Learning to Win, a Hudson Institute Report

I was pleased to contribute to a recent Hudson Institute report on operational innovation and the importance of learning to today’s U.S. Navy. The whole report is available in PDF at this URL.

In my section, I focused on the importance of creating a learning organization by coupling individual learning outcomes to the Navy’s operational objectives. The Navy of the 1920s and 1930s was able to do that and there are lessons from that experience. More effective communication, more detailed orders, or more thorough checklists will be insufficient. Instead, the Navy must create an environment that stimulates and rewards creative problem-solving.

High-level goals should focus attention on the most important operational challenges. Boundary conditions—such as timing, geography, available resources, and other constraints—should constrain the scope of problem-solving. Within that scope, open-ended exploration must be encouraged and rewarded.

That will allow the Navy to conduct multiple experiments in parallel. Different officers and their teams will explore alternative ways of approaching challenging problems. Parallel experimentation will accelerate learning and create a broader understanding of potential solutions and available options. Creative ideas will emerge from the bottom-up, based on new lessons and the friction between established operating concepts and desired outcomes.

Captain William V. Pratt (second from left), Commander, Destroyer Force, Pacific Fleet observes exercises with members of his staff in 1921

When those creative ideas are coupled with new tactics, technologies, or operating concepts innovations will result. Innovations will enhance the Navy’s capabilities. However, they are not the goal. Instead, the goal is a repeatable learning system that enables the Navy to readily adapt and innovate along with advances in technology, changes in force structure, and shifts in the international environment.

That is what the Navy was able to do in the 1920s and 1930s as it explored carrier air power, experimented with rigid airships, and improved fleet tactics and doctrine. The system of learning developed in those years paid dividends in World War II when the Navy developed the combat information center (CIC), underway replenishment, and the tactics of multi-carrier task forces. Experimentation and rapid learning were essential to victory during the war.

I argue that it is time to apply some of the same underlying concepts, like open-ended exercises, nested feedback loops, and parallel experimentation, so that today’s Navy can enhance and capitalize on what former Navy Secretary Richard V. Spencer called its “most critical warfighting capability” – the skill of its officers and sailors.

Is War Getting More Complex?

In a recent Modern War Institute series on Leadership in Future War, Dr. Cole Livieratos discussed the “changing context of war” and emphasized the importance of recognizing its complexity. The article is quite good; however, Livieratos’s emphasis on “increasing complexity” raised some hackles.

Subtweets, like the one below, led to an active Twitter discussion about whether it is appropriate to suggest that war is becoming “increasingly complex” or if war is inherently complex and always has been. These discussions are valuable because they illustrate the nature of the challenge that Dr. Livieratos is trying to overcome.

It is fundamentally an ontological problem. How do we come to understand the nature of something—war in this case—and then describe it in terms that are valuable for our audience? Perhaps more importantly, if that audience has a specific perspective, how can we work with them to change it? That is what Livieratos is trying to do.

He grounds his discussion in Anthony King’s Command. King argues that Western armies responded to the trench warfare on World War I’s Western Front by adopting a “complicated” approach to warfare. They assumed that battle was reducible; that it could be decomposed into component pieces and centrally managed. Division-level headquarters evolved to perform the associated management activities. King’s analysis is a valuable perspective.

In Mastering the Art of Command, I used King’s work to help illuminate how two World War II Pacific theaters—Admiral Chester Nimitz’s Pacific Ocean Areas and General Douglas MacArthur’s Southwest Pacific—approached command, control, and jointness differently. MacArthur’s GHQ reflected the U.S. Army’s view that war was “complicated.” It emphasized centralized management and delegated execution. Jointness resided at the GHQ level. Nimitz’s structure was different and reflected the U.S. Navy’s long-standing emphasis on independent action. Because he wanted his subordinates to be able to self-organize and respond quickly to emerging circumstances, Nimitz pushed jointness to low levels—even down to individual islands—and fostered decentralization. Nimitz’s approach assumed war in the Pacific was “complex.” I argue that Nimitz’s approach was better suited to the nature of war in the Pacific.

Like Livieratos, I think the Cynefin Framework is a useful frame to explore these concepts. It presents the essential difference between “complicated” and “complex” this way: in a “complicated” situation, the relationship between cause and effect can be known and accurately predicted; in a “complex” situation, the relationship between cause and effect only becomes clear in hindsight. This is an important distinction because the best way to approach a problem differs depending on circumstances. With a “complicated” problem, methodical analysis can identify the best course of action. With a “complex” problem, analysis is wasteful, especially if time is a crucial factor. Instead, it is best to experiment using multiple parallel probes.

I would argue that the core of Livieratos’s argument is not that war is becoming “increasingly complex.” Instead, it’s that war is “complex.” The problem is that his target audience has grown accustomed to using the term “complexity” inappropriately. Regardless of how often it appears in manuals and associated literature, there is a deeply embedded assumption within the U.S. Army—and the U.S. Defense establishment more broadly—that war is reducable, that crucial variables can be identified with sufficient up-front analysis, and that cause and effect can be predicted. The prevailing ontological frame is that war is “complicated” even if it is regularly described as “complex.” This leads to the belief that it can be effectively decomposed and centrally managed.

Because of that, Livieratos’s choice to emphasize the “increasingly complex” nature of war is, I think, a reasonable approach to reaching his audience and encouraging them to revisit their assumptions. It is a good way to smuggle new ideas in and challenge prevailing assumptions. However, I am not certain that this framing is strong enough. “Increasing complexity” suggests that there is a spectrum and that moving further along it is sufficient—if there are more variables, then all we need is greater analytical capacity to manage them.

However, what is really needed is an ontological shift. A phase transition is a better analogy than a spectrum. If we liken the Army’s current concept of war to water, it’s not enough just to heat it up; it must be transformed into steam, a new substance with different properties and greater potential. Army officers must recognize that many of their governing assumptions about the nature of war are flawed. War is not reducible; it cannot be centrally managed; and officers must continually encourage approaches that allow for coherent, decentralized, independent action with minimal guidance and instruction—in Cynefin terms, multiple parallel probes. Fortunately, there are numerous valuable examples from the nation’s history to draw from and build upon. Nimitz’s command organization is just one.