On December 9, the United States Institute of Peace and Foreign Policy magazine conducted a “PeaceGame” on the conflict in Syria. Most of the day’s discussions were livestreamed by USIP, and those not in attendance could also follow along via the #PeaceGame hashtag on Twitter.
Ellie Bartels (National Defence University) and I were among those following along from afar, and much like Statler and Waldorf of the Muppet Show, we couldn’t resist offering some comments along the way. Ellie has kindly contributed some more detailed thoughts on the methodology of the “game” for PAXsims below, and also included some recommendations for future iterations. I’ve appended some of my own thoughts at the end.
In addition to the recorded video which will be posted to the USIP website, you’ll also find a summary of the day’s events at Storify, as well as an op ed on Syria and the PeaceGame by PJ Crowley on BBC News.
Yesterday, USIP and Foreign Policy put together the first of what promises to be an important series of PeaceGames, which seek to “bring together the leading minds in national security policy, international affairs, academia, business, and media to “game” out how we can achieve peace in Syria. USIP and FP intend for the game to redefine how leaders think about conflict resolution and the possibility of peace.” The game produced a lively conversation, with some very interesting findings about the potential paths to peace in Syria. However, this post is looks at the event from a slightly different angle: how the first PeaceGame worked as a game, and what may be the analytical consequences of the design choices that were made.
But first a few quick caveats: I was in no way involved with the design or execution of this exercise, and thus some of my analysis is based on assumptions about the intent of the team that put the event together.
Furthermore, I was not at USIP in person, so my observations are limited to the livestream feed of the event, as well as Twitter conversations I had with other observers and participants. Thus, my observations are based on limited data – I couldn’t see side conversations or non-verbal responses of players off screen, or eavesdrop on conversations during the break. I tried to supplement this information by asking questions of the participants who were active on Twitter during the game.
Last, but perhaps most importantly, I am a professional gamer who is paid to run this type of event. I have the job I do because I believe that gaming can be a rigorous, insightful tool to create shared understandings within or between communities, and that creating this type of understanding on foreign policy issues can lead to better national security policy. Thus, this post isn’t a critique of the use of games to address “serious” problems (which some observers did raise during the event) — it is a response aimed at promoting good design and analysis in this and future public games.
Overview of the first Syrian PeaceGame
The stated objective of the game was to lay out the conditions for a “best possible peace” to end the Syrian conflict. 45 experts, ranging from retired ambassadors to academics to Arab activists, role-played 19 different actors selected for their influence on the war. These also ranged widely from nation states, to sub-national actors within Syria, to the global media. All teams sat around a single table for discussion, which was directed by a moderator.
Because of impending weather in DC, the event was shortened from a planned four sessions of discussion, billed as “Achieving a Near-Term Political Solution,” “Establishing the Peace,” “Challenges to Peace Emerge,” and “Establishing a Sustainable Peace” to only cover the first, second, and fourth topics. Each discussion session started with several scene-setter slides on the current state on the ground in Syria, followed by anonymous voting on the importance of different issues, which was then used by the moderator as the jumping-off point for discussion.
The first session’s discussion was dominated by views on the potential of various processes for negotiating a peace. The majority view by the end of the session was that for negotiated peace to occur, key actors (most participants identified a combination of Russia, Iran and the Alawites) would need to sign off for the deal, or an alternative process would be needed. There was also a tension identified between outside stakeholders who were capable of reaching a negotiated deal, and internal actors who likely could not.
The second session, scheduled to be about establishing the peace, instead returned to the issue of what a peace deal might look like in Syria. However, in contrast to the focus on negotiations in the first session, here conversation focused on what the final deal would look like, with a limited discussion of implementation. In general, those with a strongly nation-state view tended to see a powersharing government or division along sectarian lines as the only viable options. In contrast, participants focused on non-state actors suggested that informal governance and local dealmaking could be empowered to form a peace with characteristics of both of the state-centric models.
The third session focused on what each group would need to see for peace to work, as well as last words from the participants. Again, a strong undercurrent in the discussion was the split between participants that focused on the role of non-state actors, and those who tended to take the perspective of nation-states. Non-state actors could often offer specific things they wanted to see prior to peace; those who saw states as the dominant actors were often reduced to talking about how conditions weren’t “ripe” for peace. There was also more of a focus on this round about the needs of Syria’s neighbors to avoid conflict spillover.
Notes on Design
Based on my observations, I think there are four areas where design choices had substantial effect on what we can and cannot conclude from the first PeaceGame. In the rest of this post, I describe the choices made about the audience, objectives, format, and moderations of the exercise, and discuss the potential impacts. I offer recommendations both for mitigating the impact of these choices on game analysis, and suggest alternative choices that may be worth exploring in future PeaceGame exercises.
Audience. The first thing worth noting is that PeaceGame was very public event, with a very high level of quality in participants. I (like many other analysts) was able to follow along with the event on a very well done live stream, with an active Twitter conversation running in virtual space (after the first session, the moderator mentioned that there had already been 1300 tweets using the event hashtag, which is more than many conferences ever get). Participants were well-established experts, with deep knowledge of the subject. That is not an easy combination to pull off, particularly when talking about a topic as politically sensitive as Syria, and I applaud the PeaceGame team for building a space safe enough for so many to participate. However, it is also reasonable to assume that had the same conversation been held in a private space with only a few select analysts listening in, discussion may well have included facts and opinions that were not included in the public forum.
Recommendation. Public game findings can be supplemented with more private, post-game interviews to allow participants to raise important issues that may not have been discussed in the PeaceGame forum.
Objectives. Unsurprisingly for the flagship event of a new program, PeaceGame had an ambitious set of objectives. In addition to the frequently repeated tasking to consider what a “best possible peace” would be, the program’s website added broader ideas like “redefine how leaders think about conflict resolution and the possibility of peace.” These broad objectives translated into a very broad conversation that sometimes sacrificed discussion on specific issues like spoiler, veto points, and second-order effects in order to cover wide ground.
Recommendation. Analysts should understand that the broad and sometime shallow conversations of the Syria PeaceGame were caused by the broad objectives, and should be seen as a feature and not a bug of the exercise. However, future games might consider laying out narrower objectives in order to guide deeper discussion of some of the issues of interest identified in this game.
Format. The most consequential choice the game designers made was to have the entire game take place at a single table with all 19 teams. While this choice is logical given the requirements of a public event, it also meant that any discussion of small negotiating teams, partnerships, and side deals could only be discussed “out in the open.” This seemed to make discussion about process somewhat more hypothetical in tone, because there was little ability to explore through the game what the dynamics of efficacy of such a tactic might be.
Recommendations. In future PeaceGames, consider allowing opportunities for smaller breakout sessions in order to play out specific combinations of actors at negotiations. This seems particularly helpful when studying issues like back channel negotiations that are often not legible to observers, since the game will allow for clearer observation than is likely to be possible for analysis in real life.
The lack of defined operating space in the game outside the large table also made for some interesting dynamics involving the violent non-state actors, who are relatively unlikely to be included the type of formal negotiations that the game format mirrors. Non-state actor players often asserted (either in their speaking time or on Twitter) that they were focusing on campaigns on the ground that made the negotiations irrelevant, going so far as to declare that they were “winning” the game. However, the format of PeaceGame did not have a clear way to connect these “independent” game actions with the broader discussion in a consistent manner.
Recommendations. Future games with more emphasis on implementation may benefit from a mechanism to incorporate unilateral actions, particularly by potential spoilers and veto points. This likely will require a more formal adjudication process than was used in the PeaceGame to reconcile different types of events into a narrative about the outcomes of player decisions.
Finally, despite the moderators’ efforts, participants often dropped out of role to speak in their professional capacity, particularly during the third session. This added good information to the discussion, but likely limited the game’s ability to reproduce emotional or psychological dimensions of peacemaking that might have been explored with stronger role playing.
Recommendations. Game analysts might consider impact of level of role play in different portions of the discussion as an important variable driving the tone and content of the discussion.
Moderation. FP Group CEO and Editor David Rothkopf had the superhuman job of corralling 45 engaged experts to keep them on topic, on time, and decorous. Having experienced trying to get this big a group to just not talk over top of one another, I was very impressed by his ability to hold and direct the group.
However, Mr. Rothkopt’s moderation style included many injections of his own opinions, which is generally frowned upon in the gaming community. Many gamers would argue that the best practice is for the moderator in a game to facilitate the participants, by ensuring everyone is able to move through the event in accordance with the rules. Our concern is that if you as the moderator are seen to be driving the discussion based on your understanding of what the content should be, you exclude participants whose views might differ, stifling the discussion.
In this case, some Twitter commenters noted that Mr. Rothkopt seemed to “have his own agenda,”even noting that he seemed dismissive of some of the non-state centric solutions which participants proposed. If participants felt the same way, this tendency may have biased which ideas were considered and even who spoke, and how much.
Recommendation. Analysis of the game may want to look at the role of the moderator and consider how the moderator’s contributions may have changed the discussion.
I very much enjoyed listened to the USIP/FP Syria “PeaceGame.” There was a remarkable degree of talent in the room for the day, and as expected their contributions were informed and insightful. Unfortunately, because of the format and large number of participants, many in attendance only managed to speak only a very limited number of times. Although a few offered comments via Twitter, I often found myself wishing for more of their input and analysis.
Recommendation. Find a mechanism whereby participants are able to express their views other than via the main microphone. Twitter is an imperfect mechanism for this, since it forces complex issues to be addressed in 140 characters or less. Instead, participants might be permitted to liveblog on a central website (or provide comments for support staff to post on their behalf), or given a period in which they summarize thoughts and recommendations mid-game for public distribution.
For the most part, the Peacegame wasn’t a “game” at all. Participants didn’t get a chance to “do” anything, and the role-playing (which, as Ellie notes, was only partial) was used more as a device to facilitate discussion than as part of some iterative, strategic process. Given that the purpose of the exercise was to explore aspects of the problem space in a way that caught the attention of the media, policy-makers, and the broader public, that’s fine by me.
That being said, introducing more game or “game-ish” elements could have been useful. The different phases of discussion might have been introduced with detailed, updated scenarios in the style of a classic three-move seminar game. Without these, some of the later discussion on stabilization and reconstruction foundered because it wasn’t entirely clear what would be the contextual environment for such activities. Also, introducing events as injects would have been useful, whether to provide atmosphere (and a sense of crisis), or to nudge participants to address key aspects of the problem that appear to have been overlooked or bypassed in the discussion.
Recommendation. The “PeaceGame” doesn’t need to operate like a wargame or crisis simulation. However, there are some game mechanisms that might enhance the process (many of them also identified by Ellie above).
While the “PeaceGame” angle did help generate buzz for the event, it did have one unfortunate side-effect: it was clear from the Twitter discussion that some Syrians and others felt that the participants were simply “playing” while real Syrians died. The occasional levity or gallows humour from the participants contributed to that alienation.
This is hard to address. Even when discussing deadly serious issues like the Syrian civil war (with its 125,000 dead), humour can make a conference run more smoothly and productively for those engaged in the meeting. Indeed it may be especially helpful at times when handling a topic that, after all, is rather dark and depressing. However, it can look rather different for those not in the room, or who do not usually participate in such meetings.
Recommendation. “PeaceGame” may not be the best title for events like these (something the the “Syria PeaceSimulation” might have been less problematic). Moderators need to recognize that the external audience may need to be reassured that human tragedy is being addressed with an appropriate degree of seriousness.
Finally, allow me to nominate “Islamic Extremist” Mona Yacoubian as the role-player of the day. She was excellent at showing how her actor(s) would manipulate the political and military situation to best advantage, whether by spoiling peace initiatives or exploiting local war economies to enhance their resources.
Recommendation. Don’t ever let Mona get real control of a jihadist group. Someone may also want to consider placing her on a No Fly List.