A recent post on Wired’s “Danger Room” blog discussed DARPA attempts to use agent-based computer modeling to better understand how social beliefs are translated into social actions.  The post was maddening for a number of reasons.  First, the author of the post cast the whole thing as Vietnam-era, social science at war take II.  Of the use of social science in Vietnam, all the author has to say is, “It’s a complicated history, but it’s safe to say that these programs did not have a happy ending.”  Yes, indeed, Ms. Weinberger, they did have a complicated history, something that I just wrote two dissertation chapters about.  But that history is not important here.  The mention of Vietnam is meant only to bias the reader from the beginning.  No “complicated history” needed to do that.  (While maddening, I’ll leave this issue of social science in the military, WWII through Vietnam, for another time.)

Even more maddening, however, is the response from retired Lt. Gen. Paul Van Riper.  According to “Danger Room,”

Critics of this work, however, say that human interactions are way too complicated to predict and relying on models is delusional.

“They are smoking something they shouldn’t be,” retired Lieutenant General Paul Van Riper told Science Magazine recently about such work. “Only those who don’t know how the real world works will be suckers for this stuff.”

And only those who use the lingo of complexity theory but who do not know the tools and techniques of complexity theory would argue that using agent-based modeling makes one delusional.  Agent-based modeling is a key technique used by complexity theorists!  What does all this have to do with Van Riper, you ask?  Let me explain.

Van Riper was/is a follower of the teachings of the late USAF Col. John Boyd.  Boyd drew on emerging sciences, especially the “nonlinear sciences” of chaos and complexity, in the formulation of his ideas.  Along the way, he generated a lot of interest in these sciences within the U.S. military community.  Many learned about these sciences from Boyd directly, or read popular press science books about them which Boyd recommended.  Van Riper was one of those folks.  In fact, in testimony before Congress on 20 March 1997, Van Riper referenced nonlinear science explicitly on a number of occasions, and often used terminology associated with nonlinear science.  For example, at one point he said, “In fact, all the trends in modern science, evolutionary biology, nonlinear mathematics, and quantum physics underline that Clausewitz’s fundamental belief that we do not live in a predictable universe was right on target.”  In explaining a number of initiatives being taken by the Marine Corps to prepare for war in the 21st century, he said, “Other of our initiatives are literally on the ‘edge of chaos,’ involving the emerging nonlinear sciences such as chaos and complexity. These ‘new sciences’ are the object of research at Quantico and are also being introduced into the curriculums [sic] of our schools.” [1]  But what did that “research” into the “new sciences” involve?  Apparently not agent-based modeling, one of the main techniques of complexity theory.  More likely, the “research” involved reading popular press accounts of the “new sciences.”  The last time I checked, that’s not considered “scientific research.”

Van Riper has been a harsh critic of the notion of network-centric warfare as articulated by the late Adm. Arthur Cebrowski.  But, as I have written before, Cebrowski also cited Boyd and also based much of his thinking upon nonlinear science and evolutionary biology as he received it through the writings of MIT professor Eric D. Beinhocker.

What does all of this mean?  It means that it’s time to lay to rest the idea that Boyd’s followers are “in the know” about the “new sciences” while everyone else is ignorant of them (intentionally or unintentionally).  Both Boyd and his followers’ ideas, as well as the ideas of the folks they oppose (NCW, Cebrowski, etc.), have been heavily influenced by concepts and language taken from the “new sciences” of chaos and complexity.  Continuing to argue that one side “gets it” while the other does not is not helpful.  The fact that Van Riper (one who is supposedly in the group of folks who gets it) does not seem to realize that agent-based modeling is a key technique of complexity theory, should raise questions about who really does “get” complexity theory among those who have attempted to apply its concepts and language to various aspects of warfare.  It has become increasingly clear to me throughout my dissertation research that there are those within the defense community who are “really” doing military-related complexity research (those doing the math, running the computer models and simulations, collecting and interpreting data, etc.), and then there are those who use the language and concepts of complexity, which they have gleaned not from doing scientific research, but by reading books like Waldrop’s Complexity or Gleick’s Chaos.  Often, this results in what amounts to a form of military romanticism with a veneer of scientific lingo pasted on top.  When confronted with actual, military-related complexity research, they often don’t recognize it as such, reject it, or both.

Whichever theory we agree with (NCW, 4GW, etc.), we need to start asking how it is that such supposedly universal concepts derived from the “hard sciences” can lead to such vastly different viewpoints.  If complexity theory can be used to develop such different theories, how useful is it really for the development of military theory?  More generally, what are the limits of science as metaphor when applied to military theory?  Should military professionals have to understand the processes, tools, and techniques of a particular science, not just popularized versions of its concepts, before they apply it to the development of military theory?

These are just a few of the questions and issues that need to be addressed if concepts from emerging sciences are going to be used effectively by military professionals.

[1] Prepared Statement of Lieutenant General Paul K. Van Riper, United States Marine Corps, Commanding General Marine Corps Combat Development Command, Before the House National Security Committee, Procurement Committee, and Research and Development Subcommittee (20 March 1997), available from Lexis-Nexis.

Technorati Tags: , , , , , , ,