Tuesday, July 19, 2011

John Wilkins vs George Orwell

I saw the following quote today on, of all places, Failbook.com, where it was used as a criticism of Facebook. It’s been many years since I read 1984, but this passage immediately took me back to a high school classroom:

"'Don't you see the whole aim of Newspeak is to narrow the range of thought? In the end we shall make thoughtcrime literally impossible, because there will be no words in which to express it. Every concept that can ever be needed will be expresses in exactly ONE word, with its meaning rigidly defined and all its subsidiary meanings rubbed out and forgotten."

Imagine trying to speak English without using euphemism, slang, idiom, puns, similes, dialect, or any other of the other workaday rhetorical lubricating techniques that keep things interesting. Even synonyms can help keep things interesting. If you take all of that away—or even just the everyday mutations, spindlings, and convolutions that go in through our eyes and ears without being recognized as such—it’s no longer English, and would in fact be a recognizably distinct language (possibly legalese).

English is a miraculous language, in many respects; from its origins as, in the words of Howard Tayler, “a bad habit shared by Norman soldiers and Saxon barmaids who discovered that if they shared that habit they could share other things,” it has evolved to produced Chaucer, Shakespeare, Milton, Wodehouse, Kipling, Dickens, Churchill, Stephen Fry, and countless others, (yes, even George Orwell) all of whom have bent and misused words to capture shades of meaning that no dictionary in the writer’s day (if there were any) would likely have approved of.

English also produced Black’s Law Dictionary, than which no book ever written is more fanatically devoted to specific meanings for words. Persons aggrieved at the outcome of the Casey Anthony murder trial should remember that there is a tremendous difference between being acquitted (“not guilty”) and being innocent. This rigidity in the use of words has its place, usually when large quantities of money or prison sentences are at issue. Virtually every modern legal or technical document—from laws down to plumbing codes—includes a section defining certain terms used in the rest of the text.

Think about dictionaries for a second. We have books that do nothing more or less than to tell us what words mean. The idea becomes stranger the more one thinks about it. No wonder many of Dr. Samuel Johnson’s contemporaries were at first amazed that England’s foremost man of letters he would spend so much of his life working on the first proper English dictionary. They probably didn’t think one was necessary. Modern English, on the other hand, likely couldn’t function without dictionaries.

With a vocabulary with six times as many words as French, English clearly needs dictionaries, and more help besides; it is the only language with so many synonyms that it needs a thesaurus to keep track of them all. Not only that, but many words have multiple meanings, some of them very different from each other. As Bill Bryson noted in English and How It Got That Way, “any language where the unassuming word fly signifies an annoying insect, a means of travel, and a critical part of a gentleman’s apparel is clearly asking to be mangled.”

For that matter, English also has such a blurred line between literal and figurative uses of terms in everyday communication that it has a batch of words and phrases used to clarify which sense of the word is intended, e.g. ‘literally’ or ‘in the strict sense.’ Perhaps inevitably, the word ‘literally’ has itself come to be abused as a mere indicator of emphasis—“I was so drunk last night that I literally barfed from here to Hartford.” If the reader took ‘literally’ in its own literal sense, doctors would be shocked, engineers would be skeptical, and Pentagon weapons designers would be grinning at the thought of Jagermeister-fueled long-range vomit projectors for the Marine Corps.

Although Orwell presented Newspeak as the tool of an authoritarian regime, rigidly-defined languages with specific meanings for specific words have their place. Any attorney, doctor, or engineer relies on these particular meanings in order to give clarity to their communications—for example, an engineer stating, “this structure is not a habitable building as defined by the state building code.” ‘Structure,’ habitable, and ‘building’ all have specific meanings that another engineer would understand, even though people not conversant in the specific meanings intended would be unlikely to grasp the same meaning.

As one of the great champions of words with specific meanings, although as a force for good, I offer John Wilkins, an English cleric, natural philosopher, courtier, and writer from the 17th Century. As a testimony to Wilkins’ openmindedness and willingness to adapt to changing circumstances, he married Oliver Cromwell’s sister Robina during the Commonwealth years, and though this should have tarred him indelibly with Cromwell’s sins, Wilkins bounced back after the Restoration in 1660 and served the Stuart establishment from 1662 until his death a dozen years later, being appointed Bishop of Chester in 1668. Wilkins owed much of his success in this transition to his desire for ‘comprehension’ in religious circles; while not quite religious tolerance, Wilkins’ hopes aligned with the policy goals of King Charles II in wanting to bring Independents, Presbyterians, and the Church of England together on points of commonality as part of the post-Restoration religious settlement, in the hope of maintaining domestic tranquility.

In his copious free time, Wilkins was instrumental in the founding of The Royal Society of London for Improving Natural Knowledge in November 1660; originally an club for inquisitive gentlemen to meet and talk about interesting experiments without being divided by religion or politics, during its heyday the Royal Society claimed as members not only Wilkins but Robert Hooke, Henry Oldenburg, Robert Boyle, John Locke, Christopher Wren, Samuel Pepys, John Evelyn, Isaac Newton, and William Petty, the least-known of the crew, but who virtually invented land surveying.

In his An Essay towards a Real Character, and a Philosophical Language (1668), Wilkins propounded a standard language for what might now be termed technical purposes, scientific research and philosophical discourse, in order that all the participants in a discourse understood what all the words mean.

In the 17th Century, writers enjoyed considerable latitude in language, even considering that most works were written in Latin for most of the period, and most of the scientific texts of the day are by necessity awkwardly studded with definitions of terms. One of the most awkward is Rene Descartes’ Le Monde of 1633, in which he defines matter as 'I conceive its extension, or the property it has of occupying space, not at all as an accident, but as its true form and essence,' though in doing so he in turn leans on two terms out of classical philosophy (true form and essence) that would have been recognizable to anyone with a contemporary Latin and Greek education, but opaque to anyone else, and which themselves embodied specific assumptions and preconceptions.

The problem was that Latin, French, and Greek, the lingua francae of the educated at the time, lacked the vocabulary or syntax to manage many of the topics that came floating to the top of the late 17th Century’s intellectual fermentation. Names of things proved a further problem; Carl Linnaeus wasn’t born until over thirty years after Wilkins’ death, and there was simply no standard system for naming and classifying plant or animal species, let alone rocks, stars, chemicals, mathematical concepts, or any of the other things that interested Wilkins’ intended audience. Old things proved enough of a challenge; in Beck's Bible of 1549, the lamentation in Jeremiah 8:22 was translated “Is there no treacle [molasses in American English] in Gilead?” rather than the correct rendering of ‘treacle’ as ‘balm’ or ‘solace.’

One outstanding example of the problem of terminology is that Sir Isaac Newton’s original narrative of the methods of calculus in the Principia, for example, are largely opaque to the modern reader (even if one comprehends the equations and diagrams) simply because he’s using different words for familiar concepts. In fact, Newton called calculus “the Method of Fluxions,” with the term “fluxion” referring to differential calculus and “fluents” referring to integral calculus.

Wilkins’ Philosophical Language was intended to provide a common set of terms for all of the contemporary natural philosophers to use, so that time wouldn’t be wasted and rivalries spawned by differences of opinion over what Leibniz’ term ‘monad’ meant. Along with this ‘language,’ Wilkins created a new, sui generis ‘alphabet’ (the Real Character), in which each character presented a specific defined concept, and an elaborate system of classifying things and concepts, rather like Linnaeus later did with plant and animal life, and a system of measurement that would have enabled correspondents in different countries to avoid the confusions caused by each using his own local system of measurement.

The attempt towards self-evidence eventually went recursive for poor Wilkins, though, and when describing his proposed unit of length, he had to include the entire method for calculating it. Unsurprisingly, he did so in using terms that are essentially a digestion of Euclid, which anyone hoping to call himself a natural philosopher in Wilkins’ day would have at least tried to comprehend:

...which being done, there are given these two Lengths, viz. of the String, and of the Radius of the Ball, to which a third Proportional must be found out; which must be as the length of the String from the point of Suspension to the Centre of the Ball is to the Radius of the Ball, so must the said Radius be to this third which being so found, let two fifths of this third Proportional be set off from the Centre downwards, and that will give the Measure desired. [A translation for mathematicians and engineers: d is the distance from the point of suspension to the center of the bob, r is the radius of the bob, and x is such that d/r = r/x. The unit of measure is the result of d+(0.4)x]

The result is, incidentally, 39.25-inches, or almost exactly one meter.

As quixotic, artificial, and overly-complicated as it might seem in hindsight, Wikins’ work could have had real value if it had been adopted widely. In the event, however, it was simply too much to ask for his contemporaries all to make the switch. Given how rapidly some of the sciences advanced after Wilkins’ death, however, it is possible that even if it had been more widely used it would have sooner or later become obsolete itself through the same lack of vocabulary and syntax to describe new things that plagued Latin writers of Wilkins’ day.

The purpose of Newspeak is control over expression, and therefore over thought. The role of the Philosophical Language was to be a common vocabulary to improve the state of scientific and philosophical discourse.

Although the means are similar, the contrast between the two ends could not be sharper.

Of course, when I say “sharper,” what do I really mean by that?

I mean that you should watch this Fry & Laurie clip.

As a footnote to the topic, one of Wilkins’ other significant works was intended to be used to conceal information rather than to reveal it. Mercury, or the Secret and Swift Messenger (published in 1641), was the first English-language book on cryptography, or the use of codes and ciphers. If nothing else, the difference between his 1641 and 1668 publications exemplifies the worlds of distance between the tense, conspiratorial, and militant England of 1641, on the cusp of the First Civil War, and the more relaxed and urbane climate of Restoration England in 1668 (at least, once the Dutch War, plague, and Great Fire of London had passed). Cryptography was a necessary skill in 1641, when a gentleman really did not want people from the other side of the incipient war reading his mail. In the halcyon days of the Royal Society, though, cryptography was less important than the ability to broadcast one’s discoveries to the rest of Europe, largely to ensure a prior claim on the discovery (c.f. the decades-long feuds between Newton and Hooke, or Newton and Leibniz).

The Littoral Combat Ship Revisited

A couple years ago I blogged an essay here about the US Navy’s Littoral Combat Ship program, and how it epitomized the United States’ badly managed defense procurement system. This post is a brief update, recounting what’s happened since then.

As a recap, the LCS program was a mess from the start, with grotesque cost overruns, contractors given leave to write checks on the government’s behalf, and most bizarrely, building two different models of LCS. In hindsight, the choice to invest in both the Lockheed Martin team’s LCS-1, USS Freedom and the General Dynamics team’s LCS-2, USS Independence, still looks like what it resembled in 2004: a crude attempt to spread the pork around as many Congressional districts and defense contractors as possible.

This is sad irony because the Navy’s public mantra for procurement at the time was “faster, better, cheaper,” and an attempt to get away from the gold-plated supership projects left over from the Reagan era. Two examples of these are the bloated CG-X/DD-X/DD-21/DDG-1000 destroyer project which is finally reached the prototype stage as the three-ship Zumwalt class, and the Virginia-class nuclear attack submarine that was supposed to be a cheap alternative to the Seawolf, but which wound up costing just as much.

Unfortunately, under Secretary of Defense Rumsfeld, “faster, better, cheaper” was implemented as farming the whole project out to contractors on cost-plus contracts that essentially gave the contractor a blank check, and allowing the ‘cheap’ LCS to mutate into a smaller version of a gold-plated supership. The final cost of USS Freedom was $645 million, and that for USS Independence was $704 million, exclusive of repairs and modifications needed after trials. This is 3 to 3.5 times the original budget for each ship, which quite defeated the point of the LCS as an inexpensive ship for primarily low-intensity conflict.

The inability to get what taxpayers pay for is not just an appalling waste of money, but also a major weakness in the US military. Being ‘strong on defense’ isn’t just a matter of spending money on defense; it must include ensuring the money is wisely and properly spent. The longer a weapon, ship, or aircraft lingers in development hell, the more likely it is to be cancelled. The money already spent on the project is gone forever, and the military has to do without the ship or missile or vehicle it needed—for example, in the late 1990s the US Navy had to bite the bullet and do without the replacement for the A-6 Intruder bomber, which essentially halved a carrier’s strike radius, and the Marines continue to make do with elderly, lumbering AAV-7 amtracs now that the Expeditionary Fighting Vehicle has sunk.

Right now, with the fleet stretched thin, the Navy really needs the LCS, and needs it to work properly.

Sighted Scam, Sank Same

The LCS almost went by the boards the same way. After four years of delays and cost overruns, in November 2007, Navy Secretary Donald Winter cancelled the LCS program outright with a vitriolic outburst at the two teams of contractors, in the culmination of a summer of discontent that saw the LCS program manager fired, the admiral in charge of shipbuilding reassigned to other duties, and work stopped on two further ships (LCS 3, USS Fort Worth and LCS-4, USS Coronado) that were under construction.

This wasn’t the only shipbuilding program that had pitted the Navy against the Navy’s contractors. Only a few months previously, Secretary Winter had blasted Northrop-Grumman (the Navy’s largest shipbuilder), Raytheon, and the other contractors involved in one of the Navy’s other projects, the mechanically-defective and grossly over-budget ($1.68 billion, $840 million over budget) USS San Antonio. The San Antonio was not only three years late entering service, but was $840 million over budget and plagued by humiliating mechanical and electronics failures, some of which were so egregious that they were introduced as evidence at the court-martial of one of the ship’s officers after one of the ship’s crew was killed in a machinery-related accident. As of October 2010, the Navy faced another $38 million in costs to repair defective work and make the ship usable. Winter complained "Twenty-three months after commissioning of LPD 17 [the USS San Antonio], the Navy still does not have a mission-capable ship. As of April 2011, the San Antonio had been in dockyard hands for a year for engine repairs and other concerns.

USS San Antonio

The USS San Antonio, at sea

Secretary Winter decried lack of competition in the shipbuilding market, an industry where virtually every major defense contractor has a piece of any given project, and specifically complained about the ‘eroded expertise’ of the government in warship design and construction, and argued that excessive delegation of design and construction responsibilities to the private sector—part of the privatization campaign of Secretary of Defense Rumsfeld-- was the root cause of the problem, since it stripped the Navy of most of its oversight of the design and construction of the ships it was paying for.

Rather ironically, Secretary Winter was himself a Northrop-Grumman executive from 2002 through 2005, after Northrop brought out the firm TRW, of which Winter was CEO.

This behavior from the Navy’s leadership was as out of character as a temper tantrum in a Trappist monastery, and the defense contractors who had been happily feeding off of Uncle Sam’s largesse were understandably shocked.

Indeed, the contractors were so shocked that they agreed to Secretary Winter’s fixed-price demand—and not only that, cut their prices in attempts to underbid each other. It was nothing short of hilarious to see corporations like General Dynamics so gobsmacked at having to accept the free market where they had expected a monopoly (or a kleptocracy, depending on your point of view).

The end result, announced in November 2010, was a resounding win for the government and the Navy, though-- if the Navy would have more LCSs, they’d be built at a fixed cost, with the builders eating any overruns. Granted, it’s 20 ships rather than the 55 originally called, for and the fixed cost is reported to be $450 million per ship (just shy of a Congress-mandated cap of $460 million each), more than twice the original planned cost. It’s still less than the cost of the prototypes.

Most importantly, in 2010 the Navy was finally allotted sufficient funds to for enough Supervisor of Shipbuilding officers for all ships under construction, in order to verify that ships were being constructed to specification.

Nonetheless, Admiral Mike Mullen, Chairman of the Joint Chiefs of Staff, said in October 2010 “If LCS is unable to contain itself … then I don’t think it has much of a future.” Further problems are entirely possible, including issues that could render the design useless, and the performance of the first several ships will be taken into account.

The rather questionable thing is that after the shouting died down, the Navy decided to go ahead with both designs after all, ordering ten ships of each type in addition to the four completed or under construction, all to be paid for by 2015 and in service by 2018. The two LCS designs are distinct, it is true, but neither of them is significantly better than the other, and the trimaran Independence design costs more than the steel monohull Freedom type, although the Independence has more cargo space. Furthermore, the aluminum trimaran Independence design produced by the General Dynamics team (now headed by Austal) will be the first warship of its type to see service, which raises a whole new set of engineering challenges and maintenance concerns. The Independence also introduces a wholly new data and communications system, while the Freedom has an improved version of the Navy’s standard system. Construction work is also hindered by the two ‘teams’ having changed their rosters—the erstwhile General Dynamics design is now managed by shipbuilder Austal, and the shipyard at which the Lockheed design is being built was bought out by the Italian firm Ficantieri. To further complicate matters, Austal and General Dynamics have recently parted ways; Austal will build the next USS Independence-type (General Dynamics type) LCS competing with them. Still worse, although Austal built the Independence, General Dynamics is responsible for “warranty maintenance.”

USS Freedom

USS Freedom at sea

USS Independence

USS Independence at sea

Independence, Freedom, Corrosion and Time

Both prototype ships are operational--USS Freedom conducted her first mission in April 2010, and intercepted a narcotic smuggling boat off the Pacific coast of Colombia. This was well after the operational date intended at the inception of the LCS program, but well ahead of the pessimistic post-2007 deadlines. The Independence entered service in January 2010, but has to date accomplished even less than the Freedom.

Neither type of LCS is anything like perfect at present—they are prototypes, after all—but it begs the question of why the Navy accepted them for service at all, when both ships are plagued with grievous problems. Both Freedom and Independence suffered severe mechanical breakdowns during their first few voyages, and when the Independence completed her maiden voyage in April 2010 the Navy had to spend $5.3 million in repairs and upgrades to correct faults identified during sea trials, after General Dynamics had already spent several million dollars correcting faults already identified during builders’ trials. USS Freedom also turned out to be 6% overweight compared to her design specifications, with a significant shortfall in the reserve buoyancy that would keep her afloat in the event she is damaged, as well as serious questions about whether she can actually manage her designed speed.

USS Independence, meanwhile, suffers from severe corrosion after only a few months in service, thanks to the electrolytic disintegration of her aluminum hull, and now needs to be dry-docked for massive repairs. The cathodic protection anticorrosion system originally planned was eliminated as a cost saving measure. The Navy blames Austal’s supposedly faulty construction, and Austal blames the Navy’s supposedly negligent maintenance, though given how briefly the Independence has been in Navy hands raises the question of how fragile the ship is, if only a few months of ‘negligence’ is enough to imperil the ship’s seaworthiness and structural soundness.

It is normal for ships to require a fair bit of adjustment and modification after entering service, to address concerns identified during builders’ trials or sea trials—after all, that is the purpose of trials. The Perry-class frigates which the LCS is intended to replace, for example, suffered structural cracks up to 40-feet long in the early ships of the class, and the Perrys still active have served well for over thirty years. The severity of the LCS’s problems, however, far outstrip most of the concerns identified in previous generations of warships.

So now that we will have at least some LCS in the fleet—late and grossly overbudget, but presumably operational—will the LCS actually be worthwhile?

Both types of LCS face intense criticism on their technical merits, in addition to fulmination from more conservative big-ship advocates and flattop boosters within the Navy. The LCS program as a whole was controversial in Navy circles, if for no other reason that the LCS being the smallest seagoing warship (other than minesweepers and the like) the Navy has ordered since the 1960s. The US Navy has historically preferred bigger ships to smaller ones, when all other things are taken into consideration—if the fleet needs a destroyer, it gets a big destroyer.

There are three main criticisms of the LCS, besides cost: armament, survivability, and seakeeping, and many of the criticisms of the LCS take the form of “it’s too small, lightly armed, and flimsy—throw it out and build a Perry-class frigate.” This is decidedly ironic, since old naval constructors will doubtless remember some of the criticisms leveled against the Perry class itself when it premiered in the 1970s—“it’s too big, slow, and lightly armed—throw it out and build a Charles F. Adams-class destroyer,” or the puzzled Congressional reaction to the Spruance-class destroyers in the early 1970s—“where are all the guns?” Admittedly, the Spruance type more than twice the size of the previous class of destroyer, had less armament, and had a superstructure that looked like a truck garage.

The general sneers from the big-ship community aside, some of the technical criticisms of the LCS designs are very significant, and should not be dismissed lightly.

The Swiss Army Warship

Concerning question of armament, granted, the factory-fresh LCS is very lightly armed when compared to foreign designs of similar tonnage (e.g. the German-designed ‘MEKO A’ family used by a number of navies around the world), with only a 57mm autocannon, a close-range surface-to-air missile system, and some machine guns. In fact, many Fast Attack Craft (the antiship missile counterpart of the torpedo boat) carry more armament than a LCS on a hull 1/6 the size.

Unlike the MEKO type, however, it’s hard to say what the LCS’ real armament would include since the basic design concept is for interchangeable “mission modules.” The basic idea is that a LCS could be equipped with antiship missiles one week and then swap them out for minesweeping gear the next. Add a mission module with a couple dozen vertical launch missiles to an LCS, and it will no longer seem so under-gunned. The concept is actually fairly well-proven. The Royal Danish Navy has had success with this model since they invented it in the 1980s, although they’ve never had to use it in combat, but it’s a first in the US Navy. Both types of LCS would also operate unmanned aerial vehicles and the other types of drones that have become so widely used in the last ten years.

The Navy originally intended to buy for the whole LCS fleet a total of 16 antisubmarine modules, 24 antiship or land-attack modules, and 24 mine countermeasures modules that would be swapped out between ships depending on what the ships were being used for. Unfortunately, the antiship/land attack missile planned for the LCS, the N-LOS program, was cancelled in 2010, and so it’s not clear what the LCS would use for antiship or land attack weapons. The minesweeping and antisubmarine modules haven’t finished development either, so the LCS will spend at least a few years making do with what it was built with, as a very large speedboat and self-propelled helipad.

So far, the design is intriguing. The 55 LCS originally ordered were intended to replace 14 Avenger-class minehunters, 12 Osprey-class coastal minehunters, and 30 Oliver Hazard Perry class Frigates on roughly a one-for-one basis, covering a wide variety of rolls with one type of ship, offering great savings in manpower, maintenance costs, and long-term expenses, in addition to not having to pay upkeep costs on ships not actively needed at a given time. To put it rather more precisely, however, the LCS was conceived as a multimission ship and the Navy promptly stuck all the jobs in the ‘miscellaneous’ category onto it.

The idea of modularity is not new—the Spruance and Perry-class frigates were designed to accommodate additional weapons. The Spruances generally didn’t receive these until the late 1980s at best, when two dozen of them received 61-cell Vertical Launch Systems capable of launching cruise missiles and other weapons.

Both the Freedom and Independence types also have a substantial part of their displacement and internal volume dedicated to what is essentially flex space, that can be reconfigured based on need. In most warships of comparable size, this space is used for weapons, crew quarters, fuel, or stores. The ships have a disproportionately large hangar and large flight deck (50% larger than a Burke-class destroyer’s flight deck, and probably the single most useful thing about either LCS design) for two medium-sized helicopters such as the SH-60 Seahawk or one heavy-lift helicopter like the CH-53 Sea Stallion. A large chunk of the hull volume in either design is taken up by a ‘roll on/roll off’ cargo deck with a capacity of several dozen Humvees or light armored vehicles, or a company of troops in tight quarters.

This gives the LCS the ability to act as a (very) fast transport for Marine or Special Forces units, though admittedly this may prove to be a “too long for Dick, too short for Richard” capability. Landing a Marine Force Recon platoon in rigid inflatable boats is one thing, and can be done from virtually any ship in the fleet, but Seal Team Six isn’t in the habit of using armored vehicles on commando operations, and in any case the LCS needs a pier to offload vehicles from. In other words, the LCS may have more capability than it needs for routine helicopter and boat insertions, but not enough for bigger operations where vehicles are actually needed. If the mission is big enough that vehicles are needed, the Navy will likely assign a real amphibious ship to the job, and some of the cargo space built into an LCS would have been better used for other purposes.

Some of the missions for which the LCS is proposed seem wildly at odds with the ship’s design. In particular, the use of a half-billion dollar, 50-knot ship for minesweeping makes little sense, as minesweeping is usually done at less than ten knots using inexpensive ships that resemble fishing trawlers. The Avenger and Osprey class minesweepers currently in service are good ships and well-suited to their jobs, but although the Navy recognizes the minesweeper’s importance, minesweeping is one of the jobs the Navy really, really hates doing, much as the Army hates counterinsurgency operations. Using a LCS for minesweeping is using a proverbial sports car for a pickup truck’s job.

Freedom at sea

USS Freedom with an aircraft carrier and a guided missile cruiser on her first cruise


Most significantly, since the ability to withstand damage is a major concern for a warship of any description, both types of LCS appear decidedly frail when compared to either US warships or their foreign counterparts. Both types of LCS were initially designed to ‘commercial’ standards, rather than the significantly more robust standards customary for naval warships, and were re-engineered for the navy’s new and evolving set of naval standards during construction. A formal Pentagon evaluation released in March 2011 concluded that “LCS is not expected to be survivable in terms of maintaining a mission capability in a hostile combat environment,” or in other words that the ship could easily sustain enough damage to be put out of action. The Navy has, in fact, opted to forego the usual blast testing for prototype ships—necessary to evaluate the ship’s resilience to damage-- “due to the damage that would be sustained by the ship.” That sounds like an alarming lack of confidence in the ship’s ability to keep sailors alive in battle. The Congressional Research Service confirms these findings.

Both LCS designs are engineered for what the Navy terms “Level 1 survivability” – in other words, expected to operate in the least severe combat environment and not expected to "fight hurt." This is the level of durability you build into something cheap and expendable, like a harbor patrol boat or a coastal minesweeper, not something you expect to carry commando teams in on the coast of Iran. If damaged, a Level 1 ship will be out of the fight and will probably sink, but will last long enough for the crew to get off. By contrast, Burke-class destroyers and pretty much every other surface warship in the US Navy are engineered for Level 3 survivability—the ability to take severe damage, not sink, put out fires, and keep shooting and steaming.

The Royal Navy learned a great deal about the abrupt destruction modern naval warfare can bring—HMS Sheffield was destroyed by a single Exocet missile, and a number of other British warships were also sunk or crippled by one or two hits from missiles or bombs. “Everybody had always said that modern warships are ‘one-hit ships,’ ” Sheffield’s captain observed after the attack. “Nobody had thought about the implications of a ‘one-hit ship’ 8,000 miles from home.” (Battle for the Falklands, Hastings and Jenkins 1984, 155). By the end of the conflict, the Royal Navy had barely held out through a grueling war of attrition against Argentine airpower, and the sheer number of ships lost had nearly cost the UK the victory. The USS Stark, a Perry-class frigate hit by two Iraqi Exocets in 1987, was seriously damaged but survived. Both HMS Sheffield and USS Stark were much larger than the USS Freedom or Independence, which as much smaller Level 1 ships would probably go right to the bottom.

HMS Sheffield

HMS Sheffield, 1982

USS Stark

USS Stark, 1987

In view of the LCS’ proposed role, of fighting along coasts, within reach of missiles, mines, airplanes, other peoples’ warships, and even plain old gun artillery on the shore, the flimsiness and weak armament of the LCS designs is nothing short of ludicrous. The LCS are supposed to be ‘stealthy,’ with low radar and infrared signatures, but stealthy sometimes only goes so far—the Yugoslav Army managed to shoot down a F-117A Nighthawk “stealth fighter” during NATO’s 1999 intervention in Kosovo—and anything billed as a warship should be expected to suffer damage.

Still more importantly, though, is the question of how the Navy can in good conscience send sailors into a combat zone in a ship not designed to survive? The survivability issue is a big part of the US Navy’s prejudice against smaller ships, since larger ships can have more reserve buoyancy and other design features that make them more likely to survive damage than a smaller ship suffering with the same damage.

The Boat Had Better Float

Seakeeping ability is also a major concern. LCSs are intended to spend a lot of their time on patrol missions—hunting pirates, terrorists, smugglers, or mines—and the ability to stay at sea for long periods of time is important. The Navy’s big destroyers and cruisers were designed for the expanse and rough weather of the Pacific and North Atlantic, can steam and fight in anything short of a hurricane, and can stay at sea for as long as the machinery, food, and fuel oil hold out. The fast, nimble, and lightweight USS Freedom’s main engines are fuel-guzzling gas turbines which are “essentially the engines of a 777 jetliner,” and generates more horsepower than the powerplant of a much larger Ticonderoga-class cruiser. Her top speed is alleged to be nearly 50 knots, making her one of the fastest warships afloat. On the other hand, she has a maximum range of a mere 3,500-miles at a poky cruising speed of 18 knots on her diesel engines (half that of a typical frigate or corvette of the same size). That may sound like a lot, but remember that a ship has to be burning fuel and moving anytime she’s not at anchor. Freedom’s gallons-per-mile appetite also means that she could potentially go through her entire onboard fuel supply in a half-day’s high-speed running. As fast as the 50-knot Freedom is, if she wants to stay moving she’ll be forever dependent on a 20-knot fleet oiler to keep her tanked up. Current government estimates hypothesize that fuel costs will account for between 8% and 18% of the Freedom’s total operating cost over her operational lifespan. It also remains to be seen how well the Independence and Freedom will handle heavy weather, and although they’re not intended to do high-seas hunting for Soviet submarines in a North Atlantic winter, Navy warships have a two-century-long tradition of being used in roles and conditions other than what their designers expected.

Crew size, though one of the LCS program’s main selling points from the Pentagon’s point of view, is also a potential weakness. In an effort to reduce manpower needs (and the related costs) the LCS has a baseline crew of just 40 officers and men, compared to the 150-200 on a typical frigate. Although other personnel would join the crew to operate helicopters and specialized mission module equipment, for a total of perhaps 75 personnel, as it is the ships are shorthanded enough that crew coming off watch have to do double-duty even for routine tasks like refueling. On the trial cruises, this left many personnel functioning with very limited sleep. How this limited manpower would impact the ship’s company’s abilities in a combat zone, when prolonged sea duty and high-intensity operations would push the exhaustion limits, remains to be seen.

Ultimately, the LCS appears unlikely to live up to the expectations for it, in the absence of the necessary multirole equipment. In any case, it is not likely to prove a gamechanger Although small by US Navy standards, it is probably too large to be cost-effective in operation. It is too lightly constructed to survive a real war, and its armament is likely to prove inadequate. Worst of all, the cost overruns and bureaucratic disasters that have accompanied it have damaged the entire procurement process, to the point where it is a valid question whether the US Navy can actually get a usable warship on-budget and on-time. In the case of the LCS, the Navy struck out on all three pitches.

Systemic Dysfunction

In the end, this is 2011, and the US has to support two prolonged wars in two different theaters, as well as maintaining commitments in the western Pacific (particularly keeping an eye on North Korea) and other parts of the world. The economy is fragile, the government’s budget is strained, and ideological differences are splitting even the normally lavish-on-defense Republican party over the F-35 Joint Strike Fighter.

Defense cuts will continue. In the current economic and political climate, that is a certainty. Secretary of Defense Robert Gates was thoroughly justified in selecting most of the defense programs he proposed cuts for in April 2009 and January 2011, mostly because the projects cut, like the F-22 Raptor stealth fighter, are not relevant to the nation’s current needs. There are more economies to come, but twenty ships at a flat fee, even if they aren’t perfect, are better than no ships at all.

That in itself is one reason the Navy has fought as hard for the LCS as it has—whether they’re good ships or mediocre ships, the Navy needs ships, even if for no more complicated reason than replacing ships that were launched during the Nixon administration and that are now worn out.

Still, the simple truth is that the US Navy paid a combined $1.35 billion for just two warships, neither of which is actually in operational condition. That same sum, incidentally, should have paid for six ships if the program had managed to stay on-budget. The French Navy, meanwhile, managed to get the first of their Gowind-class corvettes, the L’Adroit, from the drawing board to sea trials in only two years. By contrast, the Independence took seven years (nearly as long as the aircraft carrier George H.W. Bush) and still isn’t operational. The Gowind is smaller and slower than either LCS design, but it’s perfectly serviceable and capable of performing many of the same missions. This speed is also an example of how painfully sclerotic the US Navy’s design and construction system has become. In the current economic and political climate, the longer a project lingers in development hell, the more likely it is to be cancelled. The Navy needs a ship that is what the LCS was originally supposed to be—simple, versatile, inexpensive, and possible to build and bring into service quickly and in large numbers.

In broader terms, the LCS program is an ominous foreshadowing of what could happen with other warships. Aside from the LCS planned, the Navy has approximately 50 other warships on order or under construction for the 2012-2016 period, including three DDG-1000 Zumwalt-class destroyers optimized for land attack missions, six more Arleigh Burke class guided-missile destroyers, several Joint High-Speed Vessel fast transports, several amphibious assault ships, three new fleet oilers, three nuclear-powered aircraft carriers, and five nuclear-powered attack submarines. All of this adds up to a lot of money, time, manpower, and R&D—hundreds of billions of dollars spread over years, and most of it going through the hands of a fairly small number of prime contractors like General Dynamics.

Sadly, even though about 50 other ships are under construction, that’s not even treading water. As it is, when Reagan-era ships such as the Ticonderoga-class cruisers, some Los Angeles class submarines, and the remaining Perry-class frigates reach the ends of their effective lives by the early 2020s, the Navy’s fleet will shrink by about 70 ships, or about a quarter of its current strength. The Navy is already shorthanded; although the number of missions is increasing every year, the fleet’s strength has actually declined 18% from a 2001 strength of 337 ships to a current effective strength of 285 warships, compared to a goal of 313 ships as described in a recent Congressional Research Service report and numerous Department of Defense publications. This 313-ship fleet is a long way from the pie-in-the-sky 600-ship goal of the Reagan years, but is more in line with what the Navy actually needs. The problem is in getting there, and staying there.

A navy will never have 100% of its ships in tip-top shape all at once. Right now, however, about 20% of the warships in the fleet are not combat-ready—up from only 8% in 2007—due to a combination of hard use (ships used more wear parts out quicker, and US Navy ships spend more time at sea than those of most other navies), sheer age, and deferred maintenance ($173 million worth); the Navy gets five billion dollars per year for maintenance, upkeep, and repair, which covers everything from aircraft engines to hull paint, and even $5 billion doesn’t cover everything. As it is, deployed ships on average spend nearly 40% of their time while on deployment—which could mean potentially being in harm’s way-- with at least one major system on the fritz. These system failures could include something as vital as the ship’s main gun, sonar, communications system, or missile launch and guidance controls, failures which could leave the ship undefended and endanger the lives of the sailors aboard. This lack of preparedness and the sheer numbers of handicapped warships has been an open secret for several years, but has become a matter of Congressional attention.

Even new ships need repairs—in February 2011, the USS Gravely’s topmast broke off while the ship was cruising off the coast of Florida. While accidents happen and things just break sometimes, the Gravely was only commissioned in 2010, making her practically brand-new, and this collapse raises serious questions about shipyard quality control.

The main problem is that more than half of the current fleet—from aircraft carriers on down to frigates and transport ships-- dates to the Reagan era if not earlier, and older ships require more maintenance. One reason these older ships (notably the Perrys) are still used so intensively is that their replacements have been cut from the budgets or delayed in design and construction. Another reason for intensive use of older ships is that manpower and budgetary demands during the late 1990s and early 2000s compelled the Navy to scrap entire classes of warships wholesale (for example, the 31 Spruance-class destroyers) at an accelerated rate between 1998 and 2005, faster than the ships were being replaced by new construction.

The problem is even worse for aircraft. Fifty-five percent of deployed Navy and Marine Corps aircraft are not completely mission-capable. Given how much the Navy and Marines depend on aircraft for everything from strategic reconnaissance to bomb-dropping to medical evacuations, this is a grave problem, part of what Congressman Larry Kissell phrased as “multiplied shortness there of everything that you could want or imagine or need with that ship… for every ship, then, what comes with that ship, we're missing?”

In the meantime, the number of missions needing warships has increased, thanks to two wars, a UN intervention, an anti-piracy campaign, and humanitarian commitments on several continents, on top of routine demands like training, while the number of available warships has steadily shrunk. On average, approximately 40% of the fleet’s strength is currently on deployment, twice what it was in 2000. Vice-Admiral William Burke testified before the House Armed Services Committee that the Navy high command considers this rate of deployment to be unsustainable.

There is no simple fix to this systemic failure. Over the last twenty years, the numbers of ships ordered have fallen, while prices per ship have skyrocketed, technical failings have multiplied, and delay has piled on top of delay, stretching into years of scheduling and cost overruns. Even when ships entered service, they are frequently found to be defective; this is one reason the Littoral Combat Ships haven’t been deployed for combat duty or even long-distance noncombatant operations, since the Navy is concerned that the ships could suffer serious and embarrassing breakdowns.

In other words, the Navy is putting money into the system but not getting usable ships out of it, but still has to play coverup on ship flaws because otherwise the ships will be cancelled. Bad ships, from a certain perspective, are better than no ships at all. The other armed services have much the same problem with their suppliers, but have not been so bluntly outspoken about the problem.

On the whole, there’s a hole.

Too many military development programs have collapsed over the last twenty years due to abuses and waste in the design and procurement process. Some of these programs were poorly conceived or justified—for example, the Crusader self-propelled gun. Others were badly needed but poorly-executed (the Marine Corps’ Expeditionary Fighting Vehicle comes to mind), but for a variety of reasons the project died of bloat and budget overruns. The program that eventually produced the Zumwalt-class destroyer hemorrhaged money for well over a decade before the Navy decided it couldn’t justify the program after all, and elected to build only three ships.

One hint at the future is the USS Florida, one of four Ohio-class ex-ballistic missile submarines originally built to sling nukes during World War III, but retrofitted to carry 180 Tomahawk cruise missiles instead of ICBMs. The Florida essentially shut down Libya’s air defense network overnight in March 2011, making her the most useful asset in the conflict to date, in sharp contrast to the F-22, which is for all intents and purposes useless for the sort of missions conducted in Libya or similar conflicts. A massive countrywide strike like the Florida’s is a job that even during the first Gulf War in 1991 would have required at least two aircraft carriers and their escorts, and which would have taken several days of sustained air attacks. The future will likely see increasing pressure to use more makeshifts, improvisation, or repurposed ships in order to fill needs, rather than relying on new construction.

Secretary Gates, wisely emphasized discarding decades-old gold-plated projects like the F-22 in favor of simple things that actually work, that were immediately relevant to the actual missions in Iraq and Afghanistan, and that are readily transferrable to other jobs in other places. The outstanding example of this policy is the use of unmanned aerial vehicles, which has grown by leaps and bounds since Gates entered office in 2006, and which has become a military staple in only a few years.

All parties—the Pentagon, the defense contractors, Congress, and the public—should be prepared to be very hard-nosed and pragmatic about defense spending and contracting. The era of blank-check contracts and 100% cost overruns is over. If the Pentagon expects to get its money’s worth in a harsh economic climate and with defense budgets already creaking under the strain of two wars and dozens of other operations around the world, it needs to mind its pennies just as taxpayers do.

Wednesday, July 6, 2011

Are Aircraft Carriers Obsolete?

Are aircraft carriers becoming obsolete?

That is an interesting and complex question.

In the current issue of the US Naval Institute’s Proceedings, Captain Jerry Hendrix, USN and Lt. Col J. Noel Williams, U.S. Marine Corps (Retired) argue that the aircraft carrier is in its twilight, faced with being eclipsed by long-range missiles and sensors (“this battle of signatures and long-range strike”) that threaten to ‘outrange’ the carrier and its aircraft in the same way that the carrier’s aircraft outranged the battleship’s guns. The authors look particularly towards China as a challenger to the US Navy’s control of the sea, and warn of a metaphorical ‘21st-Century Pearl Harbor,” meaning a situation in which the Navy is caught a technological step behind, a specter that has in one form or another haunted that Navy arguably since the CSS Virginia wrecked two wooden Union warships at Hampton Roads in 1862.


The USS Abraham Lincoln and her escorts

Hendrix and Williams float several propositions for the navy’s future, which they argue would help maintain control of the sea in the face of rising costs and potential technological upsets:

  • Smaller, 45,000-ton aircraft carriers at a cost of $2 billion each, compared to a $14 billion supercarrier (about $20 billion with aircraft). These smaller aircraft carriers would serve as a forward-deployed force that can be maintained less expensively than the supercarriers, with the ‘remaining supercarrier inventory’ available for a ‘surge capability.’ The authors imagine carriers similar in size to the Navy’s amphibious assault ships, 45,000-ton ships that have flight decks for helicopters and ‘jump jet’ (Vertical/Short Take Off and Landing, or V/STOL) Harrier fighters, which can take off and land vertically or in a short distance. In fact, Hendrix’s ideal sounds an awful lot like the forthcoming USS America, an amphibious assault ship designed around airborne assaults and that trades the usual well deck for launching landing craft for a big increase in aircraft capacity.

  • The proponents argue that with two squadrons of the future F-35B fighter-bomber aboard, a ship like the USS America could deliver an appreciable fraction of a supercarrier’s firepower for a much more reasonable investment in money, manpower, and risk. As the US Marine Corps intends to purchase the same aircraft, Marine squadrons could be used from carriers when not needed on land. In fact, their proposal depends so heavily on the F-35B and its ability to take off and land on smaller flight decks that without the future aircraft, their proposed ships would be nearly useless.

  • More reliance on unmanned aerial vehicles (UAVs), including the proposed UCLASS, which can operate from smaller ships than fixed-wing fighters can.

  • Cancelling the Littoral Combat Ship program and devoting the resources to sixty small amphibious assault ships of 10,000-tons displacement to serve as ‘utility infielders’ in baseball metaphor, supposedly “providing a tremendous platform for engagement missions and humanitarian-assistance/disaster-relief response at one end and amphibious operations and sea control at the other.”

  • Maintaining a light/heavy balance of forces, with smaller inexpensive ships deployed around the world, and a “heavy surge force” held in reserve, ready to deploy to trouble spots.

The proposition as a whole has some merit—supercarriers do represent an immense cost and potential loss packed into a single target. Still most of the authors’ proposals would carry more weight if they were actually new, but with the possible exception of the 10,000-ton multirole ships, they’re not. The Navy has been here before. Smaller aircraft carriers have been proposed a number of times in the last few decades—most notably the Sea Control Ship of the 1970s, championed by the late former Chief of Naval Operations, Admiral Elmo Zumwalt. All of these foundered on the reefs of institutional prejudices and operational expectations—the Navy didn’t really want them and couldn’t fit the concept into its plans well enough to satisfy Congress.

The first major shoal in the waters of institutional prejudice is that the Navy has for the last six decades usually detested the idea of small ships, a category which in terms of surface ships came to include virtually everything smaller than a 5,000-ton guided-missile destroyer. The Navy complained that smaller ships were too limited in their abilities, couldn’t be upgraded with new weapons, required too many men and too much money, and so on, and generally argued that the money would be better spent on a smaller number of big ships. The same is true of aircraft carriers—if you want one at all, the Navy would argue, then you should get a big one.

Part of this attitude stems from genuine apprehension at the thought of sending men to sea in ships that were unequal to circumstances (witness the furor over the fragile and lightly-armed Littoral Combat Ship), which is why the Navy has never really embraced the idea of the high/low or big/cheap split, where a core of high-quality ships was supported by larger numbers of smaller, more expendable units. The archetypical example was the Victorian-era Royal Navy, which had two main battle forces, the Channel Fleet and Mediterranean Fleet, that comprised most of the fleet’s most powerful ships, supported by swarms of less-expensive cruisers and gunboats on overseas stations. The US Navy flirted with the ‘high/low’ concept during the 1970s, but even the ‘low end’ ships were fairly large and complex. This high/low scheme is, of course, the exact model proposed by Hendrix and Williams, and raises the possibility that the lightweight forward-deployed forces could find themselves outgunned and alone in the event that a major regional conflict erupts.

The second institutional prejudice is the issue of the aircraft carrier itself. Ever since the Second World War, the Navy has been dominated by the carrier clique, men wedded to the idea of the aircraft carrier as the ultimate Big Stick. By the 1960s, this attitude had been taken to such an extreme that virtually every other kind of ship in the fleet was reduced to supporting the aircraft carrier—cruisers became single-purpose air-defense ships that shot down enemy airplanes or cruise missiles, and destroyers and frigates became single-purpose ships that sank enemy submarines. Either type was basically a defensive auxiliary to the aircraft carrier, and were ill-equipped for operating out of their intended roles, or even for the task of simply fighting other surface ships—the Navy didn’t even have a purpose-built antiship missile until the late 1970s, when the RGM84 Harpoon entered service.

In the particular case of aircraft carriers, bigger being better was embraced virtually as a universal constant, since the US Navy’s supercarriers were essentially designed around their aircraft. As built, they embodied certain engineering assumptions about the aircraft in service at the time, and other assumptions about aircraft expected to enter service in the next decade or so. By comparison, the rest of the ship was there to move the flight deck around. If you want to operated supersonic jet fighters, you need a carrier of a certain size, dictated by concerns for hangar space, length of flight deck for takeoff and landing, and so on. The Navy’s fleet of amphibious assault have a similar problem — the MV-22 Osprey tiltrotor aircraft won’t fit on most of the ships built during the late 1980s and early 1990s.

The reason for this bigger-is-better train of thought was simple—the US Navy spent half a century preparing to fight a war that never came to pass, only to be faced with a series of missions that it had never really contemplated. During the expected World War III, aircraft carriers would have had the unenviable tasks of sweeping the seas of Soviet ships and aircraft, and of hammering their way into the Warsaw Pact’s defended airspace. This required aircraft at least as good as their land-based counterparts, and a carrier with the capacity to launch major airstrikes. Both were major challenges. Since the war with the Soviets was the only thing on the radar (sometimes nearly literally so), the whole system of carriers, aircraft, training, tactical doctrine, weapons design, and contingency planning was focused on fighting the Soviets in certain areas under certain assumed conditions, and to the exclusion of everything else. The F-14, for example, was designed primarily to kill Soviet bombers with super-long-range Phoenix missiles weighing half a ton each, thereby hopefully preventing the bombers from getting close enough to the carriers to launch their own missiles. The result was a very large fighter plane engineered for a specific role.

The Navy’s original goal in designing its carrier fleet was to operate aircraft from a carrier that were equivalent to their land-based counterparts—for example, the F-4 Phantom II (one of the first combat aircraft to be extensively operated by both the Navy and the Air Force) or the F-14 Tomcat. Carrier-borne aircraft had previously been considered inferior to land-based aircraft because of the compromises in design necessary to fit them onto a carrier—weight and size were limited by the hanger size and deck length of the carrier. For example, the Grumman F4F Wildcat, the Navy’s mainstay during the early years of the Pacific War, was slower, shorter-ranged, and less well-armed than the Army Air Corps’ contemporary land-based Curtiss P-40 Warhawk. This general trend persisted into the 1950s as jet technology matured, with the Navy’s stubby little F9F Cougar lagging behind the Air Force’s hotshot Century-series (F-100, F-101, F-105, etc) fighters, but later advances leveled the playing field between carrier and land-based aircraft.

As combat aircraft evolved and got bigger in the pressure-cooker environment of the Cold War they needed larger aircraft carriers; the first ‘supercarrier,’ USS Forrestal, launched in 1951, displaced 59,650 tons standard (81,101 tons full load), had an overall length of 1,067 feet and a beam at the flight deck of 238 feet. She drew 37 feet of water and had a crew of 5,540 personnel. The Forrestal was an immense ship for her day, over twice the displacement and 25% longer than the Essex-class carriers built during the Second World War, and which made up the bulk of the Navy’s carrier fleet during the 1950s and 1960s. The never-built USS United States of 1948 would have been larger still, but was cancelled after the end of the Second World War.


The Navy’s newest operational aircraft carrier, Nimitz-class USS George H.W. Bush, has a nominal displacement of 102,000 tons, a length of 1,092 feet, and a beam at the flight deck of 252 feet. She draws 41 feet of water and has a crew of 5,680 personnel. The forthcoming USS Gerald R. Ford class will be at least as large.


In hindsight, it is perhaps surprising that the supercarriers haven’t grown more than they have, to keep pace with aircraft sizes—the Nimitz is bigger than the Forrestal, but not by that much; much of the increased displacement comes from her nuclear power plant. Rather, it is a credit to the naval architects’ foresight in estimating future needs, and to the aircraft engineers’ skill in fitting new aircraft into old ships, once a practicable maximum size of an aircraft carrier had been reached. The Forrestal served until 1993, deploying with several successive generations of aircraft, including the massive F-14, the largest carrier-borne fighter ever built, which was twice the size and four times the weight of the F9F Cougar that was entering service along with the Forrestal in the 1950s.


The F-14 Tomcat in the foreground aboard the USS America in the 1980s; the aircraft in the background are an A-7 Corsair II and an A-6 Intruder

Older, smaller carriers often couldn’t accommodate newer aircraft and had to be scrapped, or were limited to older, sometimes obsolete aircraft. The US Navy scrapped its herd of Second World War-vintage Essex-class carriers during the late 1960s largely because they couldn’t accommodate the then brand-new F-4 Phantom II. The Forrestal remained in service until 1993 because she had been designed to accommodate the Douglas A-3 Skywarrior, a jet-propelled nuclear-capable bomber (from the era of nuclear weapons the size and weight of a dump truck) that was bigger and heavier even than the later Tomcat, and so could handle even that enormous aircraft. The Forrestal even managed the largest aircraft take-off and landing ever, a USMC KC-130 tanker aircraft, in 1963)

Other navies faced the same problem. The French Navy’s Foch and Clemenceau, for example, were stuck with the Vought F-8 Crusader for a fighter until the late 1990s, decades after it had been retired from front-line service in the US Navy. The Royal Navy essentially gave up on conventional aircraft flown from carriers altogether when HMS Ark Royal was retired in 1979 and replaced with three “through-deck cruisers” (HMS Invincible, Ark Royal, and Illustrious) that couldn’t operate conventional fixed-wing aircraft, and which had initially been conceived as a mere escort to the never-built CVA-01 supercarrier. The Wilson and Callaghan governments’ stance on defense spending was so tight-fisted during the UK’s economic troubles of the 1970s that the Admiralty forbade the ships be described as ‘aircraft carriers,’ for fear of the ferocious opposition to defense spending in Parliament.

Other nations with carriers left over from the 1940s and 50s—for example, Argentina’s 25 de Mayo, Brazil’s Minas Gerais, or India’s two small ex-Royal Navy flattops, Viraat and Vikrant—were limited to helicopters, propeller-driven antisubmarine planes such as the Fairey Gannet or Breguet Alize, and obsolete fighter or attack jets like the A-4 Skyhawk. Even these small and limited carriers consumed an enormous share of a small navy’s resources, for all that the carrier may well have been a sort of tokenism, much as having even a single dreadnought was a considered to be a mark of status fifty years earlier.

Thus the argument for supercarriers. In short, if the operational priorities were high-intensity combat against Soviet land-based fighters and bombers, then dinky little World War II leftovers flying Gannets, Skyhawks, and helicopters were obviously nonstarters, and so for several decades the prejudice against small carriers made good sense. Small carriers were useful for protecting convoys against submarines—and indeed, helicopters remain the submarine’s greatest nemesis-- but were too limited to be useful for offensive operations.

The US Navy itself toyed with the concept of purpose-built smaller carriers in the early 1970s, as part of a general interest in lower-tech, less expensive weapons and methods that all four armed services entertained in the waning years of the Vietnam War, in the season of embarrassment resulting from the belief that the comparatively crude Soviet weapons had outperformed the costly, high-tech products of the American military-industrial complex. The Soviet Unions’ then-new Kiev-class ‘heavy aircraft-carrying cruisers’ were a further spur, in that they combined a flight deck with heavy antiship missiles and a formidable air-defense armament, making US carriers look lumbering and defenseless. Admiral Elmo Zumwalt, Chief of Naval Operations at the time, advocated a high/low mixture of inexpensive, single-purpose ships together with high-value multirole ships; some of the proposals generated under this initiative eventually saw daylight as the Perry-class frigates and the Virginia-class nuclear-powered guided-missile cruisers. The former were designed as new-generation convoy escorts, optimized for antisubmarine warfare, and the latter were conceived as carrier escorts, their nuclear power plants enabling them to keep up with the carriers themselves.

The most significant of these small-carrier experiments was the Sea Control Ship, conceived as a conceptual evolution of the ‘low end’ escort carriers of the Second World War, and as a means of bringing the number of flight decks back up after the mass retirement of the Essex class. The SCS—nobody dared call it a carrier, much as HMS Invincible was officially a cruiser— would in wartime have been a convoy escort and submarine killer, devoted to roles for which a supercarrier would have been overkill, or too rare; in other words, jobs that the aging, too-small Essexes would once have done. In peacetime, the SCS would have been used for showing the flag and gunboat diplomacy.


A 1974 conceptual drawing for the Sea Control Ship

The SCS design described a ship a tenth the size and an eighth the cost of a Nimitz-class carrier, or roughly the size of a contemporary guided-missile cruiser, and carrying an air wing of antisubmarine helicopters and a handful of V/STOL fighters. An alternative, smaller design would have been built on a converted Spruance-class destroyer hull. The USS Guam, a helicopter-carrying amphibious assault ship, was used for extensive trials for the concept on a 1974 Atlantic deployment, supporting a detachment of three first-generation Marine Corps Harriers and seventeen antisubmarine helicopters. The USS Franklin D. Roosevelt, a 1940s-vintage Midway-class carrier too small and too old for front-line use, carried Harriers and antisubmarine helicopters in a 1976-1977 test program intended to evaluate how V/STOL aircraft could be integrated into normal carrier operations.

The SCS idea as a whole showed promise, but eventually foundered on the rocks of Congressional politics in 1977, after Admiral Zumwalt’s retirement took the program’s strongest supporter off the chessboard. The Spanish Navy bought the design and built a 16,000-ton ship to a modified plan, the Principe de Asturias, which has done good service since 1988.

Save that Hendrix and Williams’ proposed smaller carriers are bigger and include humanitarian missions alongside low-intensity operations and showing the flag, their proposal sounds like a larger regurgitation of the Sea Control Ship.


A comparison of size, with the Spanish Navy's Principe de Asturias (17,188 tons) in foreground, USS Wasp (40,500 tons) second, USS Forrestal (81,000 tons), and HMS Invincible (25,000 tons)

Almost forty years after the SCS concept sank, the world is in a very different state of affairs. The Cold War ended twenty years ago and the Soviet Union no longer exists. There are no prospects for a high-intensity conflict on that scale in the foreseeable future. Most of the Navy’s combat or combat support operations involve low intensity conflicts, sometimes even very-low intensity conflicts such as the Somali pirates. These type of operations usually just don’t need an aircraft carrier. If they are available, the Navy also uses amphibious assault ships, capable of operating vertical-takeoff Harrier fighters and helicopters, as surrogate aircraft carriers for crisis areas such as the Libyan intervention, or for humanitarian operations.

Though there isn’t any occasion these days to use them to their full capabilities, the US Navy sticks to its fleet of eleven supercarriers largely because they’re what it knows best, and also because there is currently no reason good enough to justify departing from them. Supercarriers are by this point a proven thing, and despite all the drawbacks of size, cost, and (arguably) inflexibility, they represent a tremendous amount of force that can be moved around the world with considerable speed and almost total impunity. Two or three supercarriers and their escorts can outgun most of the world’s smaller countries, and are capable of shutting down a small country within days, such as the NATO air power did in Serbia in 1999. That’s not a capability to be given up or whittled down without good reason.

The argument about the development of anti-carrier weapons such as the Chinese military’s new DF-21D antiship ballistic missile certainly holds water. It would be foolish to expect possibly hostile nations not to put considerable effort into ways to destroy aircraft carriers. One or two hits from a missile that size could disable or sink a supercarrier, but as with any attack, the problem is getting the hit. Supercarriers are not passive targets, and they do not operate alone. In the long run, though, it’s just another step in the offense-versus-defense development process. The USSR sought to counter aircraft carriers with antiship missiles and the US Navy responded with better missile defenses, including the Aegis system. The same process is going on now—many of the US Navy’s surface ships have been or will be fitted with the antiballistic-capable SM-3 missile, originally as part of a national scheme of ballistic missile defense, but easily adaptable to carrier defense. More broadly, ballistic missile launches are by their nature impossible to hide, and the US has plenty of spy satellites and other detection systems. Any carrier threatened by a ballistic missile attack would have advanced from the time of the attacking missile’s launch.

The supercarrier has reigned supreme for so long that several other countries are bending great efforts towards obtaining them. The UK has two supercarriers HMS Queen Elizabeth and HMS Prince of Wales, under construction, nearly thirty years after the Ark Royal was retired. France is considering a second carrier, an improvement over the underpowered Charles DeGaulle, and both India and China are exploring the possibilities.

Now granted, there have been technological changes over the last thirty years that would make a smaller aircraft carrier more feasible. For one thing, there are now more ways to exercise long-distance power than to use conventional fixed-wing aircraft. The initial salvos of cruise missiles from the submarine USS Florida in March 2011, as part of the US Navy’s involvement in the NATO intervention in Libya, devastated the Gaddafi regime’s military and essentially shut down the country’s air defense network within a couple of hours. Twenty years ago, no ship besides an aircraft carrier could have done the job, but a repurposed ballistic missile submarine crammed to the brim with Tomahawk missiles accomplished it in a matter of hours. The short-lived idea of the ‘arsenal ship’ discussed in the years after the first war with Iraq was an even simpler concept—a large ship, similar to a freighter, carrying hundreds of cruise missiles. This concept was proposed in 1994, but funding has never been appropriated for a full study or design.

Vertical or short take-off and landing (V/STOL) aircraft have also matured as a technology, and second, the great and unexpected strides that unmanned aerial vehicles (UAVs) have made in the last ten years. Neither of these technologies existed as anything other than crude prototypes when the first Nimitz-class carrier was launched in the late 1960s. The most significant weakness to the SCS concept was the lack of a good V/STOL fighter, the early model Harriers not being designed as fighters. The Rockwell XFV-12, a purportedly supersonic V/STOL fighter designed as a complement to the SCS, proved wholly impracticable and was abandoned after millions of dollars had been spent to produce an aircraft that simply could not fly.

The Hawker-Siddley Harrier was an oddball little plane introduced in the early 1970s, initially as a land-based attack plane used by the Royal Air Force and the US Marine Corps, but it only made a big impact ten years later, when the Royal Navy’s Sea Harriers acquitted themselves well against the Argentine military’s land-based Mirage and Skyhawk aircraft during the 1982 Falklands War. Suddenly every navy that operated a carrier had a passable fighter and attack plane that could operate from a small deck, and many countries such as Spain and Italy were launching new small carriers designed with the Harrier in mind. Sure, the Harrier wasn’t nearly as fast or long-ranged or high-performance as a MiG-29 or an F-14, but it was better than most of the older-generation aircraft operated by, say, Libya or Pakistan.

Nearly thirty years after the Falklands War, however, the Harrier remains the only combat-proven V/STOL aircraft in the world. The Harrier’s Soviet counterpart, the Yak-38, never saw combat and was never produced in large numbers. The US Navy and Marine Corps (and many other nations) intend to purchase the V/STOL version (the F-35B) of the Lockheed-Martin F-35 Lightning II (formerly known as the Joint Strike Fighter) when it becomes available.


At $150 million each, the F-35B promises to be a great improvement over the Harrier, with performance comparable to a conventional modern jet fighter, better range, weapons payload and speed, and stealth characteristics. The F-35’s development has been a long, painfully slow, and heavily politicized one. The F-35 has been in development since 1996, first flew in 2006, and fewer than twenty aircraft have been produced to date. Over the objections of the Department of Defense, Congressional Republicans led by members from Ohio and Indiana have repeatedly tried to earmark funds (up to $450 million) for manufacturing a second engine, to be produced by General Electric and Rolls Royce in those states, in addition to the approved Pratt & Whitney engine. The F-35 is also the most expensive defense project in the United States’ history, with over a trillion dollars spent or projected to be spent on an aircraft that still hasn’t completed all of its tests. Each aircraft also costs three times as much as a late-model Harrier ($35 million in 1997 dollars during USMC procurement, or about $48 million in 2011 dollars). As costs skyrocketed, the number of planes proposed for purchase has been steadily reduced (the US Marine Corps still wants more than 400 of them). If the Pentagon’s other procurement programs are anything to measure by, the F-35 is highly unlikely to meet it’s planned in-service schedule of 2016-2018. Secretary of Defense Gates has threatened to cancel the F-35 program unless it starts meeting its goals without continued financial drain.

Without aircraft, a carrier is useless, and the performance of the available aircraft dictates the carrier’s abilities. For example, the Navy’s carrier force was significantly weakened when the A-6 Intruder attack aircraft were retired due to age in the late 1990s. Although they were replaced with an equal number of F/A-18 Hornet fighter-bombers, the Hornet has half the range and two-thirds the weapons payload of the Intruder. This not only limits the carrier’s offensive power (in terms of number of missiles or bombs that can be launched at a target) but both forces the carrier to operate closer to shore, more likely within range of enemy attack. Aircraft range also limits how far inland the carrier can reach-- from a carrier in the Persian Gulf, an Intruder could hit Baghdad, but a Hornet can’t. The Intruder’s replacement, a carrier-based stealth bomber named the A-12 Avenger II, was a financial boondoggle and engineering failure that died in development.

On the other hand, improvements in guided weapons have made it possible to effect damage on the enemy using far fewer aircraft sorties than were previously necessary.

Unmanned aerial vehicles are the newest addition to military aviation; the RQ-4 Global Hawk, MQ-1 Predator, and MQ-9 Reaper drones have been one of the most conspicuous success stories in military technology of the last twenty years, springing off the drawing boards to become a mature and indispensable tool within only a few years. After 2004, they became one of the most useful weapons the US military had. UAVs were used in 118 air attacks in Pakistan alone during 2010, in addition to doing yeoman service in Iraq, Afghanistan, Yemen, Somalia, and now Libya. Successive generations of UAVs have become longer-ranged and more useful (or in the case of the Marines, small enough to be flung into the air by hand), with better sensors, loiter times, and weapons payloads; the General Atomics-manufactured MQ-1 Reaper is a major improvement on the Predator, carrying fifteen times as much ordnance as the Predator, while the Navy’s Global Hawk has a range of 15,000-miles and an endurance of 36 hours, enough to get from California to Hawaii and back. Compared to a manned aircraft, UAVs are slow and limited, but they are inexpensive ($10.5 million for a Reaper), capable of loitering for long periods of time, hard to see, avoid, and kill, and if fitted with missiles, can more or less immediately attack any target they can see. If more firepower is needed, they can be used to bring other aircraft or artillery fire onto the target. Many of the UAV’s jobs until recently required manned aircraft patrolling the area, with increased costs and risks, and which required a land base or an aircraft carrier. UAVs, by contrast, can be launched from the back of a truck or from the helicopter deck of a frigate, and could as easily be launched from a smaller carrier as from a large one. Helicopter-type UAVs are also joining the arsenal, in addition to the fixed-wing Global Hawk and Reaper. A small carrier remaining on-station in the Gulf of Aden with a hangar full of UAVs could maintain surveillance over an area encompassing Yemen, the Red Sea, and Somalia.

Captain Hendrix proposes that three smaller (40,000-ton) aircraft carriers could be built for the cost of a single Nimitz or Ford-type supercarrier. This assertion is best taken with a grain of salt. Given the US Navy’s ship design and construction contractors’ apparent inability to get any project completed without years of delays and hundreds of millions of dollars in cost overruns, (e.g. the USS San Antonio, completed three years late, $840 million over budget, and still grossly unfit for service) the argument that any smaller ship is going to be cheaper is on very shaky ground to begin with. As a more direct analogy, the Littoral Combat Ship was originally supposed to cost $200 million, 1/5 as much as a Flight IIa Arleigh Burke class destroyer. As of early 2011, however, the going rate for a Littoral Combat Ship is $450 million, or nearly half as much as a Burke for a ship that was supposed to be smaller, simpler, and less expensive. So much for the ‘low’ end of the balance.

Setting aside the cost-effectiveness proposition, however, there are numerous advantages to having a greater number of smaller aircraft carriers available:

  • Availability is, in fact, the key. The US Navy has shrunk from a circa-1990 peak of nearly 600 warships, including fifteen carrier groups and four groups built around the reactivated Iowa-class battleships, to a total of 286 ships in 2011, including eleven carriers. Granted, the 1990 peak reflected the Reagan administration’s arbitrary and fiscally profligate ‘600-ship navy’ goal, as opposed to actual need, and included a large number of warships retained past their truly useful lifespans, at considerable cost in manpower and maintenance. Granted also, such a large fleet simply wasn’t necessary after the Soviet Union collapsed. Still, the US Navy has commitments spanning the globe, and a much smaller fleet with which to meet these commitments. A Burke-class destroyer is considerably more powerful than one of the 1980s-era Spruance class, with a far greater radius of effect thanks to her cruise missiles, but it still can only be in one place at one time. The same is true of supercarriers, only more so. The carrier off the coast of Libya cannot be of use in the Persian Gulf.

  • If the resources equivalent to a supercarrier are spent on three smaller ships, three crisis areas can be covered instead of one. This metaphorically allows us to spend small change on a cup of coffee, rather than having to hand over a $50 bill.

  • Given that of the eleven operational supercarriers, two or three are generally out of service at any given time for maintenance or other work, a fleet of smaller, more numerous carriers would in theory have more ships available for service at a given time.

  • Smaller carriers could be maintained using smaller shipyards, without need for facilities that can accommodate supercarriers. This point may be moot, however, given the considerable infrastructure the US Navy has built up for supercarrier maintenance over the last half-century.

  • Smaller, ‘routine’ missions (though no mission is truly routine when human life is at risk) are more likely to occur than the need for a major air assault. For the last ten years of the ‘war on terror,’ with the exception of the invasion of Iraq, the usual role of naval airpower has been to conduct small attack missions on specific targets, rather than the sort of blitz that savaged Iraq in 1991 or that turned out the lights across Serbia in 1999. Given appropriate aircraft, smaller aircraft could handle routine missions more cheaply and with less need to weigh carefully the small number of supercarriers against the number of missions.

  • Many missions do require air support, but aren’t big enough to justify a supercarrier—for example, a limited intervention such as in Libya.

  • Assuming three small carriers collectively possessed the same combat power of a single supercarrier, the sinking of one ship would leave the other two operational.

  • Improvements in communications and networking among ships make it possible for ships to operate more closely.

  • Smaller ships would theoretically be quicker to construct and bring into service, though in view of the current state of naval procurement this is highly unlikely.

  • Supercarriers very rarely operate at maximum capacity; in fact, the most recent full-capacity combat operations were during the invasion of Iraq in 2003, and smaller ships could handle the quotidian operations at less cost.

  • The logistics of supporting smaller carriers would be simpler; current supercarriers are limited to certain bases and harbors by their size and draft, and cannot pass through the Panama Canal. Dredging of the Suez Canal since 1975 has made it possible for supercarriers to pass through the canal rather than sailing around the Cape of Good Hope.

There are also several important arguments against smaller aircraft carriers.

· The small carrier’s actual usefulness would depend on suitable aircraft being available, and in this case it means the small carrier is married to the F-35B. Granted, the Harrier is available now, but the F-35 is unlikely to be available in suitable numbers for as long as ten years, and may be cancelled outright like the A-12 Avenger II was. It is generally a very bad idea to develop one’s ships or operational doctrine around an airplane, vehicle, or other weapon that hasn’t entered service yet—for example, consider the jolt to the Marine Corps when the Expeditionary Fighting Vehicle was cancelled, essentially junking over a decade of research, testing, and planning. If a new small carrier is forced to rely on Harriers, it will lack the punch that Hendrix expects. For that matter, even Hendrix’ example of the USS America is still under construction, due to commission in 2012 but probably several years from entering service.

· For Libya-sized operations, or issues such as the Somali pirates, an available amphibious assault ship could be used as a surrogate carrier (as in the authors’ own example), filling the small carrier’s proposed niche with an existing ship and rendering the need moot.

· Operating costs can outweigh the construction cost of a ship in cost-effectiveness evaluations. An amphibious ship of the Tarawa or Wasp classes—the America’s predecessors—is second only to a supercarrier in manning, running, and maintenance costs.

· The 10,000-ton multirole ships they envision instead of the Littoral Combat Ship—the size of an Arleigh Burke class destroyer-- would likely be too small to be of much use, facing serious tensions between cargo and troop space, power plant, offensive and defense weapons, and other design requirements. In other words, the same tensions that resulted in the LCS being such a bastardized compromise of design. Even the South Korean navy’s small Dokdo-class amphibious assault ship is 50% larger than Hendrix and Williams’ proposal.

· In the event of a major regional conflict, such as a hypothetical war involving North Korea or Iran, a smaller carrier with thirty aircraft would be seriously out of its depth. Granted, an F-35 would be more than a match for most of the Soviet-era relics operated by North Korea, but maintaining high operation tempos (keeping many aircraft in the air and maintaining that status for a long period of time), requires lots of aircraft, lots of man-hours, and lots of deck and hangar space, since for every aircraft airborne there will usually be one or two on deck rearming and refueling.

· Any threat formidable enough to ‘outrange’ and threaten a supercarrier, as Hendrix and Williams discussed, would by definition also outrange and threaten a smaller carrier. The only advantage to a smaller carrier would be that a cheaper ship with fewer crew would be at risk.

· Given current budgeting and manpower limitations, any construction of smaller carriers would almost certainly have to come at the expense of something else, such as the retirement of a supercarrier. The USS Enterprise, the first nuclear-powered supercarrier, is scheduled to decommission within the next several years, but as two more supercarriers (USS Gerald R. Ford and USS John F. Kennedy) already in the construction pipeline, any such trade-off is unlikely to happen anytime soon.

· If the number of aircraft carriers is significantly increased, there may not be enough escort ships to go round. Aircraft carriers have a limited self-defense capacity, and defend on their aircraft and other ships to protect them. At present, a Carrier Strike Group theoretically consists of a carrier, one or two Aegis-equipped cruisers, two destroyers, and two frigates. In practice, however, escort ships are frequently detached for other missions, simply because there aren’t enough ships to go round. The US Navy’s fleet of surface escorts currently consists of 22 Ticonderoga-class cruisers (the five oldest were retired circa 2003, as they had Mark 26 twin-arm missile launchers instead of the Vertical Launch System) and 60 Arleigh Burke-class destroyers of various models. There are also 28 Oliver Hazard Perry-class frigates, which are essentially “worn-out and maxed-out” in the words of Vice Admiral Barry McCullough, and which are scheduled for retirement. From the Navy’s perspective, this is few enough surface ships as it is (110 all told) for the missions already on the table, and one of the reasons of the hue and cry over the Littoral Combat Ship is that the Navy needs them to free up bigger ships for other roles.

· Lastly, the design and construction of smaller carriers would unfortunately be subject to the same delays and overruns as any other defense (particularly warship construction) project. A ship initially proposed as a ‘small and cheap’ 40,000-ton carrier would be very unlikely to remain either small or cheap, and if the past is any guide (e.g. the Littoral Combat Ship, the prototypes of which cost 3-4 times the original budget) it would soon balloon to the point where it would have all the drawbacks of a supercarrier but none of the advantages. In that situation, the Navy might as well follow its historic practice and spend the money on a supercarrier. If that happens, the ship could also face the same budget axe that killed the Air Force’s F-22, the Marine Corps’ Expeditionary Fighting Vehicle, and a number of Army projects over the last few years. The ship could also turn out to be as much an unfit-for-deployment white elephant as the LCS USS Independence, whose aluminum hull is disintegrating, or the USS San Antonio, who almost shut down the Suez Canal in 2009 with a narrowly-averted collision and grounding caused by the failure of her steering equipment.

It is a sad thing indeed that one of the most damning arguments against an idea like this is that the defense industry can’t seem to produce anything that is small, inexpensive, and easy to implement, and can’t produce anything on-time, on-spec, and on-budget. Secretary of the Navy Winter and Secretary of Defense Gates have had essentially to go to war with their own contractors in order to try to get the system on track, including canceling the entire Littoral Combat Ship program in 2007.

In conclusion, Hendrix and Williams’ proposition as a whole has some merit—supercarriers do represent an immense cost and potential loss packed into a single target. Smaller carriers do have many arguments in their favor, as well as many arguments against them. The Navy currently faces the dilemma of not having enough ships to go round, a stretched-thin budget, and a procurement and shipbuilding system that is grossly incapable of providing the ships the Navy needs.

In the current climate, with a stretched-thin Navy and an even more stretched budget, however, the issue of introducing an entire new type of warship is a hard one to greet with open arms.