Conflict simulation, peacebuilding, and development

Tag Archives: University of Exeter

Discussing political simulations and gaming at the University of Exeter

Today I spent an enjoyable afternoon discussing simulations and serious games in the classroom at the University of Exeter, in a workshop organized by Prof. Mick Dumper of the Department of Politics. Slide01 In my own talk I first situated the value from using simulations. Here my bottom line was that while simulations are not always more effective than more conventional teaching techniques, they offer an opportunity for “intellectual cross-training,” can be highly engaging, and are particularly good for exploring policy processes, coordination challenges, mixed and adversarial agendas, and decision-making with imperfect information. Slide02Slide03 I then discussed examples of several different approaches to using games and simulations:

  • quick and simple in class games/simulations
  • assigning commercial games os reading or review assignments
  • roleplay and negotiation simulations
  • “game shows” for large classes
  • matrix games
  • custom-designed boardgames
  • student-designed games
  • complex and hybrid games

I went on to offer some thoughts on “best practices” for simulation/game use in an educational setting: Slide23 Slide24 Slide25 Slide26Finally I said a few words about the use of simulations for research and policy analysis. You can find the full presentation (pdf) here. Next, myself and several Exeter faculty members (Duncan Russell, Amy McKay, Sandra Kroger, and Mick Dumper, who also use simulations to explore topics ranging from global climate change negotiations to the EU, lobbying the US Congress, and the Arab-Israeli conflict) held a broader discussion with the audience on the topic.

Finally, a group of us played a game of AFTERSHOCK: A Humanitarian Crisis Game. I’m happy to report that, after initial setbacks—including transportation bottlenecks, a major aftershock, flooding due to heavy rains, and a severe outbreak of cholera—everyone pulled together well and won the game.

Syrian refugees in Lebanon simulation — the video

Abigail Grace has produced an excellent video of January’s refugee simulation at the University of Exeter. You’ll find it below.

Simulating the Syrian refugee crisis in Lebanon


Last month I had the pleasure of running a classroom simulation on the Syrian refugee crisis in Lebanon at the University of Exeter with Prof. Mick Dumper for his POL 2046 course on The Refugee Crisis in the Modern World. Gamers extraordinaire Tom Mouat and Jim Wallman came down for the day to assist, along with graduate student Abigail Grace. Today I ran the same simulation at McGill University for some of the students in Prof. Megan Bradley‘s POLI 359 course on the international refugee regime, together with a few from my own POLI 450 course on peacebuilding. This time ICAMES graduate research fellow (and teaching assistant) Ecem Oskay was there to help.


The Exeter Control team, complete with empty sack of goat food.

Both simulations involved around two dozen students. Both simulations went very well, I thought.

A variety of roles were represented in the simulation:

  • The Lebanese Prime Minister, plus various cabinet ministers (from the Future Movement, Phalange, Hizbullah, Free Patriotic Movement, Progressive Socialist Party) and the Lebanese Armed Forces. This gave some differentiation in terms of portfolios and responsibilities, and also recreated some of the political and sectarian tensions between the “March 8” and “March 14” coalitions within the Lebanese government.
  • Various UN agencies (UNHCR, UNRWA, UNICEF, and World Food Programme)
  • A (fictional) local charitable association.
  • Human Rights Watch.
  • The European Union ambassador (representing the donor community more broadly).
  • The refugees themselves. Each of these had a different back story in terms of geographic origin, occupation/social class, family needs and situation, sectarian affiliation, and political views. One was a female-headed household.  Two of the refugees were secretly opposition organizers, for the Free Syrian Army and ISIS. Some were Palestinian refugees from Syria, rather than Syrian citizens.

The Lebanese cabinet makes a joint announcement, hoping to dampen down sectarian tensions. (Exeter)

In designing the simulation I wanted to avoid a simple seminar-type negotiation exercise in which the stakeholders all sit down around a table and try to achieve an agreement on something. For a start, such an approach wouldn’t generate the sense of overbearing crisis that Lebanon feels, a small country hosting some 1.2 million refugees from the bloody and dangerous civil war in neighbouring Syria. In addition, it would also misrepresent the dynamics whereby refugee policy emerges. Refugees do not, as a rule, play any sort of direct role in policy formulation. Instead, their actions and coping strategies provide the context.

Consequently, this simulation was really two linked simulations in one.


The Lebanese Armed Forces questions refugees, looking for evidence of militants and paramilitary activities. (McGill)

At one level, refugees were tasked with simply trying to survive. Each hour they would have to make choices about how to try to earn money (beg? work illegally in Lebanon? try to cross back into Syria?), where to live (a squatter camp? a squalid flat? a middle class apartment?), and what additional goods did they want to buy (basic durables? medicine? forged papers?) Choices had consequences–they might be arrested, deported, or shot crossing the border, or their children might get ill from poor accommodations.

The refugees sit in their make-shift shelters while aid workers undertake a needs assessment. (Exeter)

The refugees sit in their make-shift shelters while aid workers undertake a needs assessment. (Exeter)

The refugees were also given tarps, ropes, cardboard, old carpets, and other materials and were required to construct their own makeshift shelters in the classroom—which at one point were then torn down by angry Lebanese farmers seeking to reclaim their fields. They were required to undertake manual labour, representing the sort of unskilled jobs refugees typically take: in Exeter this consisted of endlessly moving furniture from one end of the classroom to the other and back again, while at McGill they had to carry heavy bags up and down four flights of steps. In their spare time they might beg, or protest, or even smuggle weapons.


Refugees build shelters near the United Nations compound as a member of the Control team looks on. (McGill)

Each hour a random event card would be drawn. Some of these were good: relatives in Europe might send money, or a refugee might reconnect with old friends. Many others were negative: agonizing moral choices, sexual assault, sickness. Refugee resiliency was tracked with tokens. If refugees ran out of these their coping skills were sharply diminished, or they were instructed to just sit and sob in their shelters until someone offered them some help. Throughout, all of the refugees kept handwritten diaries of their experiences.


The Lebanese Army arrests a refugee. (Exeter)

Everyone in the simulation was provided with lunch—except the refugees, who were expected to “buy” it with their meagre simulation income. Depending on their luck and decisions, some didn’t eat for hours, and others not at all. Refugees were also prohibited from sitting in chairs or accessing their telephones or laptops unless they “paid” to use these too. Their rooms were often plunged into darkness, unless they illegally connected to the Lebanese power grid. In Exeter we opened the windows on what was a cold and damp day to increase the refugee discomfort level (it was -18C in Montreal, which didn’t really make that a viable option).

This unfortunate refugee didn't make it—shot by Syrian border guards. (McGill)

This unfortunate refugee didn’t make it—shot by Syrian border guards. (McGill)

The aid actors had some resources (cash, food, other items), but not enough. The UN in particular had to register the refugees and undertake a needs assessment to make sure that the most vulnerable received priority.


Angry refugees protest their treatment. (McGill)

At another level, this was a more traditional policy simulation. The UN team was tasked with drawing up a comprehensive refugee strategy to which the Lebanese government might agree. The Lebanese government was concerned not only with this, but also with a number of other challenges that cropped up (a bomb attack, jihadist suspects hiding in a Palestinian refugee camp, complaints that the Syrians were pushing Lebanese workers out of jobs, crime, illegal electrical connections, a measles outbreak among the refugees—among others). The EU sought to promote a more effective response to the refugee crisis, and had some funds to support this. Human Rights Watch tried to raise human rights issues with Lebanese policymakers and the international community. The refugees were largely absent in any direct sense from these discussions and negotiations, although their choices or even protests fundamentally shaped the policy environment.

All of the policy actors were expected to take notes and minutes, and prepare formal presentations or reports that were submitted during the simulation.


Officials listen to a presentation by Human Rights Watch. (Exeter)

In both the Exeter and McGill simulation runs, the Lebanese grew increasingly concerned at the economic, political, and security challenges presented by the refugees. The UN proposed an integrated refugee strategy after several hours of consultation, but in both cases the Lebanese government rejected the proposal and called for further discussions.


The Lebanese cabinet poses for a photograph, shortly after rejecting UN proposals and calling for further discussions. (McGill)

In both simulations, despite significant local and international aid, the refugees felt they largely had to fend for themselves, and grew resentful that more wasn’t done to help them. In the debrief, many of the well-meaning internationals were rather surprised to hear this.

In the debrief session we were careful to identify the artificial aspects of the simulation—for example, more simulated than real refugees were involved in paramilitary skullduggery, and real refugees would be less likely to organize protests for fear of arrest or deportation. But there were also many, many realistic outcomes that we could point to and discuss. The refugees in particular got a sense of marginalization and vulnerability, but also how refugee communities could organize to help each other in sometimes small but important ways.

This was not a simple simulation—it was 6-7 hours of intense activity, involving a 3-4 person control team. However, those who participated seem to find it well worth the time spent.

Reflections on a humanitarian policy simulation


Earlier this year, my colleague Mick Dumper (University of Exeter) and I organized a policy simulation that explored potential near-term challenges to the United Nations Relief and Works Agency, the agency which deals with Palestinian refugees in the West Bank, Gaza, Jordan, Lebanon, and Syria. As a brief account on the University of Exeter website notes, the exercise was set one year in the future, in a Middle East that looks much like the current one, only worse:

UN Agency Learns from Simulation at Exeter

It is March 2014, Syria has imploded, refugees are massing at the borders, the Palestinian Authority is bankrupt and its security services have been unpaid for 3 months.  How should the UN agency responsible for 5 million Palestinian refugees (UNRWA) react when it itself is adjusting to US Congress-inspired funding cuts of 20%?

This scenario was played out in a simulation organised by the Department of Politics, with senior management of UNRWA, including its Commissioner-General and Director of External Relations, leading academics in the field, some donor representatives and diplomats.

The exercise involved three “Policy Teams” (one of which was composed of three College PhD students) who were set the task of coming up with clear recommendations for UNRWA by interviewing a number of stakeholders representing the EU, US, Jordan, Lebanon, HAMAS, the PA and the refugees.  Should it increase its advocacy work at the risk of alienating its main donors?  Should it cut services and where? What would the impact of that be on the host countries?

“The fact that the scenario was uncomfortably close to reality gave real impetus to the task and also weight behind the recommendations that were finally presented” commented Professor Mick Dumper from the Department of Politics who convened the simulation with Professor Rex Brynen  from McGill University.

In designing the exercise we decided against the usual two-or-three-turn crisis simulation format wherein participants assume the role of actors with an ability to take actions that iteratively affect the course of events. Such an approach, we felt, would limit focus too much energy on move-countermove and too little on developing and assessing a broad range of contingencies and UNRWA policy options. On the other hand, we did want to use simulation-type mechanisms to generate participant engagement with the complex and dynamic operational environment in which the Agency finds itself.

In the end we decided to run the exercise as a sort of competitive brainstorming exercise, with a variety of simulation mechanisms added to this. Most of the participants were allocated to one of three policy teams. Each 5-6 person team was given the same near-future scenario, and was tasked with producing policy options for the Agency that would address “key challenges, short and medium-term policy initiatives to address these, risks and associated contingencies, and possible future challenges.”

139772-unrwa appeal 2013The simulation started with a 40 minute briefing from the UNRWA Commissioner General, who did a superb job of making the scenario come alive for participants and making it clear what he hoped to see from the teams. While UNRWA staff played themselves and were available for consultation throughout the exercise, we also had a small number of experts role-playing various Palestinian, regional, and international actors. Each of these held a brief (15 minute) “press conference,” to which each policy group sent at least one member. Groups could also schedule one or more meetings with UNRWA or the various stakeholders. By listening to and speaking with each stakeholder the policy teams were thus given a simulated taste of the complex operational environment and the associated issues facing the Agency. Additional scenario events were also announced during the exercise. These injects served both to underscore the dynamism of the regional setting and underscore the pressing urgency of the various challenges.

The teams were given a full day to consult, discuss, develop ideas, and prepare their presentations. The second day was devoted to fine-tuning these, presenting their analysis and recommendations, and discussion. We also asked all of the stakeholder players to write short papers on how they thought their actor would want the Agency to act in the scenario setting, and these papers were also briefed back to the full group for discussion on the final day.

Most effective simulation designs involve a certain degree of psychological manipulation-by-design of the participants, and this one was no different. In order to motivate the groups, we made the exercise deliberately competitive, with two winners (one chosen by UNRWA staff, the other by the stakeholder players) being selected at the end of the second day. While two of the three groups were made up of academic, NGO, and policy community experts, we formed one group of entirely of graduate students specializing in refugee issues, supported by a senior mentor/facilitator. We calculated—quite correctly—that the student group would throw themselves at the task with particular energy, enticed by the prospect of beating some of the senior figures in their fields in their own area of expertise. We also calculated—again, quite correctly—that fear of losing to students would equally motivate the more experienced participants!

So how did it all work? Overall, I think it was quite successful.

The policy option papers were all quite good. I did think that some of the recommendations didn’t give adequate consideration to the rather challenging  environment we had outlined in the scenario, and instead reflected more the prior political, academic, or technical inclinations of participants. Still, that is hardly unusual, and this effect was  less pronounced than had the meeting been held in the usual conference, workshop, or seminar format.

Some of the stakeholder players may have felt rather under-utilized during the exercise, since they didn’t get to directly participate in the policy group discussions and some had periods of “down time” when they weren’t engaged in either press conferences or bilateral meetings. From our point of view, however, they did a superb job and were an essential part of the exercise: the “press conferences” were remarkably lifelike (right down to avoidance of awkward questions), the stakeholders were generally in high demand for meetings, and the short  perspective papers they each wrote were themselves very useful too.

While there were perhaps a small number of participants who felt a little out-of-place in a simulation, but most took to it with considerable enthusiasm. Unlike a regular workshop we really made them work for their supper too, with considerable participant effort going into all three presentations.

Finally, I think people enjoyed themselves. I certainly did! One of the real advantages of the simulation methodology was to provide an opportunity for an intellectual cross-training of sorts, with participants able to got at familiar problems in new ways and from new perspectives.

%d bloggers like this: