I’ve seen a couple of articles now on how we should view Russia’s invasion and annexation of Crimea through a game theoretic lens. The articles which appeared in the New York Times and the Frankfurter Allgemeine Zeitung, which suggest that unfolding events should be understood and managed as stemming from rational actors in a structured situation, need a word of caution attached to them. In the case of Russia’s actions, game theory is one possible narrative or explanation among many, and not necessarily the one best supported by the facts. And accepting game theory uncritically can be dangerous, as it means uncritically accepting concepts such as “credibility” and “deterence,” which in practice often mean laying out and following through on a sequence of increasingly aggressive actions.
I have no beef with game theory. It can be a very useful analytic and research tool for the social sciences and my current research makes use of a couple of game theoretic models. But I am also very aware of the danger of fitting my preferred policy explanations on the very complex situations I’m studying. Whether my work which hangs a lot on “trust” is falsifiable is something I am currently worrying about. And I have the advantage of studying municipal utilities, so I have thousands of potential cases which I can use to test the veracity of my work.
But here we have only one case. If a game theorist had predicted Moscow’s invasion that would be one thing, my apologies if I missed it, but even the day before predictions of invasion were scarce to be found. Furthermore we have an alternative explanation, a psychological one, that clashes with the rational one offered by the game theorists:
An examination of the seismic events that set off the most threatening East-West confrontation since the Cold War era, based on Mr. Putin’s public remarks and interviews with officials, diplomats and analysts here, suggests that the Kremlin’s strategy emerged haphazardly, even misleadingly, over a tense and momentous week, as an emotional Mr. Putin acted out of what the officials described as a deep sense of betrayal and grievance, especially toward the United States and Europe. –New York Times, March 7, 2014
This second explanation, or narrative, strikes me as closer to the facts. Which doesn’t mean that game theory should be ruled out, just that it should, in the absence of scientific verification, accept its place as a competing explanation for a complex and evolving situation. As such scientific verification is unlikely to come, it is up to Western decisionmakers to use their intuition and judgement to decide how applicable game theory is. It is up to them and their advisors to determine whether Putin is in fact behaving rationally (or more likely, when he is, and when he isn’t), and how Putin defines his interests, his allies interests, and Russia’s interest (and whether they can get him to redefine any of these).
That Dr. Cowen uses the word “lens” in the title of his article suggests that he implicitly gets this. Dr. Milinski gives us a harder line (in German), that it “looks” like Putin must have used cost benefit analysis to decide to hold the Crimea. No, it does not look that way to all observers. And one must be careful because what one sees can be determined by the lens one uses. Until evidence suggest otherwise, we need to make use of all the lenses we have.