Book Review: The Allure of Battle by Cathal J. Nolan

Image from

The Allure of Battle: A History of How Wars Have Been Won and Lost by Cathal J. Nolan is precisely what it’s subtitle promises. I wouldn’t consider myself a military history enthusiast, so I can’t easily assess Nolan’s presentation of (mostly European) war in the context of his major themes. But Nolan’s history illuminates the ways in which innovation in war has many of the same issues as innovation elsewhere. In particular, knowledge spillovers matter a great deal.

The bulk of Nolan’s book is a history of European warfare. Running through this history are a number of themes. In this post, I thought I would attempt to translate his major themes into a unified model of war. In doing so, I am cramming a more nuanced and subtle presentation of themes into a simplistic theoretical box. Apologies to Nolan.

A Model of War

Suppose war between two parties can be represented as a series of biased coin flips. Heads I win, tails I lose. Each coin flip represents a particular engagement. The “coin flips” of each engagement may be biased, because one side has better tactics, technology, elan vital, etc.

Many engagements might make up a battle, and many battles might make up a war. Losses destroy military resources, and a country can only fight so long as it can supply these resources. In general, bigger countries can supply more resources.

The crucial assumptions of the model are this:

  1. Biases rapidly erode, so that before too long both sides have an equal probability of winning an engagement.
  2. Countries have different beliefs about the extent and persistence of bias.

The first assumption is a consequence of knowledge diffusion. Superior tactics are observed and copied. Superior technology is liberated from captured foes or stolen via espionage. This is the same problem that bedevils private companies that want to retain the “edge” they get from R&D. Nolan supplies copious examples of the tendency for any one side’s “edge” to become eroded as its best ideas are diffused.

The second assumption is rather banal. It just asserts that nations (or more particularly, their leaders) do not share the same information or beliefs.

These assumptions have two major consequences, which are woven throughout Nolan’s presentation of military history.

War of Attrition

The first major implication is about the length and nature of war. Because the outcomes of engagements are initially biased, if a belligerent has a large enough bias in its favor, it may be able to defeat a larger enemy before it’s bias is eroded. For this to happen, the aggressor would need to develop a substantial military advantage in isolation, to prevent knowledge from diffusing to the enemy. Does this ever happen?

Perhaps it can. Though it is not part of Nolan’s narrative, this can account for the rapidity with which Europe subdued its colonial empires with relatively small military forces. European colonial powers were isolated from their colonies for centuries, developing powerful military technologies and tactics on the battlefields of Europe. Crucially, these developments were isolated from future colonies, allowing a substantial military advantage to accumulate by the time war began. In this case, the bias seems to have been large enough to overwhelm differences in the capacity to deploy military resources.

However, within the theater of Europe itself, the answer appears to be no. Developing a military edge appears to require military engagement to perfect. These engagements, then, become the vehicle for diffusing knowledge about superior tactics and technology. Nolan gives many examples of this kind of “in-war” innovation and learning, ranging from the discovery of how close your infantry had to advance before they could stop and accurately hit entrenched targets, to the discovery that WWII bombers needed to practice large-scale area bombing to destroy military targets because they were so inaccurate.

“God is always on the side of the big battalions.”

Except in unusual cases (such as long periods of isolation), the first assumption implies that most wars quickly devolve into a war of attrition, where each side trades losses until the one able to muster fewer military resources capitulates. Yes, there may be lucky and unlucky streaks, but in the end the law of large numbers ensures both sides trade wins and losses on roughly equal terms. Battles are rarely decisive, but merely accelerate or prolong the war of attrition. It is eventually won by the bigger country (or coalition), which can afford to fight for longer. Nolan is fond of quoting Voltaire: “God is always on the side of the big battalions.”

This is a terrible kind of war that few would willingly choose. It is costly, miserable, and long. So why do we nonetheless have war? This is the second implication of this model.

False Beliefs and Much to Gain

The second assumption allows nations to vary in their beliefs. Given that war by attrition is so terrible, wars tend to be started only by countries that incorrectly discount the probability of war by attrition. These are countries that have excessive confidence in the size and durability of the bias in their favor. In this model, war is sort of like the winner’s curse, whereby auctions are won by whichever bidder most overvalues the good for sale. Just so, war is started by whichever country is most mistaken about the scale and durability of its’ advantage.

Nolan calls this “the short-war delusion.” It is a belief that the aggressor’s advantages are so profound that definitive defeats will quickly force the enemy to accept terms. It is a delusion, because this is so rarely the case.

There are some additional and interesting complexities. Frequently, the aggressor really does have an initial advantage over its enemy (which is how the short-war delusion is born). This allows them to have a series of significant early victories, cementing the leading general’s status in the annals of military history. Unfortunately, these early victories prompt the aggressor to increase their confidence that the bias will remain in their favor, leading them to overreach in their ambitions and sustain their war-making effort after the bias has completely eroded. Think of Germany or Napoleon deciding to invade Russia, for example.

Second, war is more likely to be started by countries with more to gain from victory, all else equal. Small nations on the rise, such as Germany or Prussia, enjoy larger proportionate gains in resources from any given victory, and they often tend to be aggressors. Leaders that have a love for war in and of itself are also likely to seek it out (here, Nolan points to Napoleon, Hitler, and the Japanese military in the run-up to WWII).

Lastly, this model implies an endogenous termination of peace. The military does not stand still during peacetime, but continues to develop new technologies and tactics, practiced and developed in secret war games. However, this data is highly imperfect, because perfecting military tactics and technologies can only be done in actual battle. During this period, a lack of good data on how these new technologies and tactics will play out in a real world may allow beliefs to become uncoupled from reality. As beliefs in the extent and persistence of bias diverge from the discipline of actual data, eventually there will be a country that grows overconfident. It will initiate war, secure in the (false) belief that “this time it will be different.” They will win decisively, and avoid a war of attrition. Think here of the long peace, that finally resulted in the cataclysm of World Wars I and II.

Empirical Support

To be clear, this is my interpretation of Nolan’s model, and I have simplified. I’ve probably emphasized the technology spillover part more than he would, given my interests. Anyway, is the model true? The bulk of Nolan’s book is a reasonably comprehensive retelling of the history of European war, meant to provide the empirical support for the model I’ve sketched out above. As noted above, I wouldn’t consider myself knowledgeable enough to assess his presentation. But, I plan to keep this framework in the back of my mind the next time I read something in the same genre.