Sonar kills whales, and Yglesias links to an interesting Hendrik Hertzberg article making the following proposal:
Any real solution would have to be international, involving some serious diplomacy aimed at a new kind of arms control. The cat-and-mouse games played by submarines may have made a certain kind of sense during the Cold War, but what is the point of them now? Russia is annoying, but it is no longer our mortal enemy. The Chinese have no desire to bury us, except under piles of sneakers and kitchen appliances. Al Qaeda’s navy, as the Cole incident showed, consists of the odd motorboat. The terrorists have no submarines. Perhaps the immense sums still spent on naval toys whose strategic rationale is obsolete could be redirected to teaching the relevant personnel, humans and dolphins alike, to speak Arabic.
I’m skeptical. To overly simplify matters, there are really two kinds of arms control agreements, and neither of them fit submarines very well. The first kind, into which the Washington Naval Treaty or the SALT agreements might fit, try to limit the numbers and types of roughly symmetrical weapons held by countries with roughly symmetrical capabilities. For example (and to simplify even further) the Japanese, British, and Americans recognized in the 1920s that it was okay to limit oneself to 15 battleships apiece as long as the other major powers did the same. It didn’t do anyone any good to have an unconstrained arms race that might invite pre-emption and increase tension, but that in the end would probably result in roughly the same balance of power. Similarly, strategic arms limitation between the US and USSR was based on the idea that both states could make do with just thousands of nuclear warheads, rather than many multiples of that number. The ABM Treaty was based on more or less the same logic.
Unfortunately, this logic does not apply to submarines. Despite the fact that the United States has the largest fleet of advanced nuclear submarines in the world, submarines remain, fundamentally, weapons of the weak. Submarines are relatively cheap, difficult to detect, and can kill much more expensive capital ships. Accordingly, states that don’t feel they have a chance in a normal naval encounter tend to build lots of submarines. Kaiserine Germany resorted to a submarine campaign after Jutland demonstrated the futility of trying to destroy the Grand Fleet. The German surface fleet in World War II was hopelessly outmatched by the Royal Navy, and thus resorted again to submarine warfare. After the war, the recognition on the part of the Soviets of overwhelming and long term Western naval superiority led down a similar path. Unsurprisingly, the PLAN (deficient in surface ships) has a strong submarine arm, the Russians have preserved their submarine fleet better than their surface, and the Iranians and the Venezuelans have both concentrated big ticket procurement on submarines rather than surface vessels. The USN launched a very successful submarine campaign against the Japanese in World War II (as part of a larger Mahanian campaign), but the US submarine fleet was primarily oriented towards the defense against and destruction of the Soviet submarine fleet during the Cold War.
What this means is that, unlike the situation with battleships and nuclear warheads, an effort to ban submarines would involve a significant redistribution of world naval power. If the Chinese and Russians can’t have submarines, then they might as well pack up and go home; the USN will go where it wants and do what it wants to whom it wants without difficulty. Now, as Hertzberg and Yglesias suggest, maybe major power war is obsolete and pointless, and anyway submarines aren’t terribly effective against terrorist. That may be true, but the problem is that you would have to get the major powers to agree for the very long term that they have no disputes susceptible to militarization, and thus no reason to worry that overwhelming US naval superiority will be used against them. In other words, in order to ban submarines you essentially have to get the big states to agree that they’ll be peaceful forever. If we could manage that, great, but while we’re at it, we should probably ban all other kinds of weapons as well.
The second kind of arms control agreement involves the declaration of some particular weapon type as immoral, and thus worthy of being banned. There’s some overlap with the first, certainly; the Washington Naval Treaty wouldn’t have come about without the slaughter of World War I, and the nuclear limitation agreements (although not necessarily the ABM Treaty) were premised on the idea that nuclear weapons are icky. Bans on chemical and biological weapons and on land mines would fall into this category. There was certainly some thought given in the interwar period to banning submarines; the British considered their use barbaric and in violation of international law, for example. Those efforts really went nowhere, however, in large part because of the recognition by weak states that strong states would always have an advantage in capital ships. Future efforts to ban the submarine (especially those with an environmental logic) would probably have to be along these lines, but for the reasons laid out above, I’m not optimistic. In particular, the primary reason for declaring submarines icky was that they would most often be used against civilian merchant shipping, and that their use would almost invariably violate the rules that regulated the interaction of civilian and military shipping. That logic no longer applies; no one expects the Chinese, in the case of a militarized dispute over Taiwan, to engage in an anti-commerce campaign. Instead, Chinese subs will target American carriers, which are legitimate military targets.