A century ago the U.S. Navy commissioned its first battleships armed only with the heaviest guns, the USS South Carolina and Michigan. To a casual observer, the event represented little more than that the conservative U.S. Navy had finally caught up with the British, who had commissioned the "all-big-gun" HMS Dreadnought in 1906. In fact, the two South Carolina-class ships reflected thinking that considerably preceded that of the British ship.
The American pair introduced a feature that would mark most subsequent big-gun ships: superfiring turrets, which were higher than adjacent turrets. They enabled the South Carolina and Michigan to carry all their turrets on their centerlines, and that was one reason the eight-gun ships packed the same broadside as the ten-gun Dreadnought on 2,000 tons less displacement. The concept of superfiring turrets had never been applied to a ship's main battery, but the slow pace of U.S. shipbuilding diluted the impact of the American innovation. Conceived in the fall of 1904, the two battleships were not commisioned until nearly six years later.
The Revolution in Gunnery
The South Carolinas were part of an international trend to arm battleships with guns of a single heavy caliber (apart from small quick-firing guns intended to beat off torpedo-boat attacks). HMS Dreadnought is generally associated with the final step of that evolution, which was an inevitable outcome of the way in which battleship gunnery changed beginning in the late 19th century. Around 1880 capital ships were armed with a few large guns that fired very slowly but were the only weapons that could penetrate heavy armor. Some naval experts suspected their rates of fire were so slow the guns were largely ineffective against enemy ships, particularly fast ones. The Italians went so far as to build essentially unarmored-but fast-battleships armed with the heaviest available guns.
Later in the decade, quick-firing (QF) guns, which used brass cartridge cases, were developed. The weapons' breeches did not need to seal nearly as elaborately as on previous guns, hence they could be worked more rapidly. After the QF system was applied to medium-caliber guns, ships armed with them could tear up the less-heavily armored parts of battleships. By 1890 the world's leading navies armed their battleships with four heavy (typically 12- or 13.5-inch) slow-firing guns plus numerous medium-caliber QF guns. That, in turn, required the technology to mass produce large brass cartridge cases with sufficient precision.
American industry initially did not have that capability, so the U.S. Navy chose an alternative to medium QF guns for its first battleships (the Indiana class)-8-inch intermediate batteries. These guns fired more quickly than the heaviest ones but were slower than quick-firers. By 1898, as U.S. industry improved, there was little point in retaining 8-inch guns in new battleships; however, they performed better than the Navy's heavy guns during the Spanish-American War. Without further analysis, the service decided to bring them back for future ships. Thus the U.S. battleships designed after 1898 had three main calibers of guns: heavy (12 inch), medium (8 inch), and QF (6 and later 7 inch).
Around this same time, because of improved breech mechanisms, the rate of fire for the heaviest guns improved from more than five minutes per round to one round a minute or better. Smaller large guns, such as 8- and 10-inchers, were still faster, but they had much less of an advantage in that area. Moreover, Krupp and Carnegie armor introduced in the mid-1890s could resist 6- and 8-inch fire at expected battle ranges. Because of its superior quality, the new armor's thickness could be reduced, and the same weight of armor could then be used to protect much more of a ship. The changes resulted in the combination of medium QFs and heavy guns making less sense.
Between about 1900 and 1904 the major navies ordered (or at least designed) ships in which at least some of the earlier medium-caliber QFs were replaced by much heavier non-quick-firing guns, 9.2-inchers in the Royal Navy for example. Because by then the 12-inch guns fired faster, however, even those weapons must no longer have seemed worthwhile. The next logical step was to abandon everything short of the heaviest guns, plus very light ones to beat off torpedo-boat attacks. That happened first in the Royal Navy, but it could not have been a great surprise.
Sudden improvements in torpedoes and in gunfire control also pointed in that direction. It was generally accepted that the effect of gunfire in battle would be cumulative-ships would be gradually smashed down. On the other hand, only one or two underwater hits would likely be fatal. (The flagship of the Russian Far East Fleet sank in 1904 after hitting a single mine.) Because battleships were armed with torpedoes as well as guns, both of which would presumably be used during battle, it made sense to knock out the big vessels before they could launch their "fish." Navies, certainly the U.S. Navy and the Royal Navy, therefore became interested in gunnery beyond torpedo range.
At such ranges (perhaps 3,000 yards for the U.S. Navy in 1902, about double the expected battle range four years earlier) even the current 8-inch gun could not penetrate battleship armor. Consequently, the Navy adopted a 10-inch gun for its large armored cruisers, which were probably expected to fight alongside battleships.
A Flawed Design Process
In 1900, as for many years before and after, the materiel side of the Navy was run by independent bureaus responsible to the Secretary of the Navy: the Bureau of Construction and Repair (C&R), which designed and built the ships; the Bureau of Engineering, responsible for their machinery; the Bureau of Ordnance, responsible for their guns, torpedoes, and armor; and the Bureau of Equipment, which oversaw electric installation, for example. The Bureau of Navigation, headed by Admiral of the Navy George Dewey, ran the operational Navy and represented seagoing opinion, but in no sense was this bureau a naval staff. In the 1880s, the designs for the first ships of the modern "New Navy" had proved less than satisfactory, partly because elements from the various bureaus did not always mesh properly. To remedy the problem, Secretary of the Navy Benjamin Tracy created the Board on Construction in 1889 to coordinate the materiel bureaus.
Later, the service's experience in the Spanish-American War convinced Secretary of the Navy John D. Long to create an advisory panel mainly to conduct peacetime war planning-the General Board of the U.S. Navy, initially headed by Admiral Dewey. Founded in 1884, the Naval War College acted as the Navy's think tank, and its senior class would test new ship concepts in tactical games. Each summer the college held a conference in Newport, Rhode Island, to consider important problems facing the sea service. Also, the executive branch of government naturally exerted influence on the Navy, and once President Theodore Roosevelt assumed office in 1901, the service had a strong, interested, and energetic supporter in the White House.
Typically the Secretary of the Navy, advised by Admiral Dewey, set broad requirements for the Board on Construction. In a period of relatively stable technology, that posed no great problem; those setting the requirements had a fair idea of what sort of ship would meet them. The Board on Construction would offer alternative ship designs on a take-it-or-leave-it basis; the Navy chose one, and then the Bureaus developed that design.
Matters became much more difficult when the technology began to change suddenly, as it did about 1902. During such times, it was entirely possible for the designers to produce a ship, which though nominally meeting the specified requirements, was hardly satisfactory. Moreover, around the turn of the century there also was a growing feeling within the Navy that the operational side of the service was exerting far too little influence on the bureaus, which were perceived as overly conservative. The sudden change in battleship design caused the grievance to bubble to the surface-and changed the way the Navy conducted itself for the next 90 years.
Congress was an important factor in the equation. From the late 1890s until around 1902 the size of U.S. battleships rose from 12,500 to about 16,000 tons. Congress badly wanted to control unit cost and hence size. A compromise reached in the spring of 1903 provided two small battleships (the 13,000-ton Idaho class) and three repeat 16,000-tonners.
Movement Toward an All-Big-Gun Ship
Among the alternative designs the bureaus offered for the two 13,000-ton ships was one prepared in May 1902 for a vessel armed with twelve 10-inch guns.1 (U.S. designers had been assigned to develop a small battleship because it appeared that Congress might cut the size of all the ships it authorized.) The design was rejected in favor of a more conventional (and unsuccessful) mixed-caliber battleship.
The important overall lesson was that Congress would not happily accept growth beyond the 16,000-ton size of the largest existing U.S. battleships. Yet it was widely understood-and not just in the United States-that battleship design was due for a major change from the existing mix of calibers to the sort of single-caliber design suggested by the 10-inch-gun ship. The best-known public expression was an article by Italian chief constructor Vittorio Cuniberti in the 1903 edition of Jane's Fighting Ships calling on the Royal Navy to build a ship armed entirely with 12-inch guns. The article has often been cited as the inspiration for HMS Dreadnought. In fact, British Admiral Sir John Fisher, the driving force behind the vessel's construction, was already convinced of the need for an all-big-gun ship. Cuniberti's article is probably better seen as evidence that the idea of such ships was well known throughout the naval world more than two years before the Dreadnought was launched.2
In retrospect, the best-known U.S. advocate of the single-caliber battleship was Lieutenant Commander Homer C. Poundstone. In December 1902, he sent President Roosevelt a paper in which he pressed for much larger battleships.3 In its appendix he argued for replacing the existing mix of 12- and 8-inch guns with 11- and 9-inchers. By the time the paper was published as two articles in the Proceedings of the United States Naval Institute, however, he favored an alternative battleship (which he named the USS Possible) armed with twelve 11-inch guns.4
In June 1903, Naval Constructor Washington Irving Chambers, who seems to have developed Poundstone's designs, submitted an all-big-gun design to the Naval War College. Chambers' proposal was significant because he was a trained naval architect working for the Navy who was suggesting something practicable. In considering his and other new ship proposals, the Navy benefited from the Naval War College's ability to conduct tactical games to evaluate the designs. The games sometimes showed that radical change was worthwhile, and at other times that it was pointless.5
The games conducted during the college's conference in the summer of 1903 showed that a ship armed with twelve 11- or 12-inch guns (eight on the broadside) would equal three conventional battleships. Ultimately, the conference favored the new kind of battleship, proposing that such a ship be designed.
On that basis, in October 1903 the General Board formally asked for a feasibility study of a battleship armed with twelve 12-inch guns and as many 3-inch (antitorpedo-boat) guns as possible. The request turned out to be the genesis of the South Carolina and Michigan design. Laying out design requirements was a new endeavor for the General Board, which had hitherto been mainly concerned with war planning, although it had also suggested a long-range construction program to match U.S. national strategy. The board's unprecedented action-specifying the guns in its feasibility-study requests-suggests the new kind of battleship's importance.
The Process Grinds to a Halt
After receiving the feasibility study request, the Bureau of Construction and Repair did not act on it. C&R had limited manpower and was responsible for creating detailed plans for the many ships already authorized and under contract. There was no particular reason for the bureau, or for that matter for the Board on Construction, to appreciate just how vital an innovation the all-big-gun ship might be. C&R, which by 1904 had gone through three conventional battleship designs in as many years (the Virginia, Connecticut, and Mississippi classes), complained that it had already produced a single-caliber design (probably the one with 10-inch guns) only to have it rejected. Given the structure of the Navy, the General Board could advise the civilian Secretary of the Navy, but it could not enforce its will.
The board, however, did repeat its request in January 1904, this time calling for four 12-inch and as many other heavy guns, at least 10-inchers, as possible. Even if the ship could not have all-12-inch guns, it was pointless to keep mounting 8-inchers. (The move away from a homogeneous big-gun battery seems to have reflected doubts that the heaviest guns could be mounted on the broadside.) The 1904 summer Naval War College conference compared this kind of ship, under the name South Carolina, with the 1903 ship (all 12-inch) and with the existing mixed-battery Connecticut-class battleship (four 12-inch guns, eight 8-inchers, and twelve 7-inchers).
At both the 1903 and 1904 conferences the point was made that at the current torpedo range (3,000 yards) an 8-inch-gun projectile could not penetrate typical medium battleship armor. Even a 10-inch gun probably would not penetrate the thickest battleship armor; only a 12-inch gun was certain to do the job. Increase the range to 7,000 to 8,000 yards to account for expected improvements in torpedoes, and the argument for 12-inch guns was even stronger.
In September 1904, C&R finally began designing the least radical version of the General Board's proposals-four 12-inch and eight 10-inch guns, the latter instead of eight 8-inch and twelve 7-inch guns. Single 12-inchers, however, could replace the twin 10-inch guns. The original plans called for heavy wing turrets, which badly strained the hull, as well as large holes cut in the ship's deck (her upper strength member) for the wing barbettes, the cylindrical towers atop which turrets rotated.
On 3 March 1905, while C&R was still grappling with the design problems, Congress authorized two new battleships, to be named the South Carolina and Michigan. The legislators, however, renewed the 16,000-ton limit, which made the designers' task all the more difficult. Heavy wing turrets seemed impractical on a battleship of that size.
Constructor Capp's Revolution
The Navy's chief constructor and head of the Bureau of Construction and Repair, Rear Admiral Washington L. Capps, chose a radical solution: all big guns-eight 12-inchers-on the centerline in four turrets, two of which could superfire over the other two turrets. Other contemporary designers seem to have thought in terms of single-ship actions involving a lot of maneuvering. Hence they tried to give their ships maximum firepower in all directions. Turrets were aimed through hoods on their roofs, so it was not at all obvious that a pair of superfiring turrets could fire along the ship's centerline without their gun blasts seriously interfering with and damaging the lower turrets as well as their gunners.
To Capps it was obvious that the Fleet would normally steam in line-ahead formation, its ships firing to one side. Blast interference would be irrelevant; the issue was how many guns could fire on the broadside. The most economical approach was to put all the guns on the centerline, two turrets higher than the others. The Bureau of Ordnance, meanwhile, redesigned sights, relocating them from turret roof hoods to the turrets' sides. The South Carolinas would be the first battleships to feature the innovation.6
Within a few years every other navy in the world had chosen the same superfiring arrangement, though they generally did not abandon sighting hoods. Capps' final design was apparently the 19th (Scheme S) in a series of all-big-gun plans for the battleships. The superfiring idea probably dated from around April 1905, when the constructor asked the Bureau of Engineering to shrink machinery spaces to give him more centerline space for the main battery. The design was completed by late June 1905 (HMS Dreadnought was not laid down until that October).
Adopting the all-big-gun battery and the centerline gun arrangement saved armor weight because providing medium armor to cover a long casemate containing 7-inch guns was no longer necessary and also because there were now only four turret-barbette combinations instead of eight.
Armor Advances and Flaws
Capps also rethought the arrangement of armor. Earlier ships had armored decks that sloped down to meet the lower edge of the main armor belt (the waterline armor); higher up, the upper (casemate) belt was thinner armor protecting the lighter guns. The sloping deck would protect the ships' machinery and magazines against shells passing through the main belt, and also against debris caused by shells hitting the upper (casemate) belt. It was apparently assumed that the enemy would use his heavy guns to attack the main belt and main turrets; the largest weapons could not fire quickly enough for him to spend time on secondary targets, such as casemates and turrets housing lighter guns.
This reasoning, however, collapsed as rates of fire for 12-inch guns improved and ranges opened. In Capps' view, a thicker main belt would stop major-caliber shells. He therefore saved some weight by replacing the old sloping armor deck with a flat deck that met the upper edge of the main belt. The saved weight went into a main belt that was thicker, but narrower, than earlier ships'. Adopting thinner barbettes for the main battery was necessary because the double-height ones for the superfiring turrets added considerable weight. Bow and stern armor was eliminated to save weight and because Capps rightly concluded that it would merely serve to detonate armor-piercing shells.
The armor arrangement was an advance over previous ones but had major flaws. The South Carolinas were conceived as part of a conscious U.S. program to design ships for longer battle ranges at which only 12-inch shells (or heavier ones) would matter. Capps understood that light armor was useless, but it would therefore make sense to have the widest possible belt of maximum thickness and no armor at all elsewhere.
That was exactly the conclusion the General Board would reach a few years later when it discussed a design that became the Nevada class. Capps presumably retained the upper belt in the expectation that shells approaching steeply, at longer ranges, would tend to hit it and detonate, their fragments caught by the armor deck at about the waterline. Around 1910, when really long ranges were being achieved (or at least seemed achievable), the choice was simpler: Put a main armored deck on top of the widest, thickest belt possible, and add a thin lower deck to catch fragments that penetrated the main deck.
Changes Below and Above Decks
Capps' centerline-turret solution avoided the problem of large holes in the sides of the upper deck; however, it created new problems. Placing two more turrets on the centerline compressed spaces that earlier battleships had devoted to machinery. Given the limit on tonnage and the expectation that the South Carolinas would work tactically with the existing battle fleet, there was little point in increasing their power so they would be faster, which is what the British did by powering the Dreadnought with turbines.
Capps, however, had to provide the same power as other large U.S. warships in less space without any change in machinery technology (the Navy was not ready to apply turbine power to a battleship). He also wanted to reduce the size of boiler rooms to provide space for a degree of outboard torpedo protection. Ideally he would have cut the number of boilers by a third by using more powerful ones, but the Bureau of Engineering could not provide them. In the end his solution was to eliminate centerline bulkheads (the rearrangement is indicated by the use of two, rather than three, funnels). Capps saved hull weight (and top weight) by replacing the flush deck of the previous Connecticut class with a hull cut down aft. That in turn eliminated hull space used for officers' quarters, which Capps moved into the superstructure amidships.
The South Carolina design changed again while the ships were being built. These were the first battleships completed with what became a trademark of U.S. dreadnought design: the cage, or lattice, mast. Probably inspired by early British progress in long-range firing, the U.S. Navy conducted its first special firing exercise in 1904. Effective long-range firing was only possible if the splashes from salvoes could be spotted from aloft. In December 1905, the Board on Fire Control recommended a raised spotting platform, perhaps on a lattice mast, that could remain in place even if shells cut some of its supporting tubes.
The idea was not particularly popular. The Bureau of Ordnance preferred a light bridge below the funnel tops, so it would be smoke-free. Advocates of longer-range fire pointed out that the higher the spotters, the longer the effective range-which would mean superiority in combat. Capps does not seem to have had much interest in this issue. He merely wanted traditional pole masts located far enough from the ship's bridge so that if shells burst against them the fragments-or a toppling mast-would not destroy the conning station. Capps' inventive solution was to arrange his two masts off the centerline, diagonal to each other.
The Bureau of Navigation, in effect representing the operational Navy, pressed for lattice masts in the ships against opposition by C&R and the Bureau of Ordnance, which cited the cost of changes late in the construction process. The Secretary of the Navy made the final decision, approving high cage masts after Atlantic Fleet battle practice showed how important those structures could be. The South Carolinas were therefore modified; Capps' masts were cut down to king posts for boat cranes, and centerline cage masts were installed. All earlier U.S. battleships were also fitted with cage masts, many of them on their return in February 1909 from the Great White Fleet's round-the-world cruise.
The Design Process Comes Full Circle
The South Carolina design was the very fortunate outcome of a flawed process. Washington Capps was clearly brilliant; however, he did not fully understand the gunnery revolution then in progress, and neither did the Bureau of Ordnance. Line officers such as Navy Lieutenant Commander William S. Sims fomented the revolution, which much increased effective range. Thus Capps' design did not really reflect the evolving need for longer-range spotting (cage masts), and his armor scheme was tied to shorter battle ranges. Overshadowed by the brilliance of the centerline gun arrangement, these flaws became obvious in the next (North Dakota) class.
This time Congress was persuaded to relax the battleship tonnage limit, but the design proved less than satisfactory. Line officers criticized its armor arrangement; the main belt was too narrow and too easily submerged. Jane's Fighting Ships' American contributor, Henry Reuterdahl, added insufficient freeboard and unsafe turrets to the list of defects. Critics, including Sims, generally charged that the problems were not due to incompetent designers so much as to a flawed design process. They convinced President Roosevelt to attend the 1908 Naval War College summer conference, where the ship-design controversy was the main topic.
The conference concluded that line officers-those who would operate the ships-should have a far more intimate connection with future U.S. warship design. The General Board was assigned to develop "characteristics," or design requirements, for ships. It would monitor the development of the preliminary designs, deciding whether they were adequate before detailed design began. The first battleships influenced by the change were the New York and Texas, but the first deeply affected were the Nevadas, whose integrated design certainly reflected the new long-range gunnery.7
The General Board lost its characteristics and design-monitoring role during World War II, but the idea of interaction rather than a take-it-or-leave-it approach survived. Line officers often asked for characteristics they did not really want when they found out what sort of ship would result. Designers often did not appreciate exactly what was wanted. Only a back-and-forth process could tease out the fact that ships like the South Carolinas were a contradiction-partly conceived for the future of long-range battle, but also partly to fight alongside predreadnoughts with short-range protection.
Similarly, only interaction could reveal to the officers who wrote the characteristics for later classes what requests for higher speed or for heavier guns would cost. They had to be able to change their minds or to add things not initially wanted. It was vital, moreover, that this back-and-forth process stay inside the Navy. That way there was no great cost in making drastic changes, nor was there great embarrassment when someone's good idea turned out not to be so good after all.
The Navy is presently using a system not too different from the take-it-or-leave-it process that produced the South Carolinas. In the 1990s the Office of the Secretary of Defense (OSD) decided to standardize procurement across the services and a wide variety of weapons and systems. Military characteristics would be worked out by senior officers, and then engineers would develop design alternatives. The senior officers who had framed the requirements in the first place would choose-on the same take-it-or-leave-it basis the U.S. Navy rejected in 1908. Thus those developing characteristics had no particular idea of what they implied in physical terms. Forgotten was the expensive lesson that small changes in characteristics might have enormous effects on the size and cost of a ship.
OSD wanted all design work given to private contractors, the hope being that they would innovate freely. Once design requirements had been set, however, changing became exorbitantly expensive. The first U.S. warship designed in this way was the Zumwalt-class destroyer-the DD(X). Her large size can be traced directly to a demanding requirement to reduce radar cross section. Once those buying the innovative ship realized what they had, they likely thought that retreating a bit on the radar cross section requirement might have been a good idea.8 That sentiment harks back to the Navy's feeling of regret concerning the South Carolinas: While the superfiring turrets were cutting-edge, the ships' protection did not quite comport with the way she was expected to fight.
1. No drawing seems to have survived, but the number of guns suggests that in effect the designers would have replaced both 12- and 8-inch guns (in turrets) with 10-inch guns, in what would later be called a hexagonal arrangement (a broadside of four turrets, with a nominal end fire of three turrets). Early German dreadnoughts adopted this arrangement, as did the first Japanese dreadnoughts.
2. Fisher initially preferred a ship with 10-inch guns, which fired faster than 12-inchers; he seems to have changed his mind based on early experience in the Russo-Japanese War, where a British attaché remarked that 10-inch shots could only barely be seen alongside those from 12-inch guns. A 10-inch gun fired a shell weighing about 500 lbs., a 12-inch, about 850 (an 8-inch gun fired about a 200 lb. shell).
3. For a more complete discussion of the emergence of the South Carolina design, see the author's U.S. Battleships: An Illustrated Design History (Naval Institute Press, 1985).
4. Poundstone's paper appeared in the March and June 1903 issues of the Proceedings of the U.S. Naval Institute. Cuniberti's article appeared that October.
5. War College games shaped the first U.S. carriers, because they showed how important sheer numbers of aircraft were. They killed several proposals for battleships armed mainly with torpedoes, and they helped kill projects for flight-deck cruisers. The nascent U.S. Navy probably relied more heavily than others on gaming than other navies because it was creating a fleet so rapidly that it could not rely on sea experience.
6. Firing tests on board the modified monitor Florida, on 6 March 1907, showed that with the new sights the ship could fire her guns at any angle without suffering undue blast interference. Foreign ships with hoods could not fire along their centerlines, but U.S. ships could fire over any arc (e.g., with four guns dead ahead). Capps may have been encouraged by previous successful U.S. experience using turrets combining two 12-inch guns with two 8-inch guns on an upper level.
7. The Nevada armor arrangement was used in all U.S. battleships designed through 1919, and almost all World War II battleships had a related protection concept. Ironically, few U.S. officers seem to have realized how advanced the U.S. ships were. During the 1921 Washington Conference and afterwards, there was considerable discussion of "post Jutland" battleships, always without any realization that all U.S. battleships from Nevada on were effectively post-Jutland in the sense that they were well adapted to long-range fire.
8. The ship is large, first, because to limit radar cross-section everything must be internal. Second, in order to keep radar cross section constant despite expending fuel, etc., the ship has large internal ballast tanks. It is also stabilized, to hold the angle between the ship's side and the sea constant.
Note on sources:
This article is based largely on my book U.S. Battleships: An Illustrated Design History, which in turn was mainly based on material from the General Board (now at the National Archives and Records Administration, at that time in the Operational Archives) and the correspondence of the Bureau of Construction and Repair (RG 19, National Archives). Homer Poundstone's paper, with drawings, pressing for larger battleships is in the historical archives of the Nimitz Library, U.S. Naval Academy.