represent the problem graphically as in Figure 3–1.
PATHWAYS TO MISFORTUNE
An exercise such as this, though it may oversimplify some of the relationships between failures, is a handy device for analyzing military misfortunes. If, after looking at the chart, we draw arrows indicating relationships between various failures, we can define pathways to misfortune. We notice several things about the resulting picture. First, the critical pathway to misfortune comes in the column headed Coordination—it is through the failure of coordination at levels three through five that American forces found themselves at such low levels of military alert. Then we note a secondary pathway from box 2.1 through box 3.1 to box 4.2—the misleading nature of the warning sent out at a rather higher level, and its liability to misinterpretation by the rather literal-minded commanders on the scene.
It is in the first pathway—that stemming from failure of coordination, however—that we may find the most important explanation of the Pearl Harbor disaster. Reflected in that disaster is not simply the unpreparedness of a pacific people, or the narrowness of inflexible commanders, though both elements were present in some degree. Far more serious and dangerous was the absence of a harmonious garrison on the island of Oahu, organized and directed by a single unifying scheme, under a single commander. Had all of the forces in Hawaii been under the vigorous operational control of a single commander—most likely the commander in chief, Pacific Fleet—it is unlikely (though not inconceivable) that multiple alert system would have existed side by side, that the air above the islands would not have had a single center controlling it, or that Oahu’s fighter defenses would have assumed that long-range patrols would give them four hours’ warning of an enemy’s approach. In fact, within weeks of the debacle at Pearl Harbor the chief of staff of the army and the chief of naval operations ordered the establishment of joint commands and joint operations centers in all areas where navy and army had to work side by side.56
Arrows indicate causal links. Solid lines indicate primary pathways; dashed lines, secondary pathways.
FIGURE 3–1. Matrix of Failure
Stark and Marshall had attempted to make such arrangements in the days before Pearl Harbor, but they met the resistance of those on the spot, not only there but in other commands as well. Rejecting (with Short’s concurrence) the recommendations of both his superiors and a visiting Royal Navy captain (later Admiral), Louis Mountbatten, for the creation of a joint operations center, Kimmel wrote on November 3, 1941 that the army and navy had different tasks. “Strategic, rather than tactical cooperation, is indicated and therefore the necessity for rapid receipt and exchange of information and arrival at quick decisions is of less importance.”57 Let it be noted that army and navy relations on Hawaii were nothing if not cordial: Kimmel and Short, in particular, had friendly relations, playing golf together regularly and pledging themselves to a system of military coordination by mutual cooperation. But as the authors of the Report on the Pearl Harbor Attack concluded,
The evidence adduced in the course of the various Pearl Harbor investigations reveals the complete inadequacy of command by mutual cooperation where decisive action is of the essence. Both the Army and Navy commanders in Hawaii failed to coordinate and integrate their combined facilities for defense in the crucial days between November 27 and December 7, 1941. While they had been able over a period of time to conceive admirable plans for the defense of the Hawaiian Coastal Frontier consistent with the system of mutual cooperation when the time came for the implementation of these plans they remained hollow and empty contracts that were never executed. . . . The tendency to “let George do it,” to assume the other fellow will take care of the situation is an inseparable part of command by mutual cooperation.58
The result, the congressional committee found, was the conduct of operations in a “state of joint oblivion.”
THE MISSING DIMENSION OF STRATEGY
The congressional committee investigating the Pearl Harbor disaster was, perhaps, closest to the mark of all the studies before or since in its focus on the fragility—and in some cases, the absence—of army/navy communication and cooperation on Oahu in the months before the Japanese attack. By dissecting military misfortune in the way demonstrated in this chapter, we find our attention drawn repeatedly to what one might call “the organizational dimension of strategy.” Military organizations, and the states that develop them, periodically assess their own ability to handle military threats.59 When they do so they tend to look at that which can be quantified: the number of troops, the quantities of ammunition, the readiness rates of key equipment, the amount of transport, and so on. Rarely, however, do they look at the adequacy of their organization as such, and particularly high level organization, to handle these challenges. Yet as Pearl Harbor and other cases suggest, it is in the deficiency of organizations that the embryo of misfortune develops. And it is to the varieties of organizational disfunction in war that we now turn.
Failure to Learn
American Antisubmarine
Warfare in 1942
ONSLAUGHT
IN THE LATTER HALF OF JANUARY 1942, more than a month after the beginning of World War II, German U-boats began an assault on coastal shipping in the immediate vicinity of the United States. During the next nine months, an unparalleled massacre of American shipping took place within a few hundred miles of the coast and often within sight of the American waterfronts. Shipping in the coastal zones (referred to as Sea Frontiers by the United States Navy) suffered far heavier losses than did the main North Atlantic convoy routes—which, indeed, were successfully guarded throughout most of this time. For the first and indeed the only time in the war, Grand Admiral Karl Doenitz, commander of the German submarine force, came within appreciable range of achieving the gross tonnage losses—some 700,000 tons a month—he believed would make it impossible for Britain, and hence the Allies, to prosecute the war actively.1 During the period January-September 1942, in fact, the Allies and neutral nations lost to all causes an average of 650,000 tons of shipping monthly—more than twice the monthly average in the earlier phases of the war.2 During this time, as indeed during the earlier stages of the war, sinkings continued to outrun new merchant ship construction by something like 200,000 tons a month, despite steady increases in American shipbuilding capacity.
Of the 650,000 gross tons of losses mentioned, half or more occurred in waters nominally under the control of American forces. In March 1942, for example, the Eastern and Caribbean Sea Frontiers, together with the Bermuda area (where the United States had naval and air bases), witnessed losses of over 380,000 tons of merchant shipping—this during a month when total Allied and neutral shipping lost 840,000 tons to all causes (including accidents, mines, surface raids, and so on) in all theaters of the war, including the Pacific. And when we look at losses to German U-boats in the Atlantic alone the statistics become even more dramatic: From mid-January until September 1942 losses in American coastal waters accounted for more than half of total shipping losses in the Atlantic, and often far more. In May, for example, more than 85 percent of the merchant shipping losses suffered by the Allies occurred in American eastern and southern coastal waters. We can judge the magnitude of these losses by comparing them with the heaviest attacks on the North Atlantic convoys, which occurred in March 1943. During that month U-boats sent nearly 300,000 tons of Allied shipping to the bottom; during most of 1942, however, losses in American coastal areas exceeded this figure month after month.
Military Misfortunes Page 8