A possible Chinese blockade of Taiwan, on the other hand, clearly has even higher stakes, and history suggests that crises may again occur in connection with Taiwan. Beijing might convince itself that a blockade-centered operation against Taiwan, designed to force it to reverse whatever purportedly offending action or rhetoric had caused China’s reaction in the first place, might not be hugely risky. In principle, China could scale back or suspend enforcement of the blockade at any point if it needed to, and could do so while saving face, especially if the blockade was conducted principally by submarines. Moreover, Beijing might believe that even a partially effective naval blockade could be a potent instrument of coercion against Taiwan. Such a blockade would almost assuredly be nowhere near complete or airtight.84 However, China would not need to stop all commercial ships transiting into and out of Taiwan. It would simply need to deter enough ships from risking the journey that Taiwan’s economy would suffer badly. The goal would likely be to squeeze the island economically to the point of capitulation. This solution might seem quite elegant from Beijing’s point of view: it could involve little or no loss of life, little or no damage to Taiwan itself, and the ability to back off the attack if the United States seemed prepared to intervene or if the world community slapped major trade sanctions on China in response.85 In any such blockade, China might well combine various elements of military power, including cyberattacks, into a multidimensional operation.86 It could attack command and control as well as reconnaissance capabilities both on land and in space (the latter intended largely to blind the United States). The centerpiece of the approach would probably be the PRC submarine fleet introducing a significant risk factor into all maritime voyages into and out of Taiwan by occasionally sinking a cargo ship, either with submarines or with mines it had laid in Taiwan’s harbors.87 The PRC submarine force has improved by leaps and bounds in recent decades. Over the past twenty years, China’s fleet of modern attack subs has grown from roughly two to forty.88 China’s precision-strike capabilities have improved to the point that China could conceivably use a preemptive missile and air attack against Taiwanese airfields and ports and associated infrastructure to hobble Taiwan’s ability to strike back, though it might choose not to attack Taiwan’s territory in the first instance.89 To allow humanitarian supplies to reach Taiwan, Beijing might offer countries the option of first docking in a PRC port for inspection before sailing to their destinations. In this and other ways, it could limit the dangers to innocent civilians. Since this strategy might require the Chinese submarine fleet to sink only a few ships to achieve the desired aims, even in the worst case Beijing might believe that it was acting humanely— threatening the lives of only 100 to 200 commercial seamen. Given the perceived stakes involved, Beijing could well consider this a reasonable risk. If they chose to try to break the blockade, the basic concept of operations for the United States and Taiwan would probably be to assemble enough forces in the western Pacific to set up a protected shipping lane east of Taiwan. To carry out that mission, the United States would need to establish air superiority throughout a large part of the region, together with Taiwan (and perhaps other countries such as Japan). The United States and Taiwan, and perhaps others, would also need to protect ships against Chinese submarine attack while coping with the threat of mines near Taiwan’s ports. And they might have to do all this without assured access to much of their satellite architecture, especially for imagery and some types of signals intelligence. That is because China’s abilities to shoot down or disable low Earth orbit satellites, through direct-ascent interceptors or directed-energy weapons or other means, have improved in recent years (even as some aspects of U.S. access to space have become more distributed and resilient as well).90 China might well be willing to shoot at American satellites even if it sought, at least initially, not to attack U.S. ships or aircraft. However, its willingness to do the latter in addition should not be dismissed, in light of Taiwan’s importance to Beijing. And its willingness would likely grow very quickly after the first Chinese submarine was sunk, which could happen if the U.S.-Taiwan blockade-busting and convoy escort operation shot back at a PLA submarine that had just attacked a cargo vessel. Escalation could happen fast. Later in a conflict, China might even consider using nuclear weapons in such an attack, despite its official no-first-use policy, as Columbia professor Thomas Christensen has argued.91 While the term “surgical nuclear strike” is almost oxymoronic and Strangelovian, to the extent it has any faint resemblance to reality, it is probably in a situation like an attack on a lone ship at sea. Establishing air superiority has become much harder for the United States and Japan in these kinds of scenarios because of the PLA Air Force’s modernization trends in recent years, combined with the limited options for basing U.S. aircraft in the region. Fortunately, modern U.S. stealthy or “fifth-generation” aircraft are still far superior to Chinese planes. Unfortunately, China now has close to 1,000 “fourth-generation” fighters roughly comparable to U.S. aircraft such as the F-15 and F-16. And it can base perhaps 1,000 aircraft within several hundred miles of Taiwan. A RAND simulation estimates that China might be able to surge about half that number in an attack on Taiwan or on the shipping around it, including a mix of air-to-air, air-to-ground, and electronic-warfare planes. Using a basic model and some simplifying assumptions, RAND estimates that the United States could prevent such a surge force from reaching most of its targets only by continuously keeping some two wings or about 150 aircraft airborne near Taiwan. The United States and its partners would likely succeed in such an effort because, according to RAND’s model, the United States’ fifthgeneration aircraft—F-22s and F-35s, and to a lesser degree F/A-18E/F Super Hornets—could have 50 percent more lethality and up to 90 percent less vulnerability than Chinese combat jets.92 But the success would come at a price, and only with considerable effort and difficulty. China could choose the time and place of its surge, and the United States (with any allies) would therefore have to be vigilant at all times. It would need to keep fighters airborne near Taiwan essentially for as long as the crisis endured. It would also need the continued presence of airborne warning aircraft in the vicinity. Bases on Okinawa are about 750 kilometers away from Taiwan, or about an hour of flight time; aircraft carriers might be kept roughly that close too. But other bases in the area—Misawa Air Base on Japan’s main Honshu island, Andersen Air Force base on Guam— would be 2,500 kilometers or more distant, meaning some three hours of flying each way to get to station and then return after flying a patrol. If aircraft and crews are limited to a daily flight average of about six hours a day, jets from Okinawa could average 1.5 sorties per day and those from the more distant bases slightly less than one sortie per twenty-four hours. That would translate into three hours a day on station flying from Kadena and 1.5 hours a day flying from Guam or Honshu. Put differently: Eight aircraft (plus or minus) based on Kadena would be needed to sustain one on station near Taiwan. Sixteen aircraft or so based at Misawa or Andersen Air Force Base would be needed to keep one on station near Taiwan. Averaging this out, keeping two wings of fighters aloft at a time could require about ten to twelve times that number being based in the region, or twenty to twenty-four wings—more than half the total U.S. military aggregate. (The RAND study actually estimates that fourteen to thirty would be needed, depending on specific assumptions; I have simplified the analysis above.93) Moreover, the United States could lose a number of aircraft in this process, perhaps even dozens. China could lose dozens or even hundreds. The backdrop would be set for escalation. Both sides would be increasingly tempted to attack the land bases and aircraft carriers from which planes operated.94 Ballistic missiles and ballistic missile defense would be important in this kind of engagement, too. China now has missiles, such as the medium-range DF-21 in the Dong-Feng series, that are capable of being fired from its homeland and reaching ships east of Taiwan. It is not clear whether the United States could blind China’s sensors adequately to deprive the PLA of targeting information. Any PLA attack against military facilities in a place like Okinawa therefore could well shut down runways for at least some stretch of time and destroy aircraft or ordnance and fuel stocks not in underground areas or hardened shelters.95 Again, Chinese nuclear attacks against American and any allied naval assets in the region are hard to dismiss categorically. The antisubmarine warfare (ASW) effort could have multiple aspects. The United States would probably be tempted to deploy its own attack submarines as close as possible to China—certainly in the Taiwan Strait, maybe just outside PRC ports. This approach would provide American submarines a good prospect of destroying PRC submarines at their source, before they were in position to fire on commercial shipping (or U.S. aircraft carriers) in more distant waters. However, this type of ASW could be escalatory if it involved attacks in Chinese territorial waters. Whatever happened near Chinese shores, there would surely be additional layers of American ASW farther out to sea. American ASW assets on ships and planes would accompany convoys of merchant ships as they sailed in from the open ocean waters east of Taiwan. These convoys might form a thousand miles or more east of Taiwan and enjoy armed protection from that point onward as they traveled to the island, and then later as they departed. The ASW assets would use sonar to listen for submarines, and for the sound of any torpedoes being fired. The United States would have to deploy significant numbers of surface combatants and airplanes such as P-3s and P-8s, as well as helicopters, to the region for this mission. Some would help protect U.S. aircraft carriers east of Taiwan. Others would provide additional protection to merchant ships or mine warfare vessels as they operated near Taiwan’s shores. Any Chinese submarine wishing to fire a torpedo at a merchant ship or aircraft carrier would then first have to run a gauntlet (if it were firing long-range antiship missiles, it could avoid some of the gauntlet, but would depend on targeting information provided by another platform). It would likely have to evade submarine detection as it left port, avoid any openwater search missions that the United States and Taiwan established, and then somehow penetrate the defensive ASW perimeter of whatever convoy it was attacking as it approached its target. To survive the overall engagement and return to port, it would then need to successfully negotiate all of this in the other direction. Unfortunately for Chinese submarines, this would be risky business. Unfortunately for the United States, Taiwan, and any other allies participating in the effort, it would also be risky for their assets, and some would surely be lost in the operation. During the Cold War, the effectiveness of ASW operations was commonly assessed at 5 to 15 percent per barrier. (Cold War barriers were often more linear and literal perimeters than would be likely here, but the fact remains that Chinese submarines would have to survive perhaps three types of pursuers on three different parts of their journey to or from home base.) By those odds, the typical Chinese submarine would do well to survive for two or three round-trip missions from base.96 But it might succeed in getting off several shots against valuable, and vulnerable, surface ships before meeting its own demise.97 A recent major RAND study on the U.S.-China military balance concurred with this broad result, especially in cases where Chinese submarines could be cued by sensors to a general area where a target such as a U.S. aircraft carrier might operate. In that case, a given submarine might have multiple opportunities to get off shots at lucrative surface targets.98 China might hope that a quick strike that sank a major U.S. ship and killed hundreds of Americans (or even thousands, in the event of a carrier sinking) would cause Washington to waver in its future commitment to the defense of Taiwan. The United States might face serious challenges in waters near Taiwan as well, despite the proximity of land-based assets that could join in the ASW hunt. Shallow waters are complex sonar environments in which sound waves bounce back and forth in multiple and unpredictable directions. This makes ambush a real worry, especially for the mine warfare vessels and surface ships that would have to escort commercial vessels all the way into Taiwan’s ports.99 Toward a Net Assessment Where does this leave us? Although I have not provided a detailed scenario analysis or combat modeling exercise here, the broad contours of the situation drive home some central points. They are also consistent with the more detailed modeling work of others, such as RAND. Once a Chinese blockade operation against Taiwan began, it would be difficult to be confident that it would end without a significant and escalating conflict pitting the United States, Taiwan, and perhaps Japan against China. If that occurred, sustained naval, aerial, and missile combat in and around Taiwan would likely lead to substantial losses on all sides. American advantages in fifth-generation combat aircraft and modern attack submarines would give the United States and its partners in the operation a significant edge. However, that edge is less than it used to be. China would clearly have the edge in geography, and it also increasingly possesses good attack submarines and precision-strike missiles as well. Myriad uncertainties, including the survivability of space and cyber systems, as well as ships and land bases, make it hard to be confident in predicting outcomes. Indeed, when I modeled uncertainties with respect to a possible U.S.- led invasion of Iraq in 2002, I produced a likely band of possible casualties in which higher estimates were literally ten times greater than lower estimates. This broad approach was eventually vindicated: it initially appeared, in April 2003, that U.S. losses might be very modest, but over the course of several years it became clear they were much higher. A U.S.- China war would be far more uncertain. For most scenarios involving blockades and other possible intense operations over and in the waters of the western Pacific, the United States and allies could probably muscle their way to victory over China at this point, as we approach 2020. Taiwan could also likely endure the privations that would be experienced during the military campaign before victory was assured.100 My best guess is that both sides would lose most low-altitude satellites and some cyber as well as command, control, and communications capability, so the war would ultimately be fought in a certain fog. Yet American high-tech advantages in weapons platforms would ultimately provide an important and probably decisive edge. This somewhat sanguine conclusion is far from certain, however. It hinges in part on a willingness in Washington to accept many thousands of American battlefield dead and at least some risk of nuclear escalation. Even though U.S. nuclear forces far exceed those of the PLA, China might conclude that its disproportionate interests in the Taiwan issue warranted nuclear brinkmanship, especially if China had already possibly lost thousands of its own people in conventional combat, which would heighten the stakes as well as the reputational importance of the outcome of the conflict. These concerns could be amplified if the United States, intentionally or not, began to strike the nuclear assets of the PLA in the course of conventional fighting near Taiwan. If that happened, China might face a “use them or lose them” dilemma.101 Nuclear attack against a carrier, or a high-altitude nuclear airburst over a base like Kadena (designed to destroy people and equipment immediately below, without generating lots of fallout), might seem particularly attractive options to Beijing. Nuclear escalation might happen in another way as well. Perhaps China would attempt a large-scale cyberattack against the United States (and quite possibly Japan too) in the course of such a conflict. Its hope might be, for example, to incapacitate the U.S. electricity grid—as National Security Agency director Admiral Michael Rogers said in 2014 that China might well be able to do. Such a result could badly incapacitate many of the Defense Department systems in the United States that require civilianproduced electricity to function.102 Even if such systems could be restored quickly, the U.S. electrical system as a whole might remain damaged for many months, as might national infrastructure needed to deploy military forces abroad. Whether the United States could muster a major military response at a time when its national capacities were needed to take care of the many Americans left in the cold and dark, without medicine or sanitation or clean water or viable living quarters, is an open question. China might hope that U.S. military assets would be devoted to missions closer to home, to prevent mass famine, privation, or disease, after such a catastrophe. China might do this, if already losing in a conventional conflict or expecting to lose, even if its own cyber systems could be just as vulnerable to attack.103 Countries sometimes start wars or escalate not out of complete confidence that they will succeed but after concluding that the risks of inaction outweigh those of even risky action.104 And if China conducted such a cyber operation, it might be the very type of non-nuclear strategic attack that the Trump administration’s Nuclear Posture Review envisions as a plausible basis for possible American nuclear first use.105