Simulation of a Crisis in the Taiwan Strait

Together with the ICONS Project at the University of Maryland’s Center for International Development and Conflict Management, Brookings held a two-day exercise on May 20 and 21, 2009 simulating a crisis in the Taiwan Strait. The participants were divided into three main teams—for the fictional governments of the U.S., PRC, and Taiwan—as well as a control group and various facilitators.

Three separate two-hour sessions were held over the two days, followed by an afternoon debriefing and discussion on day two. The control team tried not to interfere with the actions and choices of the three governments, except in its creation of the original scenario for day one and then its modification prior to the resumption of play on day two. Otherwise, the control team interacted with the players through its ability to represent opposition groups, media, and other nongovernmental players in all three places, its ability to act for countries like Japan, and its prerogatives to “create events” such as military accidents and unconfirmed news reports about government activities that did not presuppose any actual decision by one of the three governments.

The setting was early 2013. No change was presumed in the leadership of the three governments relative to mid-2009. Relative continuity was also assumed in the nature of the international environment (e.g., continued Chinese military modernization, continued development of cross-Strait commercial ties, ongoing difficulties for the United States in the struggle against terrorism with particular challenges in Afghanistan).

At the beginning of day one, the essence of the scenario was that China’s People’s Liberation Army (PLA), deliberately sought to provoke a Taiwan response to assertive military maneuvers. The PLA took this initiative without the knowledge of its country’s own civilian leaders as a ploy to disrupt an improvement in cross-Strait relations that would harm its institutional interests. Before the second session of the first day, the most important modification to the scenario was a revival of the Taiwan independence movement among opposition politicians and other key actors.

On day two, the main change to the scenario (presumed still to be in 2013, but several months later) was a never-confirmed but moderately credible report that Taiwan was pursuing a nuclear weapons capability—which led to a Chinese decision to sink a ship headed toward Taiwan and suspected of carrying nuclear-related technology. China subsequently was assumed to impose a partial naval blockade on Taiwan in regard to six additional ships it considered particularly suspicious.

Several key results emerged from the simulation. On day one, despite the fact that the PLA “got away” with its provocation, and that the civilian PRC government sided with it in the ensuing crisis, cool heads generally prevailed. Despite a military accident and modest loss of life, no escalation occurred and no shots were deliberately fired with the intent to do harm. That said, the various sides did try to maneuver for position, with the Chinese leadership going so far as to attempt to impose (at least by declaration) no fly zones near Taiwan after the resumption of the independence movement. These results tended to argue against the notion that accidental or inadvertent war would be easily caused.

On day two, while war was still avoided the situation became much more tense. China believed the press reports about a nuclear weapons ambition on Taiwan and found that prospect unacceptable. But other means, most notably support for enhanced international weapons inspections, were clearly seen as the preferred tool for handling the problem, and all three sides agreed to this notion. Still, there was a fair amount of ongoing tension and potential for further worsening of the situation even as the crisis began to be defused towards simulation’s end.

A number of other key moves by the various actors were noteworthy:

• Once tensions mounted, and limited uses of force had been carried out, all sides tended to increase military readiness levels. This was a natural reaction for signaling purposes, and also for defensive purposes, since ships in port and airplanes exposed on runways are easy targets. But of course it could also be interpreted as an escalatory action and preparation for possible offensive measures. One factor contributing to misinterpretation is the unique terms that each actor uses to refer to alert levels, the meaning of which is not always clear to others. Thankfully the parties to the simulation seemed to understand each other’s reactions, even as they also felt the need to ready their own forces. It is to be hoped that real-world participants would display a comparable level of understanding, and of steady nerves.
• When accidents occurred that were quickly recognized as accidents, such as ship collisions that could have served little deliberate purpose for any government, responses tended to be calm. When accidents or misunderstandings occurred but were not recognized as such, however—as with suspicions about Taiwan’s possible pursuit of a nuclear weapon program, something that Taiwan players never acknowledged (and that the control team never fully clarified for the various governments in the simulation), the situation was somewhat harder to control.
• Expanding the number of players can change the dynamic of a crisis. U.S. contacts with Japan increased pressure on Chinese leaders primarily because they had difficulty in gauging whether those contacts were benign or malign. Public opinion on all three sides of the triangle can intensify a crisis because it puts leaders’ domestic power at some risk.
• Communications between two of the actors in a three-actor game can create anxiety in the third actor.
• China’s continuously growing military capabilities and economic strength did not make its government reckless, in this simulation. But they did seem to add confidence to the Chinese leadership and make the PRC somewhat more assertive and somewhat more inclined to pursue an outcome that advanced its interests and protected its pride.

Several purposes were served by the overall exercise. The first, noted by participants who had handled crises in real life, was the simple matter of practice. This may sound simplistic, or unrealistic—after all, how could such a contrived setting be akin in any way to true crisis management? Of course, there are limits to the learning process. But the need to process incomplete information, react to the actions of others with whom one’s communications are limited, take sequential steps for dealing with a scenario that takes on a life of its own, and handle as well the internal dynamics of one’s own country or government can be mimicked to an extent.

A second benefit of this exercise in particular was to see which red lines, and which dangers, were most serious. We found that accidental war was less likely than an escalating crisis over real, substantive issues of importance to the two sides. In particular, any hint of nuclear weapons programs or independence-oriented proclivities from the Taiwan side were very hard for the Chinese side to accept. This was known to all in advance; less easily discerned in advance, however, was how a crisis sparked by such issues could intensify through its own dynamics even when no one intended that. Again, there was a generally successful defusing of the problem in this game. But all parties witnessed the potential for things to get out of hand—especially since parties had the dual motivations of trying to resolve the crisis short of war while also trying to be sure their own country benefited from the crisis (viewing it in part as an opportunity).