Problems and Prospects of Defense COIN Wargaming

In the Winter 2007 issue of the hobby wargaming journal Fire & Movement, the editor, Jon Compton, relates his experiences in playing counterinsurgency games at the Military Operations Research Society (MORS) conference in Monterrey, California. His last comments are worth circulating in this group and expanding upon:

All in all, it was an enlightening experience, and it was fascinating to get a glimpse of what professional wargame developers are doing in the military. Although board games are highly respected in this group, they are not taken seriously as modeling tools. To some extent I found that disappointing in that there is, in my opinion, too much dependence upon computer based agent interaction and stochastic processes, and too little upon the actual human interaction, which is where board games excel. The other problem I see is the black box issue related to computer-based simulations. This issue became very apparent as I quizzed the developers of the wargame we participated in and discovered that many of the governing assumptions were not based upon any sort of empirical or theoretical structure, but were simply invented out of whole cloth. This is information you would not know by playing the game, whereas with board games the system is open to examination and critical evaluation.

The Center For Naval Analyses (CNA) - certainly no stranger to MORS or to those well-read in counterinsurgency studies - published a very interesting monograph in September 2006 on the possibilities of wargaming such situations in board wargame formats, most notably using Card Driven Game (CDG) method pioneered by Mark Herman (currently at Booz, Allen and Hamilton) in his commercial hobby wargames. Entitled Wargaming Fourth Generation Warfare, authors Peter P. Perla, Albert A. Nofi, and Michael C. Markowitz would seem to solve some of Compton's complaints - if only commercial game designers could be taken seriously:

Our process of design then begins with identifying each player's worldviews and purposes. This leads to an assessment of the actions and means at their disposal, along with any constraints that may apply. This is perhaps the most difficult element of the design process to explain or envision. The designers must overcome the challenge of designing, in effect, multiple games, and then tying them together into a single coherent system.

We chose the currently popular mental models of DIME (Diplomatic, Information, Military, Economic) and PMESII ( Political, Military, Economic, Social, Infrastructure, Information) as the structural underpinnings for our assessments, as well as for many of the fundamental game-design parameters. Key among these parameters is the structure of the basic cards that drive the game's play. Use of a well-defined structure helps keep the design of the basic cards focused on critical elements, and helps us knit together the cards into a coherent fabric of play. What's more, this structural framework for the cards allows us to specify a process through which the players of the game might themselves define cards according to their own creativity and insights into the processes that the game proposes to investigate.

Such a design benefits from the strengths of a rigid-kriegspiel system, in which careful research underlies most assessments of actions and outcomes. At the same time, it opens the game to free-kriegspiel-style flexibility by allowing the players (and Control, for that matter) to create and invoke new ideas, but within a strong but flexible framework of game mechanics. The game we envision is largely player driven and action-centric, unlike games whose tendencies toward a Control-centric approach are, at times, regrettable....

Many of the techniques we have sampled or envisioned are similar to or adapted from techniques employed by commercial boardgame designers of card-driven wargames. As a result, we are confident that the approach we espouse can achieve many, if not all, the goals we set for it. What is most important at this early stage is that we know, in fact, that solutions to most - and hopefully all - the basic problems of wargaming 4GW exist. We have seen them in commercial games and in Naval War College games.

The investigations we have conducted and the ideas we have proposed have only started the ball rolling. All it will take for future game designers to develop effective new techniques - grounded in proven methods - to wargame Fourth-Generation Warfare much more successfully than in the past is thorough research, careful design, and expert execution.

Why, given the ready ability for commercial wargame designers such as Herman, Perla, Nofi, and others who comfortably sit within the Department of Defense operational analysis community, do we seem to nevertheless suffer from the problems that Compton notes above?

Over my nearly 27 years of active duty service as a Marine officer and as a hobby wargamer, I have sympathesized with thoughtful observers like Compton and wondered why it seemed so difficult to integrate commercial methods into "serious games" held within DoD. Below are some opinions on the issue:

Deus Ex Machina Syndrome: It's too easy to put blind faith in those "black box" simulations and believe in the Wizard of Oz. We do this to ourselves - I've been to a number of technology demonstrations in the modeling and simulation community. These demos tend to be long on "eye candy" graphics and short on trade show personnel who had the technical knowledge to talk intelligently about the variables and algorithms behind the design. Our oracles today are computer simulations and games and we tend to trust them too much. Conversely, board game systems are usually endlessly debated and tinkered with by game designers and players who have their particular take on the variables and adjudication procedures. This makes people who like to get "the answer" (and usually only ONE answer will do) uncomfortable.

Suspicion Surrounding Entertainment Games (and their Designers): Philip Sabin of King's College in London described this best in his excellent 2007 book, Lost Battles: Reconstructing the Great Clashes of the Ancient World, in the introduction:

People have been refighting ancient battles for decades using counters on a map or miniature figures on a tabletop, but it is only recently that the activity has acquired a higher public profile through the BBC television series Time Commanders, based on the 2004 computer game Rome: Total War. The popularity of wargaming as a leisure activity brings a certain stigma that has hitherto deterred its employment by academics (even those who are themselves wargamers in their spare time), but the technique is actually of much wider application....

Many existing battle simulations produced for the popular market are compromised by inadequate research and historical documentation and by the sacrifice of entertainment value--in Rome: Total War, for instance, battles last just a few minutes but involve enormous mutual casualties rather than the one-sided losses attensted by the sources.

These computer games aimed at the popular market engender suspicion quite naturally among those pursuing serious simulation to solve real problems. Yet even Sabin will acknowledge that within the small niche market of "historical enthusiasts," there are games of such depth and thoughtfulness that he claims "put many books to shame." The problem is that these kinds of games are not widely advertised or available and their players tend to keep to themselves. Many of these games are not very accessible, comprising of what seems to the uninitiated to be thick and opaque rulebooks, many pieces, and very slow playing time. Computer games are generally faster and easier to immerse oneself into.

Imprecise Metrics of Gain Versus Investment: The last problem is that the game experience is difficult to quantify, particularly when the "operating system" is laid open for examination and debate such as boardgames provide. Game sponsors want to know what benefits accrue from playing these "serious games," particularly when they see distinguished analysts, civilian senior leaders, and military officers rolling dice and moving tokens on a paper map. Actually, the problem occurs whether computer "black box" games or manual games are used; the best games focus on human interaction between players and insights gained from that experience. Participants walk away from the experience with more questions, better questions, and better ideas of where to look for the answers than they did before the game. But how can one put a price tag on this?

Commercial wargame designers have been sinking their teeth into the counterinsurgency gaming problem for some time and we can expect the fruits of their labor to finally see publication soon. The question Jon Compton implies still remains - will Department of Defense "serious games" eventually incorporate their design features or not?

0
Your rating: None

Comments

Thank you for the great blog post, Eric. For those who attended that MORS conference, I wanted to clarify that I participated in the Stabilization Operations work group, and played the PSOM simulation. I found many of the concepts in PSOM to be quite good, but I did quiz the developers extensively about many of the assumptions that were behind the resolution engine and discovered that there were many gaps in theory and research (which I'm sure will be filled as the project progresses). My main point about the black-box comment was that these gaps would not be apparent by playing the game. My biggest worry is that policy makers would not think to ask given any particular result. Governing assumptions are crucial to resolution engines, and these assumptions are usually well hidden in computer games. A board game, as a modeling tool, does not have this problem.

Thank you for the great blog post, Eric. For those who attended that MORS conference, I wanted to clarify that I participated in the Stabilization Operations work group, and played the PSOM simulation. I found many of the concepts in PSOM to be quite good, but I did quiz the developers extensively about many of the assumptions that were behind the resolution engine and discovered that there were many gaps in theory and research (which I'm sure will be filled as the project progresses). My main point about the black-box comment was that these gaps would not be apparent by playing the game. My biggest worry is that policy makers would not think to ask given any particular result. Governing assumptions are crucial to resolution engines, and these assumptions are usually well hidden in computer games. A board game, as a modeling tool, does not have this problem.

Thanks for this post.

Unfortunately, I haven't found the F&M article referenced online, so I can't respond directly to that. I was however, at two MORS-sponsored wargaming seminars in September and December of last year, the second of which was at Monterey.

I obviously can't speak for everyone there but I can tell you we used commercial wargames as the basis for our analysis. I was a controller for a wargaming project called "Algernon," in which we used the rule set for "Algeria: War for Independence" (not certain on the exact title of the game) by Brian Train (who was also in attendance). The primary metric was popular support level. The game is highly kinetic, so we missed a lot of the non-military aspects of COIN operations. However, the game was useful in getting people top think about the problem.

I think we generally received positive feedback from the December MORS attendees in Monterey.

Thanks for a fascinating entry, Eric! I've always been interested in the seeming disconnect between outside game designs and those used by DoD. There's a big difference between the "Command & Conquer" style of computer games and the much more detailed and in-depth boardgames and some of the more advanced computer models. I've always been a fan of a MUD/MUSH style framework for a COIN simulation...something that has a control cell and at the same time allows for direct (or reasonably direct) interaction between a number of players. The interaction's the thing, and this can only be improved with clear, concise rules (although the design pendulum seems to be swinging back into the complex and esoteric side).

Thanks again for a good entry and link to an interesting CNA study!