|Hat tip to Paul Huang for finding this comic.|
Two years back I wrote an article for Foreign Policy with the title “Taiwan Can Win a War With China.” In a recent interview with Jordan Schneider (for his podcast ChinaTalk) I stated that I can no longer endorse the declaration in that title. While I discuss my change of heart on the podcast, I think it is best if I fully write out why my assessment has changed.
I wrote that article in the early Spring of 2018. Around 70% of its contents reflected research presented in Ian Easton’s book, The Chinese Invasion Threat, another 15% or so was drawn from a journal article published by Michael Beckely in the International Security Review, and the last 15% or so drew from my own analysis. A lot of the writing and research behind Easton’s book comes from 2015-2016. Things have gotten worse, not better, since then, and if Easton’s more recent op-ed pieces are a fair judge of his opinions, he has also grown more pessimistic in the years since.
My pessimism is grounded in the nine months I spent in Taiwan in 2019. During my time there I had the chance to interview two dozen recently discharged conscripts, meet with security folks in the DPP (Taiwan’s governing party), security researchers in both the partisan and non-partisan think tanks in Taipei, engineers working on Taiwanese arms development, and members of the ROC Navy and Army officer corps. I traveled across the island to visit potential invasion sites in person and to get an assessment of how ordinary people across the country thought about defense issues. I had the chance to talk with a substantial number of visiting experts from other countries. More important than the security experts, however, was my exposure to Taiwanese democracy itself. I traveled to rallies for both the DPP and the KMT, I met PhD students conducting research on the Taiwanese political system, met up with local journalists, and watched Taiwanese political news every night with dinner. Now that I have returned to the United States I do not pay nearly as close attention to weekly ins-and-outs of Taiwanese politics as I did when I was living there. I do not consider myself and would certainly never bill myself as a “Taiwan expert.” But my 2019 experiences have caused me to fundamentally rethink my confidence in Taiwan’s ability to preserve its own autonomy.
A few analysts and researchers have been key to shaping my understanding of this issue. I don’t think any of them will endorse all of what follows–and one or two might object to the broader argument of the entire post–but I want to give credit where credit is due. Among Taiwanese, this includes Kitsch Liao, Crystal Tu, Paul Huang, and ‘Yuster’ Yu; among foreigners, Wendall Minnick and Tom Creedon deserve special mention. There are two or three others who cannot be named without causing them trouble with their current employment. They know who they are.
In the Foreign Policy article I penned in 2018, I described certain advantages Taiwan would have in any armed conflict. Many of these advantages still hold true. Taiwan is a mountainous island. A proper invasion of Taiwan would mean the largest amphibious invasion in human history. An operation on this scale could not be disguised or hidden. We would know about it weeks, and perhaps months, in advance. The weather in the Strait is treacherous; there only a few months every year where such an invasion could occur, and only a few beaches where an invading force could safely land. Our era is defined by a precision-munitions weapons regime. This sort of weapons regime heavily favors the defense. The same “AD/A2” logic that keeps the U.S. Navy away from Chinese shores can work to scare PLAN vessels away from Taiwanese shores. Missiles, drones, and mines can destroy invading ships many times their cost (and for what it is worth, my friends in the U.S. Navy judge that Taiwan’s anti-ship missiles are better than ours). The PLA is an organization with human-capital problems. I have met kids in the PLA; I naively expected them to exude the confidence, competence, and intelligence of America’s enlisted servicemen. They don’t. The PLA is an organization with no combat experience, and having told their people for so long that Taiwanese “reunification” is inevitable and pacification easy, it is not clear to me PLA peasant-soldiers and the broader Chinese public will be all that resilient in the face of military setbacks or disruption. The ROC military has spent decades preparing for a Chinese invasion. Some of these preparations (like the nuclear-proof air hangers built into the Hualien mountainside) are impressive.
All of that is still true. Even in a worse-case scenario, an invasion is a risky gamble for the Communists. The inherent advantages of the Taiwan’s geography and current technology mean that if the Taiwanese willed it, they could make their island an impenetrable fortress.
But there is the catch. The Taiwanese must will it.
You might divide the challenges Taiwan faces into two parts: problems of military strategy and problems of training, culture and morale. These problems can be laid at the feet of the ROC military (especially the ROC Army), but behind them lies another, more serious layer of dysfunction. This layer is more serious because it infects not the military but the civilian leadership tasked with reforming the defense system. Responsibility for military strategy and morale ultimately lies with Taiwanese politicians, and to a lesser extent, the voters who bring them to office. But Taiwan is marred by a dysfunctional civil-military relationship, destructive partisan infighting, and a spirit of defeatism. These political dynamics make it difficult for Taiwan to make the reforms that might guarantee its safety and autonomy.
The problems with Taiwanese military strategy are well known. The essential issues are these: for the last decade, Taiwanese force procurement has been weighted towards expensive, high-end platforms that are high on prestige but of limited utility in an actual conflict with the PLA. 20 years ago doubling down on the high-tech edge made sense, as it was seen as a force multiplier that might counteract the weight of numbers China could throw into the fight. But the situation has changed: the PLA has parity on just about every system the Taiwanese can field (or buy from us in the future), and for some systems they simply outclass the Taiwanese altogether. The Chinese thus not only have more equipment, but better equipment on top of it.
The solution to this problem is not to double down on buying expensive items like M1 Abrams tanks, which have difficulty maneuvering in the jungle, rice-paddy, and dense urban patchwork that is Taiwan (and which will be difficult to use once the PLA has gained air superiority, the necessary precursor to the amphibious invasion in which these tanks might be used), but to instead to adopt the military model of other small and middle powers facing large threats. That is the Iranians, the Israelis, and the Eastern Europeans. The Iranians keep the US Navy at bay through investments in sea mines, missiles, drones, and fast attack boats. In Poland and the Baltics, normal citizens are trained in sabotage and insurgent tactics to provide a deterrent beyond the normal line of battle. The logic behind Iran’s A2/AD screen is obvious. The Baltic countries do what they do not because they think guerilla fighting will be enough to destroy the Russian army, but because they understand how critical it is to keep the fight going for as long as possible. It might take a while for NATO to get troops on the ground; they must do everything in their power to prevent the Russians from taking their country in one fell fait accompli. They also know that the Russians face a different calculation if they must worry about putting down an insurgency while also fighting a conventional war with the rest of NATO at the same time.
In an ideal world the Taiwanese would be able to make a similar set of threats. “Invade us,” they might tell Beijing, “and it means cyber attacks on your critical infrastructure and missiles screaming towards sensitive targets as far away as Shanghai. If you are willing to bear that cost, and think you have the missile stores to degrade our own air defenses while reserving enough to use against the Americans, then you might be able to launch an invasion. But know that your invasion fleet will face thousands of mines and hundreds of missiles and drones before your first troops can make landfall. They will be faced with a military prepared to fight backwards inch by inch, using jungle, mountains, and cities for cover, specially trained to drag this fight out for weeks. Know that if even you manage to beat most of them, you will then be faced with an entire populace that has been trained insurgent tactics, with ready access to weapons stored all across the island. You will be forced to spend months putting down an insurgency in terrain perfect for insurgents…. all while fighting a conventional fight with the Americans and Japanese in the airspace and oceans of the entire Indo-Pacific!”
The interesting thing about all that is that everybody knows this is what the Taiwanese need to do. Every single study of every single think tank that travels to Taiwan comes away making these recommendations. Every touring academic from some American defense university makes these recommendations. Every ex-military attache or civilian appointee who has manned the Taiwan desk at the Pentagon makes these recommendations. I was struck by the lunacy of it all last July when I talked to a hard green ex-official one week, and then had a meeting with a National Policy Foundation defense intellectual the next week, and was surprised to find that their recommendations for Taiwan’s defense were exactly the same! (By way of context: a hard green would be a critic of Tsai Ying-wen from the pro-independence side, and the NPF is a KMT affiliated think tank). These guys would have trouble having a civil conversation with each other, but here they were identifying the exact same problems and endorsing almost identical solutions to them.
So everyone knows what real deterrence would require. But it is not what is strategized, nor what is trained for. Every year the Legislative Yuan drags the Minister of National Defense in to grill him on defense issues. He is invariably asked, “How long can we hold out in a war with China?” Every year whoever holds the position gives the same answer: “two weeks.” This reflects a belief on the MND’s part that the war will be decided on the landing beach. This is not an odd belief; the PLA shares it, and historically speaking amphibious invasions not stopped on the beaches are usually not stopped at all. But this is no longer a sufficient deterrence posture. It will likely take far longer than two weeks for the Americans to pop the formidable AD/A2 bubble China has built to keep INDOPACOM forces away from Taiwan, and the Taiwanese must be prepared to fight for as long as that will take.
Taiwan’s 2019 National Defense Strategy made some formal motions towards the strategy I discussed a few paragraphs ago, endorsing a conceptual shift from a decisive fight on the berm towards a posture which allows for a more multi-layered defense. The problem is that the ROC Army is not training for this. Or at least they weren’t in December 2019, when I last asked Taiwanese soldiers if they had ever trained in the tactics of a coordinated, fighting retreat or in using land based platforms to hit targets in the near littoral. The sad truth is that the ROC Army has trouble with training across the board. I have met artillery observers who never seen their own mortars fired, and shared drinks with an infantry officer who traveled to Thailand on his own dime to get basic TCCC training his own military did not offer. Those were professional soldiers; the situation with the conscripts is worse.
When people outside of Taiwan talk about the problem with the conscript system, they tend to focus on its dwindling size. Yes, the inability of the ROC military, especially the Army, to fill its own ranks is a problem. But the trash they fill it with is an even larger one. I would ask ex-conscripts questions like, “Would you know how to find cover if you were ambushed?”, “Were you ever trained on how to move around if the other side controlled the skies?”, “Were you ever taught what to do if the guy next to you was shot in the arm?,” “Did they ever tell you anything about the weapons, organization, or tactics of the PLA?” or “Did they teach you how to get from point A to point B without cell service, you know, using a map?” Negatives across the board. What they could tell me were stories of officers communicating orders through Whatsapp, time spent learning Army songs and doing yard-work instead of on maneuver drills, and how the totality of their marksmanship training consisted of firing one magazine from a single (prone) position on some eight to ten occasions.
One reason for the lax training is a shortage of supplies. The ROC Army has a shortage of bullets. Again and again I was told stories of officers who would fake training exercises in order to save on spare parts.  Han Kuang is a joke put on for propaganda purposes, not serious training. The military is risk adverse; real training might lead to training accidents, and a series of high profile accidents that led to unnecessary deaths has led them to soften training for the entire force. While reservist weapons stores are scattered across Taiwan, the million reservists that are supposed to use them are not drilled. Official reservists reported to me that they have no idea what they are supposed to do if ever actually called up. These troops exist only on paper. The problem is broader: the Taiwanese population is not seriously trained or mentally prepared for conflict. Nor do they take care of their soldiers. A military career is a low status profession (“好 鐵 不 打 釘….”). Military pensions were just slashed; military basing often does not provide housing for family members. Unlike service in the U.S. military, service in the Taiwanese military rarely provides marketable skills that can be used in different career fields. Most of Taiwan’s best minds flee service altogether. Officers willing to challenge outmoded tactics, or who study abroad in an attempt to learn from foreign militaries, are seen as a threat by the upper brass and side-lined.
Not all of the ROC Armed Forces fare so bad. The ROC Air Force, in particular, is highly esteemed by those who work with it, and their pilots are trained in the United States. The real stinker is the Army. The trouble is that this worst branch of the ROC Armed Forces is precisely the branch the population of Taiwan has the most exposure to. The vast majority of conscripts serve in the Army. Many enter the Army excited, hoping they might serve their country, or barring that at least become tougher men. Their disappointment is captured by the comic I have placed at the beginning of this piece. The first panel reads, “What I assumed military service would be like.” The second reads, “the reality.” This reality is demoralizing. Of the two dozen ex-conscripts I interviewed last year, only one was more confident in Taiwan’s ability to resist China after going through the conscript system. Most left their service more cynical and defeatist than when they entered it. I will go out on a limb: the greatest danger to the security of Taiwan is not the PLA Navy or Rocket Force, but Taiwan’s own demoralizing system of national service.
These problems are understood in Taiwan. I am not claiming some unique insight in typing them all out here. Anyone who spends a month in-country digging into these issues finds the rot. The harder question is what to do about it. I have lost confidence in the Taiwanese political system’s ability to manage the problem without some sort of external intervention. I described some of these issues in a piece I wrote for Foreign Affairs last year. You can read excerpts from that piece here. But there are additional sources of dysfunction. The ROC Army’s sorry state flows in part from its past job as the iron fist of the Chiang dictatorship. The DPP old guard remember the ROC Army as torturers and murderers; the second generation does not bear these scars, but has inherited their mistrust. This is seen even at the level of the general population: one of the grimly amusing results of Dean Karalekas’ opinion surveys is that the portion of the populace most eager to fight for independence is also the group that is least willing to let their children devote a career to the ROC Armed Forces. The upshot is a party run by activists, academics, and lawyers who do not understand the ROC national security state and have a poor working relationship with most of the blue-leaning generals and bureaucrats who man it.
Taiwan’s authoritarian legacy also makes it hard for Taiwanese political leaders to employ useful tools for shaping public opinion and preparing their country for crisis. In the old days, conscription was a tool of ideological indoctrination. Democratization defanged the Political Warfare Bureau, and made it illegitimate to use conscription as an explicit system for boosting morale or building a sense of national identity. Public education campaigns whose rhetoric or imagery is too militarized are difficult sells for very similar reasons. Authoritarian rule casts a long shadow in Taiwan. Anything too reminiscent of the dictatorship is a non-starter—especially for the Greens now in power.
Behind all of that is a more fundamental problem. Taiwan’s leaders are afraid to ask their people to make meaningful sacrifice to preserve their freedoms. Deterring away the PLA has costs. It will require taxes. It will require a real conscription. It will require real reservists. It will mean turning Taiwan into something like a garrison state. Inspiring the Taiwanese people to make these sacrifices is the job of President Tsai and the leading members of her party. Thusfar she has elected not to do this. Indeed, her administration has focused on almost everything but this. There are sound electoral reasons for this. The details of national defense do not push any Taiwanese identity culture-war buttons. Real preparedness would bring with it real costs. It is far easier to settle for pageantry and symbolism and hope that if there ever is a crisis, the 7th Fleet will come to save you.
Should we so easily promise that the 7th Fleet will be there to save them?
At this point an admission is due: it is easy for me to write all that. It is easy for me to judge, to pontificate on why some other country needs to militarize its society. The costs are not my own. I will not be called up for a year’s conscription. I will not be paying higher taxes; it is not my culture at stake. But the call is now going out for America to commit itself to Taiwan’s defense. I cannot advocate sending American servicemen to die for the sake of a country that is not serious about defending itself. Unless American diplomats deliver a similar ultimatum to the Taiwanese, I am not sure they ever will be.
 Tanner Greer, “Taiwan Can With a War With China,” Foreign Policy (25 September 2018).
 Ian Easton, The China Invasion Threat: Taiwan’s Defense and American Strategy in East Asia (Washington DC: Project 2049 Institute, 2017); Michael Beckely, “The Emerging Military Balance in East Asia,” International Security Vol. 42, No. 2 (Fall 2017), pp. 78–119.
 For example, see Ian Easton, “America Should Put Troops in Taiwan,” Taipei Times (9 March 2020). Included in that piece is the following statement, stark in its judgement:
Neither America nor Taiwan seems to have a realistic strategy for responding to an all-out Chinese attack. Both countries have defense plans for raising the costs China would have to bear to conquer Taiwan, and there is little doubt they could wage a bloody war of resistance, but these defense plans are reportedly uncoordinated, underfunded, and uncertain. No one seems to know how they would actually win the war. Sinking the first few hundred ships to cross the Taiwan Strait is necessary, but hardly sufficient, for victory.
 I spent an earlier stint in the country for language learning purposes in 2015-2016. For the curious, I think Taiwan is a far better place to learn Mandarin than the mainland is. That may be a topic for a different day.
 This clarion call was first sounded in English with William S. Murray’s article, “Revisiting Taiwan’s Defense Strategy,” Naval War College Review, Vol. 61, no 3 (Summer 2008). Several dozen reports, testimonies, op-eds, etc. have been written on the same theme since then. For a selection of recent examples, see J. Michael Cole, “How Taiwan Can Defend Its Coastline Against China,” The National Interest (30 June 2019); Colin Carrol and Rebecca Friedman Lissner, “Forget the Subs: What Taipei Can Learn From Tehran About Asymetric Defense,” War on the Rocks (6 April 2019); Drew Thompson, “Hope on the Horizon: Taiwan’s Radical New Defense Concept,” War on the Rocks (2 October 2018); Michael A. Hunzeker, Alexander Lanoszka, Brian Davis, Matthew Fay, Erik Goepner, Joseph Petrucelli and Erica Seng-White. A Question of Time: Enhancing Taiwan’s Conventional Deterrence Posture (Arlington, VA: Center for Security Policy Studies, 2018), pp. 63-105; Michael Bekely, “The Emerging Military Balance in East Asia: How China’s Neighbors Can Check Chinese Naval Expansion,” International Security, vol 42 , iss 2 (2017) pp.78-119; Ian Easton, Mark Stokes, Cortez A. Cooper III, Arthur Chan, Transformation of Taiwan’s Reserve Force (Santa Monica: RAND, 2017), pp. 55-59; Lauren Dickey, “Reestablishing Deterrence: A Guide for Taiwan’s New President,” War on the Rocks (20 Jaunary 2016); Michael J. Lostumbo, David R. Frelinger, James Williams, Barry Wilson, Air Defense Options for Taiwan: An Assessment of Relative Costs and Operational Benefits (Santa Monica: RAND, 2016), 73-91.
 On the PLA’s two week timeline, see Easton, The China Invasion Threat, ; on the historical success rate of amphibious operations generally, see Theodore Gretchel, ,At the Water’s Edge: Defending Against the Modern Amphibious Assault (Annapolis: USNI Press, 2013), passim.
 To say nothing of the political requirements of rallying the American public in favor of Taiwan’s defense.
Huang Jiezheng recently had a funny riff on the whole issue which captures MND’s frustration with this question:
Huang is correct here. The oft-asked question is a dodge on the part of civilian officials, an attempt to pin the military with problems they have not been willing to try and solve themselves. And yet…. “two weeks” is still the wrong answer to give!
黃介正, “拜託不要再問國防部長「台灣可以撐多久」了！” 風傳媒 (25 August 2020)
 For examples, see Paul Huang, “Taiwan’s Military Is a Hollow Shell,” Foreign Policy (15 February 2020); Vanessa Molter, “Taiwan’s All-Volunteer Force Transition Still a Challenge,” The Diplomat (31 August 2019.
 Paul Huang recently wrote an excellent piece on this topic. See “Taiwan’s Military Has Flashy American Weapons but No Ammo,” Foreign Policy (20 August 2020). See also Wendell Minnick, “How To Save Taiwan From Itself,” National Interest (19 March 2019).
 Tanner Greer, “Taiwan’s Defense Strategy Doesn’t Make Military Sense,” Foreign Affairs (17 September 2019).
 Dean Karalekas, Identity and Transformation: Perceptions of Civil-Military Relations in the Republic of China (Taiwan) (PhD Dissertation, National Chengchi University, 2016)
 For a similar analysis see Philip Caruso, “Taiwan Needs More Than Election Victories to Fend off China,” Foreign Policy (17 January 2020).