World War II: The Sensory Revolution
Scouting Measures and Countermeasures
Many of the postmortems of World War II express more awe of the sensors that revolutionized naval combat than of the weapons, including the airplane, that fought it. But any treatment of sensors, weapons, and tactics in isolation is artificial; all three together decided battles. To get at the way the new sensors affected tactics and weapon performance during the war, the first step is to establish a framework for discussion. Then we are able to look at radar as foremost among the new scouting devices. Night surface combat of 1942 and 1943 in the Solomons campaign will serve as a premier illustration of radar’s tactical potential and its underexploitation by the American cruisers and destroyers that fought those action-packed battles. Nearly as important was the rise of communications intelligence in World War II. A brief look at the submarine war in the Atlantic, with emphasis on sensors, will show the great extent to which the information war was the decisive element in the defeat of the German U-boat. The full significance of cryptology as an element of scouting measures and countermeasures can be better appreciated now that so many closely held accounts of code breaking in World War II have at last been declassified.
The sensor war was a duel. As the reach of weapons increased, electromagnetic science raced to stay abreast, inventing the means of detecting and communicating at long range. Concurrently, countermeasures to thwart the new electromagnetic technology were being worked on. It is useful to think of the signals put out by these electromagnetic systems as measures that contributed to force effectiveness in either of two ways: aiding in the detection, tracking, and targeting of the enemy (that is, scouting) or in the execution of a commander’s battle plans (that is, controlling). Simultaneously, the force was trying to diminish the effectiveness of the enemy’s scouting and control systems.
One common way to categorize these countermeasures is to divide them into techniques for destroying, disrupting, deceiving, denying, or exploiting enemy signals. The possible actions, by no means exhaustive, may be laid out as in table 5-1.
Table 5-1 is a static display. It gives no sense of pace or timing of signals exploitation, of buying time or intruding in the timing and coordination of enemy action. Except subconsciously, military men do not think in the destroy, disrupt, deceive, deny, exploit framework but leap at once to the possibilities of manipulating their own and the enemy’s specific equipment and command structure. A more useful way to visualize signals exploitation is first to lay out the sequence of measures that must be taken for a fully effective attack, as table 5-2 does.
Evidently some scouting systems can perform more than one step. A surveillance-communications satellite could perform measures A, B1, B2, and B3. One of the advantages of aircraft has been that they can perform the tracking, targeting, attacking, and damage assessment functions in a single flight. By way of contrast, external scouting for surface-to-surface missiles must perform functions A, B1, and B2 before an attack. To ensure full effectiveness, thorough scouting may be necessary. To be sure of hitting his enemy’s carrier or carriers, the attacker may need to know with precision not only the enemy’s position but also his formation. In addition he will need to know whether his enemy has cruise missiles of his own, located on ships other than carriers, that can be used in a punishing counterattack.
Table 5-1. Countermeasures Against Enemy Signal
*In ships at sea, attacking the weapon system also usually meant attacking the signal system. From this derives the current interest in offboard command, satellites, land-based over-the-horizon radar, and other approaches, all of which will complicate tactics.
†Jamming a scouting system usually buys range. Jamming a controlling system usually buys time. Either translates into a sought-after positional advantage.
‡Deception also can be used against enemy weapons. Chaff used against missiles and noisemakers against homing torpedoes are prominent examples.
§Submarines are immune to visual, radar, and infrared detection and sometimes, through quietness, to passive sonar detection.
‖There is asymmetry here, because covert signals like low-probability-of-intercept (LPI) radio transmissions are measures to confound major enemy countermeasures, namely, intercept and exploitation. Since this table’s utility is limited, I have not pursued the logic of the problem further.
#Exploitation may be the most important component of this table, considering the serious possibility that the enemy will detect a scouting system before the scouting system detects the enemy. And radio transmissions may be heard and decrypted. In addition, the threat of counter-countermeasures enters in when the enemy suspects his traffic is being listened to. In that case, he can plant misinformation, taking due care to avoid self-deception. While there is a long history of strategic deception with assorted agents and double agents of espionage to do the communicating, this is not our concern. For an excellent and rigorous study of this subject, see Daniel and Herbig’s Strategic Military Deception.
In principle, the chain of measures necessary for an attack can be broken anywhere by one countermeasure. A successful countermeasure can defeat, delay, or reduce the effectiveness of the attack by any break in the chain. It is not our purpose here to pursue the measures and countermeasures in detail. Suffice it to say that each side strives to maintain the chain from A to C with some combination of redundancy, cover, covertness, cryptology, and sheer electromagnetic power while attempting to break the other’s chain decisively at its weakest link. The sensor war aims to shorten the time it takes to complete one’s own chain, and it aims to lengthen the enemy’s time. A modified table of measures, 5-3, emphasizes the relative consequences of a successful countermeasure.
Table 5-2. Measures Required for Effective Attack
Observe that the countermeasure called exploitation has no place in table 5-3. That is because its effect is different from that of destruction, disruption, deception, or denial. Exploitation enhances one or more of the measures B1, B2, B3, and B4, which permits one’s own attack. Exploitation has its own dynamics and fits into the tactical picture in a way best shown in the force-on-force model of modern warfare described in chapter 11.
For the purpose of seeing the several roles played by the new sensors in World War II and the information war they fostered, let us observe the dynamic processes as signals warfare shot into prominence in World War II.
Radar and radar countermeasures were the most important of the sensory tools that came of age in World War II. In our discussion we should also include the proximity fuze (which was a tiny, shock-resistant radar) and its enhancement of weapon effectiveness. Against aircraft that were closing on a five-inch dual-purpose gun, the proximity fuze not only increased the allowable fire-control dispersion error by two or three times but also simplified a three-dimensional fire-control problem, making it two-dimensional.
The possibility of using pulsed radio waves to detect ships and aircraft occurred to scientists well before World War II. Several countries had secret research under way in the 1930s. In December 1935 the first five radar stations were established by Britain on the east coast of England. Though radar was indispensable in the aerial Battle of Britain, it was not a secret weapon in the sense that cryptanalysis was, that is, a wartime tool unsuspected by the enemy. The presence and significance of radar was quickly appreciated by all sides. Radar and radar countermeasures were part of a fast-paced technological race drawing on vast scientific resources, including, for example, a one-hundred-fold buildup of personnel at MIT’s radiation laboratory.*
Table 5-3. Effect of Successful Countermeasures
As a tool of war, radar was ubiquitous. By the end of 1939 shipboard prototypes were being tested for long-range aircraft detection, antiaircraft fire control, and surface tracking. A series of remarkable breakthroughs came with the British and American collaboration that began in 1940. Centimeter-wavelength radars were ready for production in 1942, and they had sufficient definition to be used for detection of single aircraft, for fighter direction, day or night, and for accurate surface and AAW gunlaying. By 1943 radar was fitted in enough reconnaissance aircraft to have a major influence on search and attack against surface ships, and in enough antisubmarine patrol aircraft to reverse the momentum of the U-boat campaign in the Atlantic. From 1940 on radar was vital to fighter defenses over land, and during the offensive fighter and bomber sweeps before the Normandy invasion it was the key to their effectiveness. For antiaircraft defenses radar was just as important over land as at sea.
Radar quickly became an indispensable navigating tool. It allowed high-speed surface operations in narrow seas, and it came to be so much relied on that when a ship lost her radar at night she was literally and psychologically lost. As a more direct instrument of war, radar started helping guide bombers to their targets over Germany in 1943. Loran was used to aid ship and aircraft navigation everywhere operations were intense.
Radar was developed in the United States, Britain, Germany, France, and Japan. As aircraft offensive and defensive operations became dependent on radar, measures to counter it acquired the highest tactical significance. The first great sensor war occurred in the air over Europe. World War II—specifically, from the time of the Battle of Britain to 1945—offers the best case study of the measures, countermeasures, and counter-countermeasures taken in scouting and weapon delivery (it is better for study than the Vietnam War because analysts have been denied North Vietnam’s tactical picture).
We have seen the big edge radar gave the United States in the air over the Pacific. At the Battle of the Eastern Solomons, U.S. air search detected the approaching Japanese at eighty-eight miles, far enough out that fifty-three fighters—the works—were scrambled with full fuel tanks. Radar gave the United States time to put interceptors in the air after the raid was detected; all fighters were vectored to the attack aircraft without fear of surprise from another quarter.
Because the Japanese were slow to develop radar, the U.S. Fleet had an unparalleled opportunity to exploit its surface-search advantage well into 1943. Even after that, until the end of the war, it was technologically well ahead of the Japanese fleet. In the Solomons, long-range patrol aircraft and coastwatchers gave the United States early warning of the approach of almost all Japanese aircraft and warships. (Cryptanalysis played an insignificant role in this strategic detection since Japanese naval codes had just been changed.)
Because Japanese surface warships had to be well clear of U.S. air cover by daylight, the United States knew within a few hours when they would arrive. In eleven major engagements from August 1942 to November 1943, radar gave the United States the means to detect, track, and target an approaching surface force and blast it out of the water with guns and torpedoes before it knew of the American presence. There was, however, a clumsy lack of recognition on the part of the U.S. Navy that radar offered unparalleled opportunities and required new tactics. And there was a second problem that American tactical commanders had to solve at the same time. The Japanese had a secret weapon too. It was the Long Lance torpedo.
Night Surface Actions in the Solomons
The tactics of the night battles were a competition between these two new tools of war, each capable of exploitation: American radar and the highly lethal long-range torpedoes of the Japanese. In the five battles from August to November 1942 Japanese preparations paid off. The Imperial Navy developed a coherent system of night tactics well before the war and practiced it assiduously. Night action was part of Japan’s prewar recipe of equalizers designed to whittle down the U.S. Fleet before a conclusive battle-line engagement. U.S. Navy practice before the war concentrated on daylight fleet engagements built on the column of capital ships. The American tactical concept was to seek a position that would facilitate (or at least not hinder) capping the enemy T. American tactical training proved counterproductive in the Solomons.
In a strategic context, the Solomons campaign from beginning to end was a contest over airfields on land and flight decks at sea, and for control of the air space around them. For the first six months, from August 1942 to January 1943, the campaign centered on Guadalcanal. In daylight the United States controlled the air around Henderson Field. The Japanese had the same advantage around Rabaul, New Britain, six hundred miles to the northwest. But when the sun set air power lost its grip, and surface combatants met and fought once more. Every night the Japanese threatened to rush warships south through the narrow waters of the Slot between the dual island chains that constituted the Solomons. To the Americans the enemy warships were the fearsome Tokyo Express, which was bent on delivering reinforcements to Guadalcanal or a brutal gunnery bombardment to Henderson Field. Whenever American warships tried to stop the Japanese the result was a lethal surface battle in the blackness of night.
It is well to pause long enough to stress that in naval duels one side or the other nearly always has the complicating problem of a beachhead or a convoy to protect. Every night battle in the Solomons was fought over some objective on land. At the first, Savo Island, the United States was defending the beachhead at Guadalcanal. At the next to last, Empress Augusta Bay, it was defending Bougainville. In the battles between those two the Japanese had the enduring problem of fighting an engagement while trying to reinforce or withdraw a garrison in the Solomons. Two carrier battles, the Eastern Solomons and the Santa Cruz Islands, were tied completely to events on Guadalcanal. The flawed strategy of piecemeal commitment of force in the Solomons gave Japanese tactical commanders the harder task. Keeping reinforcements flowing to sites where ships could not stay was hard, because Japanese warships could only move for a few hours around midnight. It was a sharp disadvantage that American tacticians failed for too long to exploit.
The Japanese did well in the early battles (August 1942 to July 1943) despite their handicaps. This was because:
—The United States failed to grasp that the killing weapon was the torpedo.
—The United States had no tactics suitable for night battle at close quarters.
—The United States was slow to learn. Because of the rapid turnover of tactical leaders, the pace of the battles overwhelmed the Americans.
—Above all, the United States did not exploit its potentially decisive radar advantage—the edge in first detection and tracking that surface-search radar gave and in targeting that fire-control radar gave. While not all ships had both advantages from the start, the radar equipment there was should have been better utilized.
From August 1942 to July 1943 the U.S. Navy suffered from these four shortcomings. From August 1943 to the end of the surface fighting in November 1943 it finally took advantage of the latent potential of radar, using new and compatible tactics.
Phase One, August 1942–July 1943
From the outset Japanese tactics were usually to approach in short, multiple columns, get all ships into action at once, and maneuver in defense against torpedoes. Sometimes destroyers would be positioned ahead as pickets to avoid ambush. On detecting an enemy force, they would close, pivot, fire torpedoes, and turn away. Sometimes they would not fire their guns at all.
The U.S. tactic was to use a long, single, tightly spaced column. The navy expected and achieved first detection and tried to position its column so that all guns would bear across the enemy’s axis of approach, crossing his T. A range of ten thousand yards would, it was thought, be safe from torpedoes and perfect for guns, and if the enemy held to a steady column, the battle would be settled by guns before torpedoes entered the picture (if they did, it would be at a conjectured effective range of under five thousand yards).
But the range closed too fast, the Japanese would not stand still, and their torpedoes were devastating. At first U.S. tactical commanders did not put themselves in radar-equipped flagships, commands were vague or tardy, and the battles were fought at point-blank range, sometimes in great disorder. Later the Americans learned to respect the deadly effectiveness of a massive torpedo barrage against a long, tightly spaced column, but because the actions were closer than they wanted or expected, they didn’t grasp the fact that the Long Lance in barrage was effective at ranges equal to the effective range of cruiser and destroyer guns.*
The Battle of Cape Esperance, fought at the northern point of Guadalcanal on the night of 11–12 October 1942, was the first time the United States could put together a force to take on the Tokyo Express after the debacle of Savo Island five weeks earlier. It illustrates the phase-one tactics of both sides and the tempo of these actions. The United States had a nine-ship column capping the Japanese T, crossing ahead of an approaching force of three cruisers and two destroyers in “perfect” position. The four American cruisers were spaced at six hundred yards between ships, and the five destroyers were at five hundred yards. The light cruisers Helena and Boise had their SG surface-search radars on. In Admiral Norman Scott’s flagship, the San Francisco, the radar was off because it was an older model with a longer wavelength susceptible to Japanese intercept. Let us watch the speed of events as the battle unfolds (see figure 5-1).
2325 The Helena detects at fourteen nautical miles. Spotting planes are in the air but ineffective. The Helena ponders, reports nothing. Seven minutes elapse.
2332 Scott reverses course for tactical reasons that are sound only because he is unaware of the enemy closing at nearly half a mile a minute. But his ambiguous signal results in a maneuver that puts the three lead destroyers directly on the engaged side of his cruisers, racing to regain their van position. Ten minutes elapse.
2342 The CO of the Helena reports the enemy at six miles directly on the starboard beam, a position ideal for gunfire except that, without radar, Scott can’t see his three destroyers. Scott, in a quandary, asks the destroyer division commander where his destroyers are. Captain Robert G. Tobin says he is dead on the starboard beams of the cruisers, right in the middle. He is correct about only two of his destroyers, because the third, the Duncan, is not astern. Having seen the enemy on radar at four miles and believing Tobin is charging them, the Duncan is headed toward the enemy and now in the middle of no man’s land. Three minutes elapse.
2345 The five Japanese ships, still oblivious to their danger, are at two and a half miles and visible in the Helena, which with radar contact knows where to look. The captain asks permission to open fire. The signal is ambiguous—Scott thinks the Helena is asking whether another voice radio transmission is rogered by Scott.* Scott says affirmative and is shocked to see the Helena open up with fifteen six-inch and four five-inch guns. One minute elapses.
2346 The Japanese are more stunned than Scott. Until that instant oblivious to the presence of U.S. ships, they are as vulnerable as U.S. ships were at Savo. They corpen-18 away and for once fail to fire torpedoes.† The range is two miles, point-blank. One minute elapses.
2347 Scott orders check fire, even as all ships but his own flagship open up. The order is not without cause, because the Farenholt and Duncan are in line of fire and will show American six- and five-inch shell holes on their port sides after the battle. Four minutes elapse.
2351 Scott orders resume fire, but most ships have not stopped. His ships have no fire distribution plan and concentrate on what they can see, two already crippled destroyers ablaze. Gunfire distribution is a problem that will never be solved.
The Japanese fled, and no one seems to have suggested a hot pursuit. Four cruisers and five destroyers had totally surprised three enemy cruisers and two destroyers. The United States sank or damaged two destroyers and one cruiser and had two destroyers and one cruiser of its own sunk or crippled. Much of the damage was self-inflicted. Before the Duncan sank, the hits made in her by U.S. gunfire could be seen.
The U.S. Navy counted Cape Esperance a victory; with U.S. firepower potential and advantage of initiative, it should have been an annihilation. In those early days of the sensory revolution the Americans used radar, radios, electronic countermeasures (ECM), and the signal book ineptly. At Cape Esperance they did not fire a torpedo. After that the enemy was never caught so unprepared, so unready to counterattack, or so lightly loaded with the killer weapon, the torpedo.
It was a pickup force the United States had. Morison said the force had no battle plan, but in fact the gun-column was the plan. There was no sense of pace in the decisions made between first radar detection and the time for attack. What was worse, because Cape Esperance was an apparent victory, it seduced the United States into using the same tactics at Tassafaronga and later battles.
Fig. 5-1. Tempo of the Battle of Cape Esperance
Six weeks later the Battle of Tassafaronga exposed all the U.S. shortcomings: the scratch team, inexperienced leadership, and obsolete, tightly spaced single column. Again the Americans had an overwhelming force of five cruisers and six destroyers against eight destroyers, six of which were laden with supplies, and they achieved total surprise, detecting with radar in ample time. Gunfire was opened with radar fire control at the ideal eight to ten thousand yards. But, because of an inadequate sense of timing, torpedo fire was held too long, and the course was held because the Americans supposed they were out of enemy torpedo range. This time they faced the redoubtable Rear Admiral Raizo Tanaka, who had drilled destroyer teams since 1941. His destroyers were trained when surprised to wheel and fire their Long Lance torpedoes. They did, and as a result four of the five U.S. cruisers were sunk or damaged and Tanaka lost only a picket destroyer. The U.S. commander was Carleton Wright, but the plan was not his, for he had relieved someone else two days before the battle. When it was over Nimitz said the lesson of the battle was “training, training, and TRAINING.” But the tactical lessons still slipped through American fingers.
Phase Two, July–November 1943
It was a full year after Savo Island and three battles later that the U.S. Navy got it right. The Japanese continued to operate as before, except more proficiently and with better scouting: they used night air reconnaissance, radar ECM, and rudimentary radar. The United States would take on a tougher enemy, but it would also fight with more skill. First, it had trained units. Second, it trained to sound tactics. Torpedoes would be fired by small, compact divisions of three or four destroyers. Two divisions would be sent in; one would fire and turn away, then the other would fire, then both would mop up with gunfire. When cruisers were present, they were to be kept at a distance of more than ten thousand yards for fear of the deadly Japanese torpedoes. But it was best to let the destroyers do the damage: they had the killing weapon.
1. Detection range: 19,000 yds. Time: 0.
2. All possible torpedoes launched by first unit. Time: +6 mins.
3. Second unit closes, wheels, and launches torpedoes. First torpedoes impact. Time: 12 mins.
4. Second torpedo impact: Move to mop up with gunfire in prolonged stern chases. Watch out for torpedoes!
Fig. 5-2. The Battle of Vella Gulf, 6–7 August 1942
It was at the Battle of Vella Gulf on 6–7 August 1943 that U.S. tactics came together. The navy had its usual foreknowledge of a Japanese reinforcement mission. This time destroyers alone were sent hunting, and Commander Frederick Moosbrugger had the mission, the firepower, the scouting advantage, and the tactics to fuse all three. According to E. B. Potter, Moosbrugger used Arleigh Burke’s battle plan.* It has Burke’s flair. The plan looks like the scheme in figure 5-2. There were two units, both with three destroyers, well drilled and tactically disposed to stay out of each other’s way with the help of radar. They would maneuver with lightning precision. The first division under Moosbrugger carried forty-four torpedoes, and it would fire as many as possible. The second triplet of destroyers under Commander Rodger Simpson, being heavier in AAW batteries, carried only twenty-four torpedoes. Both divisions would move in, bows to the enemy torpedo threat, and wheel and launch their fish. Like Japanese tactics, all this would be done stealthily, without gunfire. After the shock of the torpedo barrage was over, gunfire and aggressiveness could be used in proportion to the damage done, but at that stage the destroyers would have to watch out for the death sting of the now-alerted enemy.
On the night of 6 August tropical rain beclouded the SG radars of Moosbrugger’s short twin columns. The islands around Kolombangara and Vella Lavella confused the radar returns and presented the usual complications of a military operation. Having been spotted by an American night-search plane earlier, the Japanese were on the alert for a U.S. force. Their mission was to reinforce the Kolombangara garrison. They had four destroyer transports, a 2:3 disadvantage. Those were odds they had beaten before.†
Moosbrugger’s force detected the Japanese at nineteen thousand yards. Every ship was told at once. Moosbrugger shifted course right thirty degrees by twin corpens to hit the four enemy destroyers port to port. Gun and torpedo fire-control solutions were cranked in as the unseeing enemy cooperated with steady course and speed. Just seven minutes later the range was four miles. With lookouts estimating the visibility at two miles in the blackness of night, it was time to act. Moosbrugger gave the order to fire, simple and unambiguous, setting all the tactical flywheels of his semiautomatic plan churning in synchronization. The portside torpedoes—twenty-four in all—hit the water from three ships. The firing range was just sixty-three hundred yards for a running range of four thousand yards—as good a setup as could be expected and all one could ask for. A minute later a “turn nine” order came from Moosbrugger. His division executed a simultaneous ninety-degree turn to starboard to clear out, combing the wakes of the predictable enemy torpedo counterattack. At the same time Simpson wheeled his three destroyers to port and bored in.
The men in three Japanese destroyers scarcely knew what hit them. On the U.S. side there was tension aplenty to grip the Americans, who could well remember the chagrin of earlier battles lost during what should have been the mop-up phase. But not this night. As in all good naval battles, the outcome was decided by a feasible plan whose tactical cohesion came from training, good scouting, and the swift thrust of a killer weapon. Three of the four Japanese ships were sunk at the cost of an American gunloader’s crushed hand. Like Nelson, Moosbrugger made sound tactics look easy.
The U.S. Navy had found the tactics to match its radar advantage and neutralize the enemy’s torpedo advantage. It could now beat the Japanese with torpedoes, their own superior weapon. Good sensors and scouting could overcome better firepower. And the United States would win with little ships because they carried the big weapon. The destroyer’s torpedo, not the cruiser’s gun, ruled at night. Hit and move was the answer, not crossing the T; units had to be nimble rather than fixed in a sturdy, steady, cohesive—and suicidal—column.
There would be a setback at the Battle of Vella Lavella on 6–7 October 1943. That was another case of an unblooded commander with a scratch force—a nominal number of six destroyers, in fact only three, facing a nominal number of nine, in fact six, Japanese destroyers. Captain Frank Walker, the American commander, chose to go in with his three ships rather than wait for three more to join him from ten miles away. He had the usual radar advantage, fired fourteen torpedoes at seven thousand yards, then gave the game away by opening up with guns while his fish were running. Staying broadside to the enemy, his three destroyers paid the penalty: two of them were torpedoed and the third collided with one of the victims. Three destroyers were put out of action in exchange for one Japanese destroyer sunk—the one that turned into the U.S. torpedoes instead of away.
But then came the masterful battles of Empress Augusta Bay and Cape St. George, fought by tacticians par excellence Stanton “Tip” Merrill and Arleigh Burke. It is not necessary to recount all the details. Empress Augusta Bay (2 November 1943) looks less decisive in the box score than it was in fact: a Japanese light cruiser and a destroyer were sunk and a heavy cruiser was damaged in collision in exchange for one U.S. destroyer damaged. Merrill’s mission was to defend the landing beach on Bougainville. His force of four light cruisers and eight destroyers was pitted against an enemy that we now know consisted of two heavy cruisers and eight light cruisers and destroyers, equal forces on paper. The tactical plan was for Merrill to keep his cruisers at longer range, about sixteen thousand yards, executing 180-degree turns timed to upset Japanese torpedo-fire-control solutions, while staying between the enemy’s heavy cruisers and the beachhead. To do all this he had to sacrifice gunnery effectiveness, a small penalty because closer in his six-inch guns could not fire long enough to be effective before being put out of action themselves by the Long Lance. The destroyers, four under Burke at the head of the column and four astern under Bernard “Count” Austin, would be the untethered mad dogs charging in and inflicting the torpedo damage.
The three separated formations lost control so the results were mixed. Some of Burke’s four ships scattered and he had to mill around reconcentrating them, at one point firing five-inch guns at Austin. Merrill kept his four light cruisers under tight rein and away from the danger of torpedoes. His ships smothered the enemy with shells and occupied his attention while the destroyers made their charge, doing what moderate damage they could. The American attack so bewildered the Japanese that they turned and got out, abandoning their mission. In the battle the Japanese ships were almost completely ineffective. The U.S. force had learned to survive by lightness afoot. Its gunfire wasn’t much—Morison estimated that it achieved only twenty six-inch hits out of forty-six hundred rounds expended. Night spotting of gunfire proved to be terribly hard; the Japanese later said the Americans were consistently off in deflection. Merrill wisely declined to pursue, because dawn was coming and this time, for a change, it was the Americans who would be subjected to land-based air attack, from Rabaul.
At the Battle of Cape St. George, on 25 November 1943, Burke had the chance to use his patented tactic of hitting with the left and following with the right. He had five destroyers split into units of three and two, supporting one another five thousand yards apart. His nominally equal enemy had two new destroyers escorting three destroyer transports which were trailing thirteen thousand yards (a twenty-minute run) behind. The Japanese commander had no radar on, and his second force was, in effect, a protected, not a mutually supporting, force. Against Burke’s force that was fatal. Burke detected the enemy on radar at eleven nautical miles and adjusted course, and fifteen minutes later at three nautical miles his three destroyers launched fifteen torpedoes undetected. They mortally wounded both lead destroyers. Then he took after the three destroyer transports. Bows to the enemy, his ships evaded a mess of torpedoes in a two-hour stern chase and sank one destroyer. It is fitting that the man who had conceived Moosbrugger’s tactics could finish the Solomons night actions with his own little tactical masterpiece.
How, then, to recapitulate a year and a half of night actions, eleven of which were distinguished as battles? Radar was the new sensor, it had to be integrated tactically, and nighttime provided a magnificent opportunity for so doing; black night was radar’s element, and it should have given the United States a decisive advantage.
In the conditions under which these battles were fought, crossing the T meant very little. The best tactic was to approach on a broad front, bows on (short columns abreast in practice), wheel anywhere within range and fire a barrage of two or three dozen torpedoes, then point sterns toward the enemy’s reply. One of the American errors was to forget that combat is two-sided competition. Line tactics were based on the strength of the broadside, which nominally had twice the firepower that could be unleashed end on. Line tactics overlooked the fact that a beams-to column exposed ten times as much hull to torpedoes as a line abreast pointed toward or away from the enemy. In force-on-force computations the U.S. Navy imposed on itself a fivefold penalty with the line ahead.
There was a no man’s land of at least five miles in which no cruiser belonged. For it had come to pass that with a torpedo barrage, a handful of small ships could destroy more than their weight of the enemy and a superior force according to conventional reckoning. Somewhere in the Valhalla of warriors Jellicoe must have looked down on those dark nights punctured with the violence of the torpedo and with a thin smile shaken his head at the Americans who took so long to learn what he knew in 1916.
In the early battles the United States was foredoomed by pickup forces, thrown together and untrained, using the only tactics surface officers had practiced—those of the fighting column. Perhaps they were the only tactics the Americans were capable of executing at first, in the face of not only the Japanese threat but also the hazards of steaming at high speed in darkness and mostly uncharted shoal waters.
Still, the impression remains that the early tactical commanders did not know better—did not know the importance of their radars and the dangers of a long column. Particularly in the early battles, they seem not to have had the sense of pace to keep control while the opposing forces were closing at speeds of a mile a minute. Tactical commanders sometimes drove their van destroyers until they were in danger of colliding with the enemy before opening gunfire. The Japanese never had that problem. Upon seeing an enemy, they pivoted and launched their spread of torpedoes. Their tactics had coherence. Before the war they built torpedoes into their cruisers, while the Americans took them out in the belief that all modern battles would be settled by guns outside of torpedo range.
From the outset the Japanese tactical commander was up front, usually in the lead ship, after the fashion of the great Admiral Togo. The American commander, in his cruiser flagship, was far back in his single column (in two battles as far back as the sixth ship). When fast action was required, maneuvering from the middle of the column was ineffective. The mess at Cape Esperance indicates the kind of problem that resulted. The placement of the American flagship in the middle was a consequence of tactical good sense that had become obsolete—an example of tradition reigning over an appreciation of new tactical circumstances. In the last five battles of 1943, including Empress Augusta Bay after Merrill turned his destroyers loose, the American tactical commanders were in front and the results were salutary. The Japanese practiced their night tactics in peacetime and knew how to fight from the beginning. To the Americans it became evident only through hindsight.
When I reread the details of these battles I expected to find that each American commander improved by learning on the job and training his units accordingly. Certainly the intense operations in the Slot—and a reader focusing just on the major battles misses the fact that ships were out night after night, patrolling, stalking, engaging in shore bombardment—helped to school and steel leaders and their crews in 1943. But with one exception we cannot find an American tactical commander who fought two night battles and improved. This is a commentary on the rapid turnover of ships and staffs Already noted were the cases of Wright and Moosbrugger, who took command within forty-eight hours of battle. Admiral William F. Halsey, as commander in the South Pacific, never let pass an opportunity to fight and so was always scratching for ships, especially destroyers. No officer ever led more than two battles or commanded exactly the same set of ships. The single officer who fought two battles and improved was Arleigh Burke, and even he was not in overall command at Empress Augusta Bay. We may wonder how it might have been if Burke or Tip Merrill had fought as many battles as Nelson. Burke would have been the best prospect against a tactician he never had the chance to fight, the redoubtable Japanese wizard, the tenacious Tanaka.*
Old ideas of massing force came into question in the Solomons. A smaller force had the firepower at the prevalent short ranges to smash a larger force and survive. The commonplace principle of victory by superior concentration of offensive force had to be held in abeyance by the Americans. No doubt we will again see circumstances in which the wisdom of the Solomons prevails: when small ships armed with many missiles have the firepower to take out more than their weight of opposing force.
It is appropriate to conclude this section with an observation on strategy, specifically the strategic significance of the tactics in the Solomons. There appears to have been a tendency on the part of the Japanese to abandon winning positions (at Pearl Harbor, Coral Sea, Savo Island, Samar) and stubbornly to pursue losing causes (the whole of the Guadalcanal campaign, Tanaka’s dauntless reinforcement of Guadalcanal, the Battle of the Philippine Sea). The successful Japanese night tactics in 1942 were hit and run.* The successful American tactics in 1943 were hit and duck and hit again. American forces were sometimes punished because it was ingrained in them to stay when they came. But what was costly tactically may in the end have paid off strategically.
There is a coda to the Battle of Empress Augusta Bay that says much about the U.S. surface fleet’s ability in 1943 to defend against air attack with radar. After the battle Merrill was well within striking range of the great Rabaul base and knew he would see a full-scale Japanese air attack in the morning, mirroring the American air attacks on the withdrawing Tokyo Express launched from Henderson Field in the Guadalcanal days.
Merrill gathered up his four light cruisers and four of his destroyers in their tight AAW wagonwheel. The attack came, one hundred strong, and if there was ever evidence that a modern surface fleet, concentrated and well handled, could deal with aircraft, this battle was it. Merrill’s ships bristled with guns, for AAW defense was their business. The cruisers alone fired one thousand five-inch and over thirteen thousand forty- and twenty-millimeter shells. The Japanese achieved two bomb hits of minor impact and lost seventeen aircraft. The attack took seven minutes, which meant the cruisers fired about thirty-five shells a second, a withering barrage.† Radar had given Merrill early warning, radar fighter direction steered his inadequate land-based CAP, and (I am guessing) radar proximity fuzes made his five-inch dual-purpose guns the most effective killers in the battle.
That was on 2 November 1943. Three days later Rear Admiral Frederick C. “Ted” Sherman’s carriers Saratoga and Princeton were sent by Halsey to deliver air attacks on Rabaul. Up to this point Rabaul had been forbidden fruit. With at least seventy fighters on the ground, the harbor ringed with antiaircraft guns, and the firepower of seven or more heavy cruisers and a flock of lighter warships, the defense should have been overwhelming. But the Japanese were gearing up for a massive sweep, and Halsey, with not one heavy cruiser to fight a night action, felt he had to risk an air attack. He sent Sherman in to strike with forty-five attack aircraft and fifty-two fighters.* Japanese reconnaissance aircraft blew their report of the force, and Rabaul, without radar warning, was completely surprised. The American strike attacked in the tightest of massed formations and got out with losses, mostly in retirement, of only ten aircraft. They damaged four heavy cruisers, two light cruisers, and two destroyers, and the Japanese abandoned all thought of another night sortie to relieve Bougainville.
To see the significance of radar, one merely needs to reverse the locations and missions of the two fleets. American aviators had asserted their ascendancy over Japanese aviators, but we should not forget the roles played by U.S. warships in accelerating the process and the radar that was so effective in early warning and AAW gunnery.
This book says little about the submarine wars. That is because amphibious operations and air strikes required surface ships, and because fleet actions offered the best chance of controlling the seas. Submarines could deny but not exploit sea control. Submarines were spoilers, and still are that—except in nuclear war, in which case they are intended to play a central role.
Submarine roles are divided into support of the fleet and attacks on shipping in a kind of guerrilla warfare at sea. To support the fleet, submarines scout where other warcraft cannot go and attack and weaken the enemy. These were major roles in the German, Italian, British, Japanese, and American navies.
The Philippine Sea (June 1944) serves as well as any battle to demonstrate the effectiveness of submarines in fleet support. Submarines first sighted the Japanese force and reported its general composition. Before the end of the battle they sank two big Japanese carriers, a better score than Mitscher’s aircraft achieved. If one counts large, light, and escort carriers together, then during World War II aircraft of all nations sank twenty carriers of 342,000 aggregate tons. Submarines sank fifteen carriers of 306,000 aggregate tons. (At Midway the crippled Yorktown was actually sunk by the Japanese submarine 1-168, but I credit the carrier’s destruction to aircraft.) Surface warships sank two carriers, of 30,000 tons total.
In the submarine’s guerrilla warfare against shipping there were three major efforts: the German campaign in the Atlantic, the British campaign to interdict German resupply of North Africa, and the American submarine campaign to isolate Japan from oil and other resources. By any standards all were formidable, and the last-named can be called successful. Even the U-boat campaign in the Atlantic, which resulted in unparalleled destruction of some of the bravest men ever to put to sea, can, as Admiral Gorshkov of the Soviet navy has pointed out, be called a strategic success because of the vast and disproportionate response imposed on the Allies, who had to expend many times more manpower and material to defeat the campaign than Hitler’s navy committed to it. But then, barring certain inefficiencies on their part, the Allies had no choice. Nazi Germany had tremendous leverage and exploited it ruthlessly; the Allies needed control of the ocean’s surface.
The submarines were the latest in a long tradition of raiders at sea. In the most rewarding of all guerrilla eras, Francis Drake and John Hawkins and their Elizabethan compatriots got double profit: they denied to the enemy the fruits of the ships they captured and took their booty to England. Later Raphael Semmes, captain of the Confederate ship Alabama, could destroy but rarely keep his prizes. By World War II surface raiders like the Graf Spee and Bismarck were doomed by cryptanalysis, aerial surveillance, and radar. To escape, twentieth-century raiders had to disappear under water. Throughout World War I and at the beginning of World War II submarines were surface raiders that submerged to evade attack. The significance of this was that U-boats forced under water by aircraft (far enough away from the convoys) were ineffective. In the middle of 1943 it was aircraft that broke the back of the peak effort by U-boats, when on an average day 104 of them lurked the seas. The Allies won the Battle of the Atlantic by a combination of offensive air patrols in the Bay of Biscay and defensive air patrols around convoys. Aircraft with radar first slowed U-boats in transit and then drastically curtailed their maneuvers in the vicinity of the convoys.
Radar was essential to the Allied effort. And the search radars stimulated one of the first big measure-countermeasure duels in the scouting business. We cannot recount here how the British kept ahead of German detection devices by changing frequency, but it is an instructive story. British operational researchers learned how to gather data that told them whether the U-boats were on to Allied radar frequencies.* In fact such analysis was not necessary. The innermost circles of Allied command knew already because the British had cracked the German code.†
Exploitation of the U-boat cipher—the Ultra secret—was the most important weapon of sensory warfare in the Battle of the Atlantic. Admiral Dönitz directed his U-boats from ashore. By 1942 wolfpacks were being used to scout and concentrate attacks on convoys. Since the U-boats could not communicate freely, Berlin played the role of tactical coordinator, opening up an opportunity for the greatest of all tactical signals exploitation. In May 1941 the British had pilfered a German cipher machine from the U-110 and started to read (intermittently) the enemy’s signals. In addition, the Allies triangulated U-boat positions with RDF to supplement cryptanalysts. German submariners thought that very short, technologically sophisticated burst transmissions would be impossible to use for triangulation; they were wrong.
The deciphered information was of the highest strategic importance, of course. Intelligence gave an exact count of the U-boat fleet, in both the Atlantic order of battle and the Baltic Sea during shakedown training. The code breakers provided bountiful information about the movements of U-boats and their plans of attack, ordered from thousands of miles away in Berlin.
Another incalculable benefit of having been able to read Hydra [the cipher initially used by all operational U-boats] for so long was the insight which it had given us into the way the U-Boat war was being conducted, and perhaps even into the way that Dönitz’s mind worked. We knew the U-Boat’s methods, the average speed of advance when proceeding to and from patrol, the endurance of the various types of U-Boat and characteristics of their many commanding officers, the types of patrol lines favoured and the exact meaning of the short signals used for making sightings, weather or position reports.*
No wonder that superb American tactical study, ASW in World War II, published in 1946 and at the time classified, is so rich in detail—it even includes the names of German Aces.† The wonder is that the secret of Ultra was kept so well. In this official study virtually the only mention of code breaking is in connection with the capture of the U-505 in June 1944. The authors write that it gave the Allies important information about the German codes. The truth is that the USS Guadalcanal and her escorts were able in quick succession to dispose of four U-boats, one of which was the U-505, because the Allies already possessed the code and knew where to send Admiral Daniel Gallery’s hunter-killer group.
From what we know now about the influence of code breaking in the Atlantic and the Pacific it is reasonable to infer that:
—A guerrilla campaign at sea—a modern guerre de course—that is not covered will fail because of the capabilities of modern surveillance.
—Due to overconfidence in high places, signals from shore to sea are especially vulnerable. Operational command from afar, which is tactical and reveals battle information, requires special signals discipline.
Tactical Interaction between Land and Sea Forces
We come to the last important factor regarding sensors, scouting, and countermeasures. As emphasized in chapter 1, the strategic interplay of events ashore and at sea has always been the major determinant of the scene and scale of most naval battles as well as the opponents’ aims. A major development in World War II was the added importance of tactical interaction, owing primarily to the new role of aircraft. In this brief section let us consider the effects the sensory revolution, which opened so many new possibilities, had on the information war and on naval command.
For the first time we see how tactical command was exercised from ashore to a significant degree: Yamamoto, Dönitz, Nimitz, and Halsey all participated to greater and lesser extents in the battle movements of their forces. In the record of Halsey’s signals from ashore in Noumea may be found orders to commence South Pacific operations at explicit latitudes and longitudes at specific times. Tactical command and, even more commonly, what in current parlance is called operational art were exercised from ashore so that the striking force of ships and aircraft at sea could keep radio silence until their presence was discovered by the enemy. Even when cryptanalysis and RDF could not be exploited, traffic analysis of the volume of signals revealed impending operations. Often when the text could not be decrypted the address headings could, revealing the commands and ships involved.
The Japanese and Americans both used land-based reconnaissance aircraft, partly to help conceal the location of ships and aircraft at sea but also because land-based patrol aircraft had very long range and endurance. The Japanese, more than the Americans, used external reconnaissance to husband carrier assets for the attack. In the Mediterranean one of the most serious deficiencies, which sapped the Italian navy’s confidence and morale, was the failure of land-based air reconnaissance.
I hardly need to comment again on the new reach of sea-based air attacks against shore targets. What might be noted is the fate of land-based attacks against targets at sea. Both the Italian and American air forces were supposed to attack warships at sea, and they largely failed. The Italian fleet, having been denied a naval air arm by Mussolini, had to depend on reconnaissance by the Italian air force and was grievously crippled by inept support. The Japanese, however, flew naval aircraft from fields ashore with success, most notably sinking HMS Repulse and HMS Prince of Wales with torpedoes. There is nothing inherently wrong with attacks launched from ashore against warships if the aircraft are trained and armed for the mission. Lack of mobility and the power to concentrate were the tactical constraints on the strike effectiveness of land-based aircraft in World War II. Weak command structure and neglect of the special training required to hit maneuvering warships were the main (and unjustifiable) reasons for the widespread failure of air strikes from land to sea.
Land-based maritime patrol aircraft proved highly effective against submarines because they could sortie on solo patrols safely and at long range. Due to the nature of their tactics they could be large and plodding. The British were slow to see their possibilities. After HMS Courageous was sunk by the U-29 in the first week of the war, the Royal Navy failed to exploit the fact that land-based aircraft were a safe way to fly against U-boats. For three and a half years the British demurred, until the exigencies of the war in the Atlantic drove them to it. The turning point came in early 1943 with the transfer of bomber command squadrons to coastal command. At the same time the British persuaded President Roosevelt to earmark new U.S. Liberators, then starting to come off the production lines in large numbers, for ASW patrol, a mission that suited the characteristics of those remarkable, long-ranged aircraft.*
Nazi Germany missed a golden opportunity to exploit land-based aircraft at sea. After the fall of France, a few sorties by the Luftwaffe demonstrated that its medium-range bombers could attack Allied Atlantic convoys effectively. But Hermann Göring’s penchant for attacks on land targets eliminated the possibility of developing and committing German aircraft to shipping attacks in great numbers. That Germany might wake up to the opportunity was a worry haunting the harried Royal Navy through much of the war.
A separate book is needed to put amphibious operations in perspective. They have usually and rightfully been studied for their strategic content. But unquestionably the reach and range of weapons and the maneuverability of ships and aircraft changed the nature of amphibious assault between the Napoleonic Wars and World War II in fundamental ways that are almost impossible to exaggerate. Amphibious landings since World War II have demonstrated the growing land-sea interface and have made use of more new tactics. We may recall the stunning operation at Inchon, the landings at Wonson that were confounded by minefields, the recent event in which the British destroyer Glamorgan was struck by land-based missiles from East Falklands Island, America’s swift use of aircraft and warship mobility in the taking of Grenada, and most recently the American amphibious demonstration and threat of an assault in Kuwait during Desert Storm operations.
The multifaceted growth in the potential of land- and sea-based forces to operate against each other deserves and has received careful study. Not to be underestimated also are the roles of scouting, communications, the control of forces, and the countermeasures against them that have risen in importance and will continue to do so. The sensory revolution was the cause of all this. The growth of land-sea tactical interactions and of sensor technology are two of the great trends in tactics.
* The number increased from forty to four thousand (Brodie and Brodie, p. 209).
* A quick example of torpedo fire: Japanese spreads were thrown in the general direction of U.S. gun flashes (sometimes search lights). A column of eight ships was about four thousand yards long. At a distance of four to eight thousand yards, the torpedo spread was more than likely to be inside the end points of the target column. Ships were over one hundred yards long and spaced at five to six hundred yards, so about one weapon in six would hit. With as many as thirty torpedoes in the water, unseen and unsuspected, the results could be, and were, lethal, especially since one hit on a cruiser or destroyer nearly always achieved a firepower kill. McKearney’s study (1985) found that the average torpedo firing range over the entire campaign was eighty-five hundred yards, which made the running range on the order of seventy to seventy-five hundred yards (p. 154). He concluded that the average hit probability for all engagements was .06, but at the battles of Tassafaronga and Kula Gulf torpedo hit probabilities approached .20 (compiled from appendix A, pp. 188–246). Not all U.S. columns were as long as eight ships, not all torpedoes would have been reliable, and toward the end of the campaign the Americans learned to comb the tracks, so his numbers corroborate the theoretical estimate.
* Interrogatory Roger was the ambiguous signal book query.
† A corpen is a follow-the-leader maneuver. Corpen-18 means to reverse course by turning in succession 180 degrees to starboard.
* Potter, p. 313.
† However, neither side had very accurate estimates before that battle or any other.
* In fairness it must be pointed out that the Japanese also fought with cruisers and destroyers that had not operated together. But doctrinal integrity and much practice in night operations kept them cohesive.
* This is not to deny Japanese courage in stopping to rescue survivors, which was done at great peril. Japanese tactics conformed to the prewar strategy of whittling the American fleet down to size.
† S. E. Morison, vol. 4, p. 321. Note that the cruisers fired over eight hundred shells for every enemy aircraft splashed.
* Observe the fighter-heavy ratio. At Halsey’s express orders, Sherman sent everything he had. His carriers were supposed to have been covered by land-based fighters.
* Blackett, pp. 222–23. Among the several accounts of this electronic duel, one of the best and most concise is in Tidman, pp. 75–80.
† By 1943 American operations analysts had deduced this. As Tidman reports, Jay Steinhardt calculated that locations purportedly based on RDF (radio direction finding) fixes were ten times more accurate than analysis showed they should have been. He took this puzzle to Philip Morse, head of the ASW operations analysis group, who confronted his naval boss with the data. The cat was out of the bag, and Morse and Steinhardt, at least, were told the truth. Yet there was never a whisper of the secret in the vast outpouring of operations analysis literature.
* Beesly, p. 116.
† Sternhell and Thorndike, pp. 4, 10, 11, 20, and 81.
* Blackett, p. 227.
