Michael KreponNaval Treaties

Before Ronald Reagan, there was Charles Evans Hughes. Both men were unorthodox thinkers, spotlight stealers and risk takers. Both were dismissive of the status quo, and sought deep cuts in strategic forces.

Battleships were the strategic forces of the 1920s and 1930s, able to traverse long distances to project power and assert national interests. The big guns of capital ships made their presence felt far from home to influence decisions about war and peace.

Hughes, Warren G. Harding’s Secretary of State, knew that his country was tired of war and international intrigue. The isolationist mood among US citizens and within the Congress would not support a vigorous naval buildup. Hughes wanted to establish a system of naval arms limitation for Great Britain, the United States and Japan that kept a lid on adventurism while bowing to domestic political and budgetary constraints.

At the opening of the Washington naval conference in 1921, Hughes electrified the world by proposing deep reductions in battleships on line and under construction. My data on Hughes’ proposals and what was ultimately agreed upon comes from Volume III of Analysis of Selected Arms Control and Disarmament Agreements Between the World Wars, 1919 – 1939 by Richard Dean Burns and Donald Urquidi (1968). Hughes volunteered that the United States would scrap plans for 15 ships under construction and 15 older vessels, totaling 846,000 tons – if Great Britain would scrap four ships under construction and 19 older vessels totaling 583,000 tons and if Japan would scrap eight projected battleships, seven under construction, and ten older vessels totaling 449,000 tons.

Hughes seized the conference agenda and generated international enthusiasm for his proposals. What emerged from the 1922 Washington Naval Treaty was an agreement by Washington to forego 771,000 tons of capital ships, compared to 591,000 for Great Britain and 415,000 for Japan. These reductions and the accompanying 5-5-3 ratio served US interests in the near term, but only also re-channeled the competition in naval ship construction to heavy cruisers and submarines that remained unconstrained. The ten year “naval holiday” on new battleship construction enshrined in the treaty was therefore a fiction. Along with these ratios came a pledge by the signatories not to create or enlarge fortifications and new naval facilities in the Pacific.

Diplomatic historian Arthur E. Tiedemann concluded that,

Without a doubt the Washington Naval Treaty assured the home territories of Japan and the United States. A 40-percent inferiority in capital ships and carriers rendered it unlikely that the Japanese could successfully invade the eastern Pacific. On the other hand, [the no fortification pledge] made the United States operations in the western Pacific extremely difficult… therefore the treaty gave the Japanese an unchallengeable control in East Asian waters.

The impetus for naval arms control had not yet run its course. The US delegation to the 1930 London naval conference was led by Secretary of State Henry L. Stimson, who, like Hughes, was operating under severe constraints of US isolationism and domestic opposition to funding naval ship construction. The London Naval Treaty extended and expanded limitations on some surface combatants, while exceptions to these rules and escalator clauses flourished. Japan was headed on a collision course with the United States, and in December, 1934, Tokyo announced its intention to withdraw from all naval limitations, providing the two-year notice required by the treaty. The United States still reacted sluggishly to Japan’s announcement.

The bill of particulars against nuclear arms control draws heavily from this negotiating record of the inter-war years. In this critique, the treaties were to blame for US unpreparedness prior to World War II. Stimson, a Republican who responded to President Roosevelt’s call to join his cabinet as Secretary of War to help build bipartisan support as war clouds loomed, didn’t believe in the lulling influence of arms control. As Stimson recalled in his autobiography, On Active Service in Peace and War (1948), “It was not the [1930 London] treaty, but Congress and the President, supported by the public, that prevented the construction of fighting ships in the years that followed.” In 1936, the year of the London Naval Treaty’s expiration, US plans for naval ship construction did not even aim to achieve permitted treaty strength until 1942.


  1. Scott Monje (History)

    Also, I think the subject matter of arms control back then was shaped by the same sort of verification issues that later influenced nuclear negotiations. The focus was on capital ships not just because of their range, but because it took two years to build one and they were built in the open. That allowed for the sort of transparency that made it hard to cheat and get away with it. It was the “national technical means” of detection of that time. On the other hand, interwar proposals to limit ground forces never had comparable success because cheating would have been harder to detect in time.

    • John Schilling (History)

      That’s a very good point, and it isn’t limited to the naval/ground forces dichotomy. The true capital ship of WWII, particularly in the open seas of the Pacific, turned out to be the aircraft carrier – and aircraft carriers do not have to be built openly. Japan, if I recall correctly, built half a dozen or so as passenger liners and/or naval auxiliaries, specifically designed to have the nominal payload swapped for a flight deck and hangars, etc, on short notice. The United states turned about a dozen light cruiser hulls, almost beneath treaty notice, into fast light carriers during the war, along with countless merchant-hulled escort carriers.

      By comparison, there was no such thing as a dual-use 16″ gun, nor much in the way of civilian applications for foot-thick armor plate. The problem posed by dual-use technologies to the effort to combat nuclear and ballistic-missile proliferation are well known, and it’s not hard to see other examples as we look forward to e.g. missile defense and space weapons.

      When crafting arms-control treaties, you need to look out for the game-changing military technologies that might not seem worth controlling, and for the dual-use almost-weapons that treaty-constrained powers can stockpile in anticipation of breakout. These are things that can put an otherwise-useful treaty beyond reach or, as noted earlier, make it worse than useless.

    • Spruce (History)

      With respect to carriers, one should be careful with hindsight. While the carrier turned out to be the main weapon of WWII, none of the navies went into the war thinking that way. See, for example, the low priority Germans placed on Graf Spee, the British reluctance to reform FAA, the US pre-war strategies (Plan Orange being the prime example), and Japanese deployment of their battlefleet and their insistence of seeking decisive battle betwenn opposing battlelines. There was really no arms race between countries in carriers pre-war because of this as they were still considered just supporting ships for the battleships.

      Also, as noted, there were tonnage limits on carriers in the naval treaties. These actually caused almost directly to the loss of USS Wasp, which was designed as a 15,000 ton version of the 20,000 ton Yorktown class – because that 15,000 tons were what was left of the US carrier tonnage. One of the weight savings was practically complete elimination of torpedo protection, which ended up costing the ship when torpedoed by a submarine.

    • Cameron (History)

      A second note on hindsight: The major pacific naval battles included Battleships, and their use was expected. The IJN sent 2 battleships to the pearl harbor attack, they sent 3 battleships and 4 battlecruisers to midway (and another 4 battleships to the aleutians)

      WW2 became an air war despite the intentions of the admirality on both sides, and in part because it’s what the US had available after Pearl Harbor and CV’s are quicker to build than BB’s.

  2. Seb (History)

    That said… remember the (entirely daft) hysteria in the Britain over German “secret Battleships” in the 19/20th century Anglo-German naval race that (amongst many other things) helped set Britain and German mistrust that contributed to WWI. Evidently not *that* long ago people were not convinced about the verrifiability of even stonking great battleships.

    I had always thought that the naval Washington and subsequent naval treaties were in part motivated by a desire to avoid that kind of drift towards conflict through competetive and destabalising dash for strategic advantage.

    • Markus (History)

      Stonking great battleships, and their capabilities, were in fact much easier to hide than might be expected nowadays. The Japanese very effectively concealed the Yamato during its design and construction. US Navy intelligence on the characteristics of various elements of the Japanese fleet (Battleships , Heavy Cruisers), and even the British Fleet was surprisingly weak.

      Apart from having a very poor understanding of the design and capabilities of the Yamato class the US Navy greatly underestimated the speed of the entire Japanese battle-line.

      The Naval Treaties did a very nice job of eliminating the possibility of a Jutland style battle by the simple expedient of eliminating the large forces of battleships that were needed to meet all of the various fleet needs in addition to maintaining a battle-line.

    • Seb (History)

      Granted, their capabilities I would expect to be quite hard to get! Which of course, is really a lot more important and relevant in terms of who will win against whom. The RN may have had a large fleet going into WWI and WWII, but if you look at the engagements before Jutland, and also some of the battles in WWII, tonnage clearly was not the entire story: radar fire control, air defences etc. mattered a lot more.

      I wouldn’t have thought (though, again, this is not something I’ve ever looked into in depth, so correct me if I am wrong) that there were people worrying excessively about verification beyond avoiding a costly arms race in tonnage, rather than trying to work out the actual military effectiveness of the other peoples fleet. As you say, that would be much harder and rather than being seen as a deal breaker seems to have been largely ignored in the interests of achieving the wider aim of trying to prevent an arms race of the Anglo-German style that was both expensive (Churchill I think it was said something like “the treasury want four, the admiralty want six, and strangely we have compromised on eight.), and helped create the tensions and distrust that led to the war. I think, IIRC, it was the naval race that led the UK to join the Entente.

      In the context of nuclear weapons, I wonder about verification. If the Russians want to keep tens of thousands of nuclear weapons, do we actually care if this doesn’t confer on them an ability to counterforce our deterrent? Sure, they make the rubble bounce, but then nobody is left to see that, so it really doesn’t make a difference to us.

      As a political issue though, that might make us very nervous about what their intent was in keeping all these nukes, and we would probably do the same to avoid looking weak costing us money, and making both countries distrust each other… but then if they keep secret nukes, do we care? As long as it is’t forcing us to do something stupid and costly. I’m thinking of all that rubbish about strategic bombers for the purposes of nuclear accounting and certain people getting worked up over the possibility for Russians counting multiple air dropped bombs as only one bomb.

      If you can rely on a sub launched force being robust against a first strike, and you can meet deterrence thresholds with 300 deployed nuclear weapons on subs… how much should we let the robustness of verification get in the way of conducting an agreement to meet the strategic goals of being able to have the smallest diversion of resources to achieving deterrence and that the issue itself doesn’t actually set us on a collision course through distrust?

      On the other hand I suppose verified agreements really is the only direct lever you have in making sure Russia is not doing things that put us at risk through adopting bad practice in an attempt to maintain parity. Surely we care more about whether they are doing dangerous and stupid things with their arsenal.

      Now, not as a substantive point but as a personal curiosity, but their existence (i.e. numbers, tonnage and gross features like the number of guns and their calibre) strikes me as being relatively easy to gauge in the 19th century… there are only so many ports and dry docks, and it was pretty easy to move across borders those days.

      I would have thought it was pretty much impossible for Germany to have reasonably constructed and held in secret X many extra battleships that the British Admirality simply wouldn’t even know about (which I had thought, though it was a long time ago that I studied all of this, was the issue), the idea always struck me as bafflingly odd.

    • Scott Monje (History)

      “If the Russians want to keep tens of thousands of nuclear weapons, do we actually care if this doesn’t confer on them an ability to counterforce our deterrent?”

      My problem with finely calibrated rational arguments about deterrence is that most people just don’t think that way, and that has consequences. Look at the current GOP debate on foreign policy (to the extent that there is one). You don’t hear anything about the size or intended purpose of the Russian nuclear force, the imbalance of tactical nuclear weapons, or any other related issue. Whether or not you consider these things threatening, they are at least real and could be the subject of meaningful disagreements. Instead, you hear a discussion of whether/when we should go to war with Iran because they don’t have a nuclear weapon but they might want one and we all know they’re trying to build one anyway even though no one can prove it and whatever it is that happens when they get one it’ll be worse than anything we can imagine. Just how do we calibrate a calculation like that?

      A minor quibble on joining the Entente. It wasn’t really anything that a country could join. Britain negotiated separate ententes (“understandings”) with France and Russia to resolve long-standing colonial disputes, in Africa in the case of France and in Central Asia in the case of Russia. There was no comparable negotiation with Germany at the time because they had no comparable dispute. The later dynamics of growing tensions in Europe pushed the British closer to France and Russia in opposition to Germany, but there was no formal defensive treaty tying the British to France and Russia (although the latter were allied to each other). Thus the Germans hoped that Britain would remain on the sidelines when World War I actually broke out.

  3. Walt Slocombe (History)

    On balance, the naval treaties get a bad rap. Until the early to mid 30s — when there had been profound (and adverse) changes in the governments of both Japan and Germany (the latter not a party to the treaties, but limited much more stringently by the Versailles Treaty), the main naval nations did adjust their building to meet treaty limits. Yes there was – on virtually all parts — some creative accounting on tonnages, but basically every player kept more or less to the rules, often with some funny effects on ship design. The treaties did achieve their actual political goal — to prevent a useless, expensive, and conflict-inducing rivalry among the WWI allies — US, UK, Japan, Italy, and France. For more than a decade, naval buildups were sharply limited, and there were no significant military-related tensions among the old allies. By the time the military had come to dominate Japanese politics, and Hitler had come to power in Germany, the political premise of the agreements had evaporated, and with it the readiness to stick to the limits very different governments had agreed to. As your post indicates, the failure of the democracies to react to the challenge — by no means limited to naval buildups — was due to blindness and sloth, not the naval treaties.
    On a couple more technical points, there were limits on aircraft carriers, including on total tonnage, and on cruisers, which did shape and limit construction.

    After the sinking at Midway of 4 of the 6 carriers that had struck Pearl Harbor and the decimation of Coral Sea of the air crews of the other two, Japan did convert a battleship (Shinano) under construction to be a fleet carrier, but so far as standard references books indicate, the idea that Japan during the treaty period built nominally civilian ships with a view to conversion to aircraft carriers is inaccurate. Japanese did convert three naval auxiliaries to be light carriers — Ryuho, Zuhio, and Shoho, and converted two incomplete liners to be larger carriers — Hiyo and Junyo — and, like the UK and US, converted several merchant freighter for use as escort carriers — but the hulls used for that purpose had all been laid down in the late 30s, after Japan had withdrawn from the treaty system.)

    • John Schilling (History)

      Chitose and Chiyoda were also built as auxiliaries (seaplane tenders) with provisions for rapid aircraft-carrier conversion, though for those two the conversions weren’t implemented until after the start of the war.

      And all but Hiyo and Junyo were laid down in 1935 or earlier, when Japan was still party to the Washington and London treaties. The American cruiser- and merchant-hulled light carriers were post-treaty developments, but it is hard to see the Japanese ships as anything other than deliberate stockpiling of dual-use hardware in anticipation of breakout.

  4. walt Slocombe (History)

    Thanks for the clarification. Janes says that the first of the Chitose group was laid down in 1934, so it is fair to say that the Japanese admiralty probably had the possibility of conversion in mind when they were designed. Such a possibility may well have been part of the design of other hulls laid down in the mid-30s, and I have no doubt was a factor in those laid down later, and quite possibly designed before the formal denunciation.
    However, by 1934, the treaty system was essentially dead, and Japan formally announced its withdrawal from the treaties at the end of that year. My point was that there is no evidence that Japan, in entering into the treaties, and, indeed more or less abiding by them for more than a decade (and no nation was totally clean on how tonnages were calculated), intended to violate their spirit by stockpiling hulls for eventual conversion to full flight deck carriers, or , for that matter in any other way. (It’s true that post war analysis showed that many of their “treaty’ cruisers exceeded the 10,000 ton limit if you used some methods of estimating displacement, but I believe that was not unique to Japan.) As I understand it, the actual conversions were not begun until significantly later, in some cases as late as 1941, by which time conditions had changed totally. But I accept your point that by the mid-30s, Japan likely designed some “merchant” and “auxiliary” hulls with a view to possible conversion.
    And I agree, of course, that the US and UK were foolish to continue to twist their designs to meet treaty standards once Japan opted out, but that was only a relatively small part of the much greater folly of refusing to recognize what was going on. The whole story seems to me simply to illustrate that arms control (and other) treaties affect nation-state action only so long as they are reasonably in conformity with the states’ broader security strategies.