Game theory and diplomatic history: Introduction to the Rhineland crisis
In this second post of our series on game theory, Frank Zagare sets the scene for the Rhineland crisis of 1936: the players, the potential outcomes, and the preferences of each player. For the basics of game theory, read the first part in our series.
Nuanced game models detail the private information each player possesses, information that is common to all the players, and each player’s belief about what is likely to take place as the game is played out. Each of these elements can be illustrated with a straightforward example that will serve as a reality test for assessing some of the benefits and pitfalls of constructing a case study around an explicit model, game-theoretic or otherwise: the Rhineland crisis of 1936.
Under Article 42 of the Treaty of Versailles of 1919, which concluded World War I, Germany was “forbidden to maintain or construct any fortifications either on the left bank of the Rhine or on the right bank to the west of a line drawn 50 kilometers to the East of the Rhine.” Like Article 231, the so-called war guilt clause, Article 42 was almost universally detested within Germany. For the French, however, it was strategically vital.
Because control of the Rhineland opened Germany to a French attack, a demilitarized Rhineland made it difficult for Germany to invade any country to its south or east (i.e., Austria, Czechoslovakia, or Poland).
Thus, after 1919, French policy makers were motivated to see that the stipulations of the Versailles Treaty were enforced (i.e., to maintain the status quo) in the Rhineland, while German leaders sought to undermine them. Until 1936, the status quo held. According to Kagan, Adolf Hitler, the German Chancellor, first spoke about the possibility of remilitarization in the summer of 1935. But he did not act until March 7, 1936, when a small contingent of German soldiers crossed the Rhine. The French government protested (halfheartedly), but did not respond militarily.
Since the Rhineland crisis is straightforward and relatively transparent, it is easy to model. The two core players, the French and German decision makers and their choices are apparent. Rather than contesting the military status quo in the Rhineland, Hitler could have decided to accept it — at least for the time being. In the contingency that he decided to unilaterally upset the status quo, however, French leaders would have to decide whether or not to resist. Finally, the historical record reveals that Hitler also considered what he would do if the French responded militarily. He had to consider this contingency since his initial choice would be made without certain knowledge of how the French would react.
Thus, as is oftentimes the case in international affairs, the Rhineland crisis of 1936 was a game of incomplete information. Neither player knew for sure the other’s preferences.
These choices and the general strategic environment that existed in March 1936 are captured by the extensive form game given in the diagram above, which also summarizes, verbally and symbolically, the likely consequences (i.e., the outcomes) of the various choices, contingent or otherwise, available to the players.
The players’ preferences and their beliefs about each other’s preferences must first be spelled out. Consider first German preferences, which are relatively easy to specify. At the time of the crisis, Germany’s armed forces were patently outgunned, which is why Hitler’s generals initially opposed the remilitarization. So to gain their support, he promised that he would withdraw should the French resist, which would only be the case if Hitler (i.e., Germany) preferred the outcome labeled Germany Defeated (with payoff GGD) to the outcome labeled Conflict (with payoff GC). Assuming that Germany was a revisionist power in 1936 (a safe assumption), it follows that, for Germany, France Concedes >G Status Quo >G Germany Defeated >G Conflict where ‘>G’ means ‘is preferred by Germany to’.
On the basis of the testimony of the French ambassador to Germany, we know that, well before the crisis, the French government had decided that it would not risk a conflict by reacting militarily should German troops cross the Rhine, that is, that it preferred to concede rather than risk a conflict with Germany. The remaining French preferences are apparent, so, rather than belabor the obvious, I will simply stipulate that Status Quo >F Germany Defeated >F France Concedes >F Conflict where ‘>F’ means ‘is preferred by France to’.
The easy and natural correspondence between the simple game form depicted in the image and the strategic situation that existed in 1936 suggests a straightforward analysis and explanation of the crisis. To be sure, the analysis is facilitated by the fact that the game form of the Rhineland Crisis is a generalized ‘off-the-shelf’ game model now called the Unilateral Deterrence Game.
Now that you have the background, look out for the next part of our Game Theory series, in which game theory is implemented in order to fully analyse the Rhineland Crisis of 1936.
Frank C. Zagare is UB Distinguished Professor of Political Science at the University at Buffalo, The State University of New York (SUNY). Professor Zagare’s main research interests lie in the nexus between security studies and game theory. He is the author of Game Theory, Diplomatic History and Security Studies (OUP, 2018).
