I had occasion to do some reading on doctrine today, looking back over last year’s emails between LTG Van Riper and LTG Deptula. I found the full email chain in the InsideDefense archives (subscription required), which provided some thought-provoking material.

Then - in classic internet ADD fashion - while I was searching for one of Van Riper’s papers, I came across an interesting Air Force Magazine op-ed. Robert Dudney argues that the USAF’s latest version of Doctrine Document 2, Operations and Organization “shows that Air Force thinking about [the effects based operations concept] has deepened and matured in the past few years.”

While I haven’t read the AF document in question, Dudney’s piece sparked some connections:

Far from being inflexibly mechanistic, EBO doctrine sees war as “a clash of complex adaptive systems”—that is, “a collision of living forces that creatively adapt” to new situations. Planners must be flexible and seek to know how the enemy will respond to planned actions, and incorporate this information in operations.

In order to know how the enemy will respond to planned actions, planners will need to know how the complex adaptive system of the enemy will respond. This requires predicting the behavior of a complex system, a challenge upon which the science of complexity has made precious little progress. For a sub-set of complex systems, what Van Riper refers to as structurally complex systems, the reductionist approaches of systems analysis can model and predict behavior[1]. For these systems, such comprehensive knowledge is possible. Incidentally, it was the successful analysis of structurally complex systems such as electrical grids and air defense networks during Gulf War I that led to the development of EBO. More recent work by John Robb has discussed how non-state forces can use similar insights to bring down structurally complex elements of infrastructure.

The superset of all complex systems, however, includes trickier beasts: interactively complex systems [2]. These systems exhibit irreducible complexity, meaning that if you attempt to decompose them some system-wide behaviors become lost. In other words, it is exceptionally difficult to isolate the relationship between the macro-level behavior of the entire system and the micro-level behavior of individual elements. Analysis through decomposition simply doesn’t seem to work (in fact, some people who study complexity essentially define a complex system as one which is resistant to decomposition analysis). Enemy forces as a whole are interactively complex, and therefore I question the feasbility of using complexity science to predict their behavior.

Failing to accurately understand the complex system of the enemy has serious consequences, as Dudney points out:

A corollary, says the doctrine paper, is that success comes “at a price,” which is a need for “comprehensive knowledge” of the battle space and its actors. US knowledge must go “well beyond” just the enemy order of battle and include enemy thinking, influences, and tendencies. USAF notes that, in Vietnam, the US failed to detect the “implacability” of Hanoi’s leaders, with lamentable results.

If enemies are interactively complex systems, is such compreshensive knowledge generally attainable? Perhaps, but given the immaturity of complexity science, I expect that these insights will come from traditional fields such as anthropology, history and psychology. In the Vietnam example, I expect that we would all agree that LBJ, MacNamara et al could (and should) have correctly identified the motives and goals of Hanoi without complexity science. If complexity science can’t currently offer the necessary comprehensive knowledge, then what is the usefulness of looking at war as a complex adaptive system? In a paper for the Strategic Studies Institute, Van Riper argues that if new doctrine cannot demonstrate an improvement upon established doctrine, then we should wait until we can seperate the wheat from the chaff.

This brings up the question of EBO’s relationship to the doctrine established by the military reform movement [3] [4]. The Van Riper-Deptula email exchange provies an excellent overview of this discussion. One of the most concrete questions is the degree to which EBO modifies the established doctrine of mission-type orders. Deptula states in his email that EBO and mission-type orders are one and the same. Dudney states in his op-ed that EBO “produces standard mission-type orders.” In “Planning for and Applying Military Force: An Examination of Terms,” however, Van Riper argues that “the new joint planning definitions and concepts [including EBO] tend to confuse more than they inform.” So, what does EBO add that established doctine lacked?

I don’t have an answer yet. At the very least I need to do some more reading of COL John Warren and LTG Deptula’s writing on the subject. But it’s friday and the tank is just about empty (and I’ve exhausted my limited knowledge of doctrine). The old brain is tired enough that all I can think of when I read “…to achieve the desired effect…” is George Clinton reminding me that “the desire effect is what you get when you improve your interplanetary funksmanship!”

And on that note, have a good weekend all.

[1] A structurally complex system, in this context, is one that consists of many parts yet exhibits relatively simple behavior due to the limited interaction between those parts. Your car is an example. Thousands of parts, yet each one has a very limited degree of freedom. Furthermore, these parts interact in a carefully designed manner. So long as these parts function as designed, the entire system behaves in a consistent and predictable manner.

[2] An interactively complex system is constituted of elements that interact freely in relatively unconstrained ways. Each element has a high degree of freedom and can interact with other elements in a variety of manners. Social systems, such as economies, are interactively complex. Consumers, companies and government regulators are some of the elements of such a system and they each have many more degrees of freedom than the pistons or rear axel of a car. Furthermore, each of these elements can influence a host of other elements - to a much greater degree than each element of your car can.

[3] FMFM 1 Warfighting (pdf)

[4] FM 100-5 Operations (pdf)