Le’ Notes #16: Rational choice theory, part 1

This post discusses the rationalistic approach to war decisions and application of game theory in Napoleon’s Battle of Waterloo.

Why do people go to war?

There are a number of approaches that attempt to explain how people, especially political and military leaders, make decisions. I focus on wartime decisions: the decisions to start, wage, and end wars.

In the study of war decisions, the first we stumble upon is rational choice theory. This theory assumes that when making decisions, humans are completely rational. Of course, there are certain caveats. A person can only be rational so far as the amount of information they have at a given point in time. Remember, hindsight is always 20/20.

I present the views of Fearon and Kirshner in regards to the rationalistic approach of war. We’d see that the rationalistic approach does have its merits and drawbacks in assessing war decisions. And to cap it all, I’ll present Mongin’s application of game theory in assessing the feasibility of Napoleon’s decision in the Battle of Waterloo.

Fearon [paywall] attempts to “provide a clear statement of what a rationalist explanation for war is.” He argues that rational arguments in neorealist theory fail to provide a prewar explanation for the behaviour of rational leaders. Among the six characteristics of the international system as postulated by neorealist theory, all six fail to provide an explanation as to why leaders choose to fight in the first place despite the presence of other solutions to conflicts. Often, states fail the bargaining process and go directly to war, Fearon observes. However, Fearon does not reject the neorealist argument. Rather, he attempts to build and enhance them even further.

Two of Fearon’s main arguments include information asymmetry, or the incentive to deliberately misrepresent or withhold information to leverage the bargaining process. Supposedly, information is power, and keeping a potential adversary in the dark (or knowing what they don’t know) leverages the bargaining position of, say, party A. A can then use their knowledge against B to put B in an unfavourable position, either through blackmail or whatnot.

However, lack of information also tends to lead to problems in conveying the right message. Because B lacks the proper information about A’s nuclear power plant, B launches a preventive strike because they think A is building nuclear weapons. Only after the preventive strike does B realise A is building a nuclear power plant and not nuclear weapons. But by then, there’s only smoking rubble left.

Another way to think about it is this: suppose that you want to know China’s military capabilities. You know their White Paper and military acquisition plans are sketchy at best. So, how do you force China to spill the beans? Declare war on them. They would be forced to reveal their true military capabilities.

States may be forced to use war as a credible means to reveal private information about their military capabilities. – Fearon, p. 400

Second is the commitment problem, where states are also incentivised to defect on previously agreed negotiations. Since the system is anarchic, there is no third party that can be relied on to prevent other parties from defecting on their agreements. Hence, I could say today I won’t attack you and sign a treaty. But, next Christmas, I might attack you. Who’s going to stop me? This problem is also reflected in the prisoner’s dilemma game. In the classic game, it is always better to be the first party to defect. This is related to first-strike capability, which, most of the time, heightens the advantage of the attacker.

Kirshner [paywall] argues directly against Fearon. First, Kirshner accuses Fearon of downplaying the political dimension of war, since:

States evaluate the costs and benefits of fighting, and not fighting, in political terms. – Kirshner, p. 145

Fearon tends to think that states go to war purely over materialistic gains, such as resources and information. However, Fearon does indeed overlook the political aspect of war. In some cases of war, states will go to war if they see the political benefits as being higher than the material cost. Whether or not the move results in a loss is irrelevant, as long as the state acknowledges that there is a higher political gain that can be obtained with the resources they have at hand. The US dived into World War II because they did not want to see a Europe (where all their buddies are) fall into Nazi hands.

He then attacks Fearon’s central premise, which is paraphrased as follows:

Given identical information, rational states should come to the same, if not, similar, conclusions. The fact that there different conclusions among actors indicates the presence of different or private information. – paraphrased, Fearon, p. 392

Can you spot what’s wrong with that? Kirshner pointed out that the entire premise is flawed because even though two rational people are given the same information, they might draw very different conclusions. Kirshner draws an example from football “analysts”. Kirshner does not doubt the rationality of these trained analysts, who supposedly know everything there is to know about a certain team, down to the very stamina of the players. However, even they could not agree as to which team would win, despite having the same access to the same information.

Of course, international relations and war cannot be likened to football. In fact, since IR and war have even more uncertain variables, analysts and experts are even more prone to misinterpreting information. Hence, the entire process becomes even more complex. Kirshner then proceeds to reject Fearon’s claim that war is fought over private information as “unlikely” and of little practical significance.

As we’ve seen from Fearon and Kirshner, the rationalist explanation for war seems too parochial in nature as it often discounts uncertain character of international politics and war. Furthermore, the rationalist explains also discounts the complexity of the decision-making process in humans and state organisations, especially in regards to availability of information and the processing of said information.

A game-theoretic analysis of Waterloo


The Battle of Waterloo


Let’s now move on from Fearon and Kirshner’s debate into the operational/tactical realm of war. For this case, we’ll be visiting Napoleon’s last battle at Waterloo and see how Mongin [PDF] applies game theory in deciding whether or not Napoleon’s decision to detach Grouchy was rational or not.

First of all, Mongin’s aim is to contribute to the usual approach to explaining historical decisions. Usually, the “analytic narrative” approach is employed. However, Mongin notes some methodological problems with this. One of them is the tendency for the historian to “take sides”, as evident in the several accounts of Napoleonic history that are either in favour or against Napoleon’s decision to detach Grouchy from his main army. Mongin’s aim is to settle the debate and enhance the methodology by employing game theory to add to the objectiveness of the debate. Mongin then concludes that Napoleon’s decision was indeed rational, considering that it was the best play he could do within those circumstances.

Since this is a short note piece and I’m not an expert on game theory, I leave it to you to assess whether or not Mongin’s interactive game-theoretic model is feasible in assessing the rationality of Napoleon’s decision.

While game theory is a mechanical way to represent war as it often simplifies the complex and confusing conditions of war, Mongin’s game-theoretic approach to analysing the Battle of Waterloo enhances the scope of the “analytic narrative”. It is often the case that historians emphasis the role of the individual, in this case, Napoleon, which prevents an objective analysis on the decisions of the figure. By employing game theory with regard to historical accounts, Mongin manages to analyse the Battle of Waterloo from an extended and more objective perspective.

The bells ring…

Here’s a quick recap of what I’ve discussed thus far:

  • Rational choice theory assumes that as long as the information is available, actors will make rational decisions as best as they can.
  • According to Fearon, states go to war due to information asymmetry and commitment problems.
  • So far, the rationalistic school of thought in international politics shows some serious disregard for the human element by simplifying uncertainty in the decision-making process.
  • Mongin’s game-theoretic model, while still far from perfect and mechanistic, does indeed allow historians to add more scope and objectivity to their analytic narratives.


Comments are closed.

Website Powered by WordPress.com.

Up ↑

%d bloggers like this: