Some of the wargaming team at HMS Victory (left to right: Paul Strong, Mike Bagwell, Colin Marston, me, James Bennett, Mike Young and Major Tom Mouat).
Public release identifier DSTL/PUB103767
In mid-July I was fortunate to spend the better part of a week with the (UK) Defence Science and Technology Laboratory wargaming team at Dstl Portsdown West, near Portsmouth. As with my visit last year, I had a very enjoyable and productive time, exchanging views, discussing challenges and approaches, and generally benefitting from their broad experience. The schedule and pdfs of all my slides are provided below. (The videos have no sound, and are just another way to present the slides.)
On the first day of my visit I made four presentations, each of which was followed by broader discussion with the group in attendance. The first of these examined wargaming as an educational (vs analytical) tool [pdf]. In this I discussed the strengths—and potential weaknesses–of serious gaming as an educational and training tool. I emphasized that educational outcomes depended not just on a game’s design, but also how it was used, and how it related to course objectives—the debate over the Statecraft international relations simulation being a case in point. I highlighted four general types of educational games, which I termed “pathway,” “strategy, “perspective,” and ‘fog and friction” games. I noted how the design of these differed from analytical games intended to answer one or more research questions. However, while games should certainly be designed for their intended purpose, I also suggested that practical realities (including limited resources) meant that it might sometimes be necessary or desirable to conduct dual-purpose games that have both analytical and educational dimensions. Much of the rest of the discussion focused on how best to do this without adversely compromising either aspect.
My second presentation [pdf] looked at wargaming unpredictable adversaries (and unreliable allies)—a topic of growing importance given the challenges of global terrorism, North Korean missile and nuclear weapon development, the current crisis in the Gulf Cooperation Council, and similar challenges, coupled with the complications posed by uncertainty in US policy under the Trump Administration, plus Brexit in the European context. Much of my talk drew upon ideas I first raised in an article on the topic at The Strategy Bridge in March. I discussed several possible approaches to representing unpredictability/unreliability in a game, including scripting, stochastic (random) behaviour, responsive variables (with stochastic elements), using the white cell, and two (or more) level games.
After that, attention turned to gaming the semi-cooperative [pdf]. Here we explored the challenge of designing games that are cooperative but retain plenty of room for competition, poor coordination, and friction. This can be done, of course, with game mechanics that offer tangible rewards for both sets of behaviours, such as the dual metric system of (cooperative) “relief points” and (individual) “organization points” in AFTERSHOCK: A Humanitarian Crisis Game. However, I suggested, there are limits to such a utility-maximizing, game theoretical treatment. Designers should also emphasize a psychological approach, whereby player engagement with the game narrative, imperfect information and communication, time pressures, game facilitation, and other methods are used to internalize sources of potential disagreement among the various participants.
The final presentation for the first day explored wargames as experiments [pdf ]. Here I suggested that wargames were rarely proper experiments: idiosyncratic variations between players, the limited number of iterations possible (often only one), and the complex, highly contingent nature of outcomes, all weighed against true experimentation. Nonetheless, some quasi-experimental designs were possible, and wargaming was extremely useful both as a way to generate questions for further study and in the context of efforts to triangulate findings using mixed methods. I pointed to a few techniques that might be used.
I also suggested that we needed more research on wargaming methodologies. One possible way of encouraging this would be to have an annual game design challenge, wherein wargamers would be invited to submit wargames exploring a set topic. This would allow methods and outcomes to be compared. (As Paul Strong pointed out to me, this is something the Society of Ancients already does in the hobby arena, annually refighting an ancient battle using a number of different rule sets.) A Dstl challenge for Connections UK 2018, perhaps?
Tuesday: Matrix Gaming and MaGCK
Most of the next two and a half days were taken up with a workshop on basic and advanced matrix game techniques [pdf], including an overview of the Matrix Game Construction Kit (MaGCK) prototype. For this I was joined by Major Tom Mouat (Defence Academy of the UK), a fellow member of the MaGCK design team. We both found it very useful to get some feedback on the kit and its contents. Most of suggestions we received will be incorporated into the final production version, which will be launched at the Connections UK professional wargaming conference in September.
Day 2 also featured a presentation on gaming foreign policy [pdf]. This examined the value of serious gaming for training and policy analysis, and reviewed some of the work colleagues and I had done over the years gaming various aspects of conflict and peacebuilding in the Middle East.
At the end of the day we sat down to play A Reckoning of Vultures, one of the sample matrix games included in MaGCK:
A Reckoning of Vultures is set in the capital of the fictional Republic of Matrixia.
There, in the ornate Presidential Palace, surrounded by his most loyal Presidential Guards, the President-for-Life is on his death-bed—and various power-hungry factions are jostling to take power themselves.
Once the President passes, competition between these would-be successors will escalate to open conflict, until the Central Committee of the Ruling Party can meet and agree on a new leader…
The Central Security and Intelligence Directorate (CSID) are Matrixia’s shadowy—and much-feared—secret police, responsible for maintaining a close watch on both dissidents and potential rival power centres within the regime. Although lacking large numbers of armed personnel, covert CSID operatives are well-placed to blackmail, influence, sabotage, subvert, or spy.
The Matrixian Armed Forces can call upon large numbers of military personnel located in three major military bases around the capital. Inter-service rivalries and the influence of other factions may mean, however, that not all MAF units are loyal or obey orders.
The Ministry of the Interior has authority over police and emergency services personnel in the capital. Although MoI units are well-positioned across the city, most are inferior in combat capability to those of the regular military.
Much of what happens in Matrixia is manipulated by a group of rich and powerful Oligarchs, who both control much of the business sector and have deep ties to the country’s major criminal syndicates. Although they have only a few private security guards and mercenaries to safeguard their position, they have considerable wealth that can be used further their political ambitions.
The National Union of Toilers represents the downtrodden workers of the country. NUT hopes to mobilize the masses and advance their political agenda through strikes, demonstrations, and direct action. If they can arm some of their followers and form a workers’ militia, they could become very powerful indeed.
Things heat up in the capital of Matrixia (left to right: me, Colin Marston, Mike Larner and Major Tom Mouat).
In this particular case, our game involved a dead President (of course); student protests, which were soon crushed by hired thugs; an amphibious landing by MAF marines to wrest control of the port; a failed airborne landing at the worker-controlled oil refinery; a spectacularly unsuccessful jailbreak (in which unguarded prisoners preferred to stay in their cells than follow the revolutionary NUT leader); sabotage of the Ministry of Information communications system; bombing of the civilian airport by government jets; a dramatic face-off outside the Presidential Palace, in which tanks were vanquished by protesters (presumably through moral suasion rather than any sort of inherent anti-armour capability); and a closely fought vote for supremacy in the Central Committee of the Ruling Party. While it was all good fun, the scenario—as intended—demonstrated a variety of different matrix game techniques. Moreover, it was possible to relate most of the game events to real life coups and succession struggles in Syria, Egypt, Turkey, and elsewhere.
Wednesday: More matrix gaming, and a dockyard tour
For me, one of the most interesting part of the programme was the opportunity to develop a matrix wargame from scratch with members of the Dstl staff, as a way of exploring matrix gaming and game design more broadly. The choice of topic was left up to me, so as a Middle East analyst I chose a possible future conflict between Israel and Hizbullah. After an overview of key aspects of the issue [pdf], we all set to work for a couple of hours. The result was a game design with four actors (Israel, Hizbullah, the Lebanese government, and civilians fleeing the fighting). Reflecting the complex, multi-sided character of Lebanese politics, the Lebanese government randomly determined each turn whether its actions reflected a common national interest, or the interests of a particular political or sectarian group. The civilian player represented the interests of Israeli and Lebanese civilians alike, and their actions offered an interesting way to model the safety-seeking behavior of local populations in wartime.
During peacetime phase, Israel and Hizbullah would each take one action each per turn, while the Lebanese government and civilians could take one action total during the entire phase. Once major fighting started, Israel and Hizbullah received two actions per turn (one military, one non-military), while the Lebanese government and civilians received one each. The war would continue until a ceasefire was agreed to by the parties, or the domestic support level of one of the belligerents fell to zero.
After all this it was down to Portsmouth, where we went on a specially-arranged tour of HMS St. Albans, a Type-23 Royal Navy frigate. Members of the crew were very informative—especially the watch officer who showed us around, and the senior engineering rating who offered a detailed look around the engineering control room and engine room (my first opportunity to get up close and personal with a Rolls-Royce Marine Spey gas turbine). The Dstl team later presented me with a copy of the ship’s crest—a lovely gift, even more so because it had been signed by everyone. We also had some time to see HMS Victory at the Historic Dockyard. Appropriately enough, dinner that evening was at the officer’s mess at HMS Nelson (Her Majesty’s Naval Base, Portsmouth), where–in the best traditions of the empire–it was curry night.
Finally, I spent a couple of hours that night, putting together a playable version of our prototype Israel-Hizbullah game, writing up rules and player briefings and using components from MaGCK.
Thursday: Playtesting and game design
The morning of Day 4, we playtested the Israel-Hizbullah matrix game. The game featured two distinct phases. The first depicted growing tensions, with a major arms build-up by Hizbullah, Israeli bombing of one particularly significant weapons shipment through Syria, the successful Israeli assassination of a senior Hizbullah military commander, and an ominous border incident. A system of random event cards left players the option of initiating conflict (at a political cost), or waiting for events to make it inevitable.
Finally, the war came. Israel, which had already called up a substantial number of reservists for a planned military exercise, crossed the Lebanese border on a wide front, hoping to destroy most of the estimated 150,000 rockets Hizbullah had amassed in southern Lebanon. Going was slow, however, with Hizbullah forces making good use of the terrain, minefields, bunkers, ATGMs, and the combat experience it had gained in the 2006 war, the Syrian civil war, and elsewhere. However, most efforts by the Shiite militia to score a major propaganda victory—for example, by downing an IDF helicopter laden with troops—were largely unsuccessful.
Contemplating the situation in Lebanon (left to right: Lt Cdr James Winsor, Stephen Ho and Ben Short).
The Lebanese government pressed for a ceasefire, and was ultimately successful in seeing a draft resolution tabled at the United Nations Security Council with the support of Russia, China, and the European Union. The Trump Administration, however, was sympathetic to the Israeli operation, and vetoed the resolution to give the IDF more time to achieve its objectives.
Tom Mouat’s notes from the Israel-Hizbullah game.
There we had to end the playtest–Israel was narrowly ahead on points, but Shiite support for Hizbullah was high, and there was little evidence that the group had suffered a fundamental setback.
Later that day I made my final presentation for the week, on gaming corruption [pdf]. I differentiated between three levels at which corruption might be represented in a game: as a complicating factor largely beyond the control of players (represented by some limit or random event); as significant secondary dynamic that players could interact with and affect (as in Mission Zhobia); and finally as the primary focus of the game. In the case of the latter I drew upon the serious games that Tom Fisher has developed for the World Bank and Egmont Group on money-laundering and anti-corruption efforts.
Last but far from least, the final part of Thursday was spent in an extended discussion of possible design elements for a project that Dstl is currently working on. I can’t disclose the topic or participants, but can say that our discussion addressed a variety interesting issues regarding:
- in-game communication, including the constraints imposed by classified material
- using the red cell in a way that both offers red “the freedom to win,” yet assures that game stays on course for its analytical or experiential purpose
- employing subject matter experts (and keeping them sufficiently busy and engaged)
- determining the level of military fidelity necessary (and deciding what of this should be communicated to the players)
- the use and abuse of marker tracks and metrics
- generating narrative engagement and immersion
With that, my visit to Dstl came to an end. It was enormously valuable to me to have had the opportunity to share ideas and insights with such a talented group of wargamers and defence analysts—and in a casual setting conducive to frank discussion, innovation, more than a few cups of tea, and a great deal of fun. I’m very grateful to Colin Marston and the rest of the team for their hospitality, as well as their support for the MaGCK project.
For those of you who want to try A Reckoning of Vultures or the Israel-Hizbullah War 201? matrix games, we will be running both during the games fair at Connections UK in September.