The technical characteristics of weapons systems had an important influence on the fighting off Guadalcanal. In the interwar period, the firepower of surface ships had dramatically increased. Improvements in fire control—including more advanced models of the Ford rangekeeper—had made gunfire and torpedo salvoes more accurate; advances in fuse and shell technology had made shells deadlier; and new torpedoes had increased lethality. These enhancements combined to make ship combat more intense than ever before. However, the maximum effective range of these weapons was still limited by that of the human eye.
The upper limit of visual range at night off Guadalcanal was almost always within ten thousand yards; often it was within five thousand. At the battle of Savo Island, the Japanese opened fire on the northern group of Allied cruisers at about ten thousand yards, using searchlights. During the decisive portion of Guadalcanal II (sometimes “the Naval Battle of Guadalcanal”), the battleship Washington engaged the battleship Kirishima at 8,400 yards with the aid of radar ranging and star-shell illumination. At Cape Esperance, the Japanese were sighted at five thousand yards, and fire was opened almost immediately. Guadalcanal I descended into a melee because of the low visibility. The leading ships of the opposing formations first sighted each other at three thousand yards; fire was opened soon thereafter.
The close engagement ranges, combined with the lethality of the ships’ weaponry, explain why the combat off Guadalcanal was so furious and deadly. They also illustrate the importance of hitting first. Surprise was decisive in the night surface battles, a better predictor of victory than any other single factor. The other hallmark of these fights was extreme confusion. The Navy had anticipated that night combat might be chaotic. A 1938 doctrinal manual for light cruisers warned that night battle would develop quickly and be fast-paced, with only brief opportunities to score hits on opponents as they appeared out of the darkness. This emphasis found its way into drills and practices, which stressed getting on target extremely quickly. Hitting first was important in daylight action; in night combat, it was decisive.
Night battle practice—scores in which made up a significant percentage of a ship’s individual gunnery merit ranking—reinforced this emphasis by training crews to acquire and engage targets in the shortest possible time. The highest scores went to the ships that got on target the fastest and scored the most hits.6 The heavy cruiser Augusta gave an exceptional performance in March 1937 while flagship of the Asiatic Fleet:
A total of thirty-eight hits, or 42.2% on run one with the main battery, of which fifteen were early hits obtained in the first six salvoes in three minutes and ten seconds can be considered little short of remarkable. . . . The target diagram of hits shows the high degree of accuracy and consistency in both range and deflection, indicating extreme thoroughness of the director check and battery line-up as well as the accuracy of the fire-control procedure and spotting.
Augusta’s ability to get on target quickly was aided by her ranging technique. She fired “a three salvo ranging ladder in 500 yard increments by individual turrets” to determine more accurately the target range. That is, the first salvo would have been fired at a range five hundred yards beyond the estimated range, the second at the estimated range, and the third five hundred yards below that. Spotting from these three salvoes would have provided an accurate target range very quickly. In effect, she used her gun battery as a rangefinder.
Lloyd M. Mustin, a future admiral, served on board Augusta during this time and described how her gunners achieved such impressive scores: “I guess there were really two keys to our whole approach of taking a target under fire at night, both much interlaid with each other. One was heavy emphasis on the utmost speed in opening fire . . . [using] an estimated range. . . . [The other was] much more streamlined internal procedures that [permitted] . . . a minimum of required transmissions back and forth between the Captain on the bridge and the man who had the firing key in his hand.” Augusta’s approach of using the guns to find the range became standard procedure. Reports of night battle practices stressed that rough estimates of enemy course and speed were all that could be made before opening fire. Hits were to be obtained by firing immediately and correcting the fall of shot; there was no time to develop a model of the target’s movements. This was the opposite procedure from that used in daylight, when firing was deliberate, controlled, and based on precise calculation of the target’s future position.
Firing without a precise solution would introduce errors. Through experimentation, the Navy discovered that the best way to secure a large number of hits was a “rocking ladder.” Once they found the range, Mustin and his contemporaries altered the range up and down in slight increments with each salvo, “rocking” the shells back and forth across the target. The variation in range compensated for errors in the fire-control solution and increased the number of hits. At night, it was extremely easy to believe that shells were on target when in fact they were slightly short or long. Continually rocking the shells back and forth overcame this problem and became the prescribed doctrine for night firing. The results were impressive; Mustin described the target raft after one practice as a “shambles”—it had been “literally shot to pieces.” The emphasis on opening fire immediately was well worthwhile; all the night surface actions of the campaign—except for the last one—were to be decided by gunfire.
Radar too played an important role in the night battles of 1942. The Navy’s willingness to experiment allowed it to realize quickly the potential of using radio waves to develop a more detailed picture of the surroundings. However, the radars used at Guadalcanal were fairly primitive. Most displays were unsophisticated, and effective procedures to harness the information they provided had yet to be developed. Radar was not yet a transformational technology, but in the right hands, it was an extremely useful tool.
Radar research and development was performed under the auspices of the Navy’s technical bureaus. There were two parallel threads. Search radars, designed to provide early warning of approaching ships and planes, were the responsibility of the Bureau of Ships. BuOrd focused on fire-control radars, which could provide range, bearing, and other information necessary to augment fire-control solutions. The Navy’s heuristic emphasizing quick and effective gunfire led to the rapid development of the latter.
Search and fire-control radars used similar technology but were employed differently. Search-radar antennas were typically kept rotating so that they could scan the area around the ship. When an unknown echo was observed, the earliest installations—like the SC radars in use in 1942—had to be stopped so that range and bearing could be estimated. When its antenna was stopped, the radar could not search other sectors, increasing the chances that the operator would miss approaching contacts.
Fire-control radars were occasionally used to search like this, but their specialty was specific targets. The open architecture of the fire-control system made it easy to integrate radar into it; there was no need for major modifications to take advantage of radar information. Fire-control radars could augment existing fire-control procedures by providing the range to the target, reducing the need for ranging ladders and allowing the target to be “straddled” more quickly.
The Navy’s most modern fire-control radars, the FC (Mark 3) and FD (Mark 4) were used extensively off Guadalcanal. The FC was designed for surface fire control; the FD was similar but configured differently to allow it to be used against airplanes as well as surface ships. Both used lobe switching—alternating transmission in two lobes on either side of the antenna—to improve directional accuracy. Although some ships considered the FC and FD sufficiently accurate for “blind” firing, fire control was most accurate when using radar ranges and visual bearings. This helped ensure that the engagement ranges at Guadalcanal were short.
Initially, the standard radar display was the “A-scope,” a two-dimensional view of reflected signals on a specific bearing. The horizontal axis displayed range, while the vertical axis indicated the strength of the reflected signal. If the radar detected nothing, there would be no return, and the display would be dominated by “grass”—low-level noise signals detected by the receiver, akin to “snow” on early television sets.14 When the radar did detect an object, a vertical spike would appear at the appropriate range. The strength of the signal was indicated by the size of the spike. Strong signals produced a tall vertical “pip”; weaker ones were shorter. If using a search radar, the operator would stop the antenna and note the bearing and range. The nature of the display meant that the radar could either search, scanning for potential contacts, or focus on a single bearing, recording the specific location of a contact. It could not do both simultaneously.
It took effort to translate this information from the display into a picture of potential contacts. Search-radar operators were trained to focus on pips and accurately determine their range and bearing. They were not responsible for tracking the contacts they identified, and the A-scope made it very difficult for them to do so. Instead, they reported a series of ranges and bearings. Other members of the crew took this data and added the necessary context to create situational awareness. On large ships, like carriers and battleships, this was done in a Radar Plot. On smaller ships, it was done in the captain’s brain. Both mechanisms were frequently overwhelmed.
On fire-control radars, the focus of the A-scope could be narrowed. Operators could increase the resolution and zoom in on a small area of the display—either five hundred or a thousand yards in range—and see more detail around the target. This allowed them to determine target range very accurately and, if conditions were right, even to observe the splashes of shells. However, this procedure limited their view to a narrow window. If the target slipped outside, it could easily be lost.
Radar signals reflected off land as well as ships, and the peculiar littoral terrain near Guadalcanal made it difficult to identify contacts. Savo Sound is surrounded by three islands. Guadalcanal forms the southern border. To the north, Florida Island hems it in and forms the sheltered anchorage at Tulagi. The southern tip of Florida reaches toward Guadalcanal, forming the eastern entrance to the sound. Reefs segment that entrance into three separate channels: Nggela, Sealark, and Lengo. To the west, the small island of Savo divides the western entrance to the sound into two roughly equal channels. The proximity of land interfered with radar signals, cluttering displays with echoes from the islands and making it difficult or impossible to distinguish targets. This interference was a surprise, and it undermined prewar faith in radar systems. These limitations informed how radar was used and how battles were fought. Radar was not yet the sophisticated technology we think of today; for the first year of war in the Pacific, it was a rudimentary tool that did not seamlessly integrate with existing shipboard information systems. This largely explains why the Navy, despite this powerful new technology, found the battles off Guadalcanal so confusing and difficult.
However, some ships benefitted from much more sophisticated radar installations. One of the most important outcomes of the prewar work with radar was the development of the “plan position indicator,” the PPI scope. It was a dramatic improvement over the A-scope and vastly enhanced the ability of operators and their ships to process radar information. The PPI was the brainchild of Dr. Robert Morris Page of the Naval Research Laboratory. Page applied a particularly nautical solution to the problem of radar displays; he took the Navy’s established plotting paradigm—the top-down view—and devised a way to present radar information in a similar format. It was the first iteration of the display with which we are all familiar, with the radar in the center surrounded by a bird’s-eye view of contacts.
As the radar revolved, potential contacts appeared as pips. Each pip would remain momentarily on the screen, allowing the operator to record the bearing and range without pausing the radar’s scanning motion. The strength of the return, giving some estimate of the relative size of the target, was reflected in the size and brightness of the pip. The new display tightly integrated radar information with the mental models of the operators, making it much easier to translate that information into a picture of the surrounding world and increase situational awareness.
The first radar to use the PPI scope was the SG, the Navy’s first microwave search radar; later, it became the standard display for all search radars. A few ships that fought near Guadalcanal, including the cruiser Helena, destroyer Fletcher, and battleships Washington and South Dakota, were equipped with the SG radar and PPI scope. These ships consistently had better situational awareness in combat, but they lacked an effective means to share that picture with other ships in company.
FLAWS IN THE NAVY’S APPROACH
They were missing because of invalid assumptions about how commanders and their ships would approach combat. The Navy recognized that a decentralized system—in which each commander applied his own judgment to his circumstances—could adequately deal with the complexity of combat. The emphasis on decentralized doctrinal development reflected this, and it made squadron and task force commanders responsible for formulating doctrines and plans to guide their forces in battle. To make the system effective, however, ships had to stay together long enough to absorb a common doctrine and create a shared context for decision making. But the demands of the Guadalcanal campaign and the needs of a two-ocean war combined to destabilize the Navy’s fighting units and undermine this process.
Creating a shared context took time and required preparation. Although the fleet had devoted significant attention to preparing for major fleet actions, it had spent very little time preparing for fights between smaller task forces. Individual task force and squadron commanders were expected to fill the gap and develop specific plans for their forces. Off Guadalcanal, they were consistently unable to do so. Forces were repeatedly thrown together piecemeal without the necessary time for indoctrination. The resulting problems are evident throughout the campaign. In the aftermath of the defeat at Savo, Vice Admiral Ghormley cited the lack of time available for indoctrination: “Detailed plans and orders for the Watchtower Operation were of necessity prepared in a short space of time immediately prior to its execution, giving little, if any, opportunity for subordinate commanders to contact commanders of units assigned to them for purposes of indoctrination.”
During Guadalcanal I, the lack of an effective doctrine was an acute problem. Rear Adm. Daniel J. Callaghan was given command of a “scratch team” formed from two separate forces the day before the battle. There was no time to indoctrinate the new ships, and because Callaghan had assumed his own post as a task-group commander (TG 67.4) barely two weeks prior to the battle, there was no doctrine even for his own force. Vice Adm. William S. Pye, president of the Naval War College, would note this deficiency in his comments on the action: “It seems . . . that the American force went into this action without any battle plan; without any indoctrination, or understanding between the OTC [Callaghan], and his subordinates; with incomplete information as to existing conditions in possession of subordinates.”
At Guadalcanal II, two days later, Rear Adm. Willis A. Lee was in a similar situation. None of his six ships had ever operated together before. His four destroyers were from four different divisions and possessed nothing resembling a common doctrine. The same held true for his two battleships. Lee suffered “from the same lack of practice in teamwork that had plagued Callaghan.”
Planning suffered along with indoctrination. Before the war, the Navy assumed the OTC would develop a battle plan that would explain his intentions and the way he expected to fight. It would provide context for the interpretation of task-force doctrine and help align decision making. To prevent confusion, these plans had to be concise and extremely clear. Unfortunately, the same circumstances that undermined the ability of task force commanders to develop common doctrines also inhibited the creation of battle plans. Captains frequently went into battle without the shared context required for effective coordination. Instead of fighting as cohesive units, the Navy’s task forces broke apart and fought as individual ships.
The Navy understood that this approach was very costly, but there was no alternative. The battles of Guadalcanal I and II came during the decisive moment of the campaign; Lee and Callaghan were fighting to ensure the survival of Henderson Field, which had become the key to victory. Success or failure depended on their ability to fight their confused collections of ships.
The Navy’s performance in the night battles of 1942 was also hindered by prevailing assumptions about how to employ torpedoes. The emphasis on major fleet action limited the Navy’s destroyer doctrine and focused destroyer commanders on attacking a well-defended enemy formation at night. In such an attack, torpedoes would be fired at the enemy battleships after the cruisers and destroyers had used their guns to penetrate the screen. This meant that destroyers were trained to preserve their torpedoes and use their guns first. Originally weapons of stealth, the Navy’s destroyers had lost the art of using their torpedoes in a surprise attack.
The 1929 Destroyer Instructions reflected these prevailing assumptions. Torpedoes were to be used primarily against the “objective,” the enemy capital ships. Only a limited number could be expended against other targets: “While the main mission of the attack is to sink the objective [enemy battleships,] . . . it must be remembered that favorable positions for torpedo fire are very seldom gained in night operations, and that every opportunity must be taken to inflict damage on any enemy ships encountered. . . . For this reason, destroyers are authorized to fire one torpedo at any destroyer and two at any cruiser or light cruiser encountered at such close range that there is a practical certainty of hitting.” The 1938 version of Night Search and Attack Operations reinforced the emphasis on preserving torpedoes for large targets: “While penetrating an enemy screen advantage will be taken of any favorable opportunity to torpedo enemy cruisers. Torpedoes will not normally be used against enemy destroyers.”
These assumptions led to simplistic interwar torpedo exercises. Battle Torpedo Practice C was the Navy’s standard night-torpedo exercise, designed to simulate an attack on an enemy battleship. It assumed that the target would be slow-moving and that the attack would come after the destroyer division had penetrated the enemy screen. This meant the target would be fully alert to the presence of the attacking destroyers, open fire to repulse them, and thereby reveal its location and provide a convenient point of aim for the destroyer torpedoes. Blinking lights simulated the battleship’s gunfire.
While the Navy’s gunnery exercises measured how quickly the guns were brought on target and how often they hit, successfully focusing crews on the importance of quick and accurate gunfire, Battle Torpedo Practice C considered only the accuracy of a single torpedo. If the torpedo hit the target, the ship received a perfect score. If it missed, the score was zero. Most destroyers, aided by the blinking lights, managed to hit. The only way they could improve upon their scores was to fire their torpedoes earlier in their attack runs. The artificialities of the exercises and emphasis on major action prevented the development of more sophisticated and effective torpedo tactics. Although there were extensive night exercises before the war, not one of those held in 1938 simulated an encounter like the battles off Guadalcanal, which were to be dominated by fast-moving cruisers and destroyers.
Immediately before the war, more complex exercises with faster targets and more challenging torpedo-fire-control problems were in fact introduced. But they came too late to alter the pervasive beliefs that enemy capital ships were the primary target for destroyer torpedoes and that most attacks would be delivered at close range against slowly maneuvering, well-illuminated targets. Off Guadalcanal, the Navy eschewed the potential of stealthy torpedo attacks and focused instead on gunfire as the dominant weapon.
The IJN approached the problem quite differently. Forced by the interwar naval-limitation treaties to accept a fleet of significantly smaller size, the Japanese had sought to redress the imbalance through technological innovation. The U.S. Navy failed to anticipate this and went to war assuming that Japanese ships and weapons would possess capabilities broadly similar to its own. That the Japanese would develop torpedoes with unprecedented range and striking power was wholly unanticipated.
The Type 93 Mod 2 torpedo, more commonly known as the Long Lance, was the IJN’s counter to the large size and fighting power of the American battle line. Introduced in 1936, it was capable of a range of 20,000 meters (21,900 yards) at fifty knots, its highest speed setting. Its warhead weighed 490 kilograms (1,080 pounds). The Navy’s contemporary, the Mark 15, had a range of only six thousand yards at forty-five knots; it carried an 825-pound warhead. The better performance of the Japanese torpedo was due to its larger size and the fact that it used pure oxygen as a combustion agent. The Mark 15 used regular air.
The Japanese also developed stealthy tactics that emphasized firing torpedoes before opening fire with their guns. At night, gun flashes created clear aim points. If torpedoes were fired and gunfire held until the torpedoes were among the enemy ships, the targets could be quickly overwhelmed by shells and torpedoes striking simultaneously. The lethality of this tactic was demonstrated at Savo Island.
American naval officers consistently underestimated the range and speed of Japanese torpedoes. They believed that submarines were firing them, not cruisers and destroyers. In the aftermath of Savo Island, Rear Admiral Turner expressed his belief that heavy cruisers Vincennes, Quincy, and Astoria “ran into [a] submarine and torpedo trap.” An intelligence summary issued after Guadalcanal I reflected a similar assumption about Japanese torpedo tactics: “Jap[anese] torpedo attacks are the biggest threat. They appear to succeed in firing well placed torpedo salvoes. They hit from the flank and also the disengaged side. They undoubtedly use destroyers and cruisers as well as submarines well placed in area.”
Ignorance regarding the range and accuracy of Japanese torpedoes resulted in tactics that played to their strengths. At the battle of Tassafaronga, Rear Adm. Carleton H. Wright maintained the course and speed of his cruisers while rapidly firing at Japanese destroyers. The cruisers’ gun flashes provided excellent points of aim, and their steady course carried them right into a barrage of enemy torpedoes. Two struck Wright’s flagship, the heavy cruiser Minneapolis; New Orleans, Pensacola, and Northampton were also hit. The Navy failed to recognize the true capabilities of the Long Lance until late 1943, far too late for the Guadalcanal campaign.