PAXsims

Conflict simulation, peacebuilding, and development

Reconstructing Afghanistan (2013)

On Friday night, a group of students from my POLI 450 (Peacebuilding) class attempted to reconstruct Afghanistan using the Afghan reconstruction game. This is the second time we’ve played the game at McGill, although this year we modified the original rules to add a Taliban team. (In the original version of the game Taliban actions are entirely determined by decks of random event cards.)

We also somewhat modified the rather confusing way that victory points (the “National Stability Index”) is calculated in the game, although in retrospect the revised formula was a little too harsh.

While there are around one hundred students in POLI 450, there were only spaces for 18 to play the game. This isn’t a particular problem in this course, because the “course participation” grade includes a number of optional activities, of which this game is only one. However, it could be a challenge integrating the game into larger courses organized in a different fashion.

As the video above suggests, the game went well, with a lot of student engagement. While the NATO team and the Kandahar provincial team came close to eking out a victory, in the end the result was a substantial Taliban victory (perhaps a not entirely unrealistic outcome).

Below I’ve collected together some of the comments that students had about the game. Not only do they provide insight into the strengths and weaknesses of this particular game design, but they show how quickly a group of neophyte players can move from playing into thinking about the underlying social models and they way they are represented in the rules. While we didn’t have much opportunity to do so on Friday, such a discussion over game design can reap significant analytical dividends. What are the essentials of reconstruction? How does development relate to political stabilization? What is missing from the game that is important enough in real-life peacebuilding to add? What effects are “unrealistic”—and why? …and so forth

FSJ, the Taliban commander in Khost province, commented:

I though the game was very interesting in laying out the fundamentals of condensing and simulating a nation-building/conflict resolution situation into a board game.  There were both some very through of design features, and also some very not-thought-through design features when we played the games.

Good features:

  • The rock-paper-scissor method of a combat made the game more engaging (and even sometimes aggressive in an entertaining way) between actors.
  • The ‘resource’ and ‘influence’ chips were very well thought of because it helped simulate reality forces that polarized decisions and outcomes on both a provincial and national level (and to win the game).
  • It was also good to have the Taliban added than any other actor in its place, because it helps simulate the reality and complications that this simulation addresses to teach.

Bad features:

  • As Prof. Brynen mentioned, as per the funding for the game, it was indeed a bit odd to see Christian-based NGOs in the realm of Afghan provinces, such as the “World Church Union.”
  • The scoring of the National Stability Index was a bit wonky in terms of being equitable to all parties, i.e. the provinces and their actors and the Taliban.
  • SInce the event cards were the same for all 3 provinces, it was not as specific to that province as it could have been. More specific event cards would have made the game more interesting on a provincial level.

All in all, well worth the 5 hours of play, and WOOHOO the Taliban had a National victory!

The “rock-paper-scissors” system mentioned above for resolving combat was one of my game modifications, and it worked really well both as a random-outcome generator and as an engagement device. These fight’s invariably elicited a great of attention, and the US Stryker battalion commander in Kandahar province become something of a NATO legend for her success in fending off multiple Taliban assaults.

MM, the Taliban commander in Kunar province had these comments:

Just thought I’d offer comments as well on the Reconstruction Game. The Game was a very engaging and entertaining mode of learning about the on-the-ground difficulties that multiple actors face while attempting to undertake sustainable reconstruction in a conflict-ridden area. It highlighted all the vectors of development which must be undertaken by different actors according to their differential capacities to achieve national stability. As I was a Taliban Representative to the Kunar province, my views may be a bit one-sided.

The merits of the Game lie in its emphasis on strategy. This is true not only of the Provincial and National forces (The Afghan Govt., NATO and the NGOs), but also of the Taliban forces. The Game highlighted the paramountcy of coordination, priority-management and foresight. At the national level, actors were faced with the task of effective devolution of resources and influence amongst local actors. On the provincial level, local actors realised their fundamental role in the maintenance of national stability. Of course, this was realised in the form of the National Stability Index. I noticed that when the Game got rolling, actors from all 3 provinces started coordinating reconstruction efforts (in particular, completion of key projects in development and security) to garner enough points to increase the Index.

Professor Brynen’s inclusion of the Taliban actors and their role in the seasonal “insurgency” was one of the most enjoyable and entertaining aspects of the Game. It also definitely made the Game more realistic. The actors in this rendition of the Game were not able to achieve a NSI over a 100 — at the end of the Game it was still below 50 and none of the Provinces had successfully completed their projects, although 2 Provinces were only one short of this mark. For all intents and purposes, I would count that as a realistic win.

Although actors could have won on a provincial level, I felt that efforts were dedicated toward the end of national stability. I am not sure exactly how realistic this would be. As Professor Brynen mentioned, in reality, Kandahar Province received most attention in terms of aid and resources during the War and recon effort because of the presence of important actors such as the US and NATO.

An obvious drawback to the Game was the identical structure of all the Provinces (they even had exactly the same maps and local leaders). It is important to be sensitive to not only the particular conditions of each locality, but also to the diversity of its people. Not to sound too harsh, but the Game ran on a principle of appeasement (in particular monetary appeasement) of the local Malik, the local religious leaders (Mullahs) and the local Majlis-e-Shura. There were no efforts to include the local actors in the reconstruction efforts, which is an important vector emphasised in lectures time and again. Perhaps another category where players can place their “influence chips” could include the local populace.

The calculation of the NSI was skewed, as all of us noted. It made it very difficult for the provincial characters to advance in the Game. The “Event Cards” seemed too harsh — it was very difficult to advance with development projects which kept being obstructed by freak accidents such as storms and cholera outbreaks. While it makes sense that a cholera outbreak should test health sector development, it does not have to erase all development hitherto achieved.

I want to thank Professor Brynen for letting us partake in this great exercise. The pizza was also yummy!

Once again, I should reiterate that the difficult in increasing the “National Stability Index” was the result of my own over-adjustment to the rules, following last year’s game when the Coalition players won too easily.

The Afghan Government player in Kunar province commented:

I agree with all of the pros and cons you’ve listed. Here are a few others:

Pros:

  • It was very interesting to see the shifts in alliances and loyalty that occurred towards the end of the game since it was possible to win with your organization instead of just with your province.
  • I think the ‘telephone’ idea of not being able to show intelligence cards to other members and only relay information verbally was a great addition to the game. It created a realistic aspect of what it could be like in the field with miscommunications or problems with translation.
  • I think that having to buy influence from the Shura was a great aspect to the game since it underscored the importance of having influence first to get things done and the possibility of having all your work halted when alliances sour. However, as we discussed during the game, there was a flaw that made it more beneficial to accumulate lots of influence instead of focussing on the projects. That flaw was partially counterbalanced by the requirement of 6 completed projects to keep the Taliban from winning in your province.
  • The impact of having a warlord was also very interesting because it created different arguments for strategy in terms of fighting the Taliban and resulted in some realistic punishments (withholding funds) from the national level because of ignored orders.

Cons:

  • As mentioned before, the point scale made it very difficult to raise the stability score, especially when the score needed to be rounded down to the nearest tenth.
  • After a certain point, it started to feel like the provincial events card had a stronger negative impact on our projects than the Taliban.
  • It seemed a bit too easy for every single actor to have all their projects completed by the deadline.

Overall it was a fun way to learn and even a little stressful to juggle so many responsibilities while on a clock. The game gives you an idea of the dilemmas of negotiations and allocations of limited resources.

SM, a national-level Taliban leader, also noted how much the introduction of active Taliban players contributed to the game:

I thought the Taliban was an excellent addition to the game. As a member of the Taliban, we were able to exert strategic decision-making throughout the game. For example saving cards up to strike at once, spreading resources fairly evenly through all provinces, using our money efficiently etc. That said, I think the game would be greatly improved if the provincial and national actors would be able to exert the same amount of strategy that the Taliban were permitted to through their role. This could be facilitated through provinces and national government following the same system of pulling cards and being allowed to save them strategically that was enforced by the Taliban (instead of just pulling cards and having it happen right away.)

I also think that the cards themselves for the provincial and national governments should be changed. I found them too harsh, it almost became guaranteed that every round a project would be destroyed in the provinces. However, this problem was solved to an extend by the ability to over-build on projects, which would be useful to do from the start the next time the game is played. It would probably also make it more difficult for the Taliban to win provincially.

As previously mentioned there were some problems with the national index scale being somewhat flawed. However, this can be easily fixed through a new scaling system.

The ability for the Taliban to build up money reserves through drug smuggling and then spend money on additional cards for insurgency purposes was a nice touch and accurate to real life.

As discussed after he simulation, the level of communication between provincial and national level of governments, provinces and organizations was perhaps a bit unrealistic. Especially with the ability to trade resources and influences between them.

All together I thought that it was a really fun game, the addition of the Taliban created a fun and competitive environment. It was a great learning tool and I would definitely participate in this kind of thing again.

The game as originally designed for National Defense University played down the military strategy and competitive components because the audience were primarily military officers who needed to focus on issues of interagency coordination, not kinetic operations. My participants were rather different, however. Compared to last year, moreover, I thought the Taliban players (who were routinely booed when they entered the room) made the whole environment more fun. One of them also uttered perhaps my favourite line when, during a slightly heated point in the game, she asked “What’s with all this negativity?”—as she plotted how best to attack Coalition projects and personnel.

LK, the national-level Afghan government player, points to some of the coordination problems that cropped up in the game—and the possible need to build in even more of this:

What an informing five hours! A lot of what I have to same or comment on has already been said, but I still want to make a comment as a national participant of the reconstruction game. Overall, I enjoyed the takeaway. A the national GIRA player and representative of the Afghanistan government, nothing was more frustrating than seeing my fellow national NATO player two chips away from ‘completing’ all of his projects while our national stability index was in the red. I didn’t expect the final result to be reflective of the real world situation where NATO’s projects may be deemed successful but overall the country is worse off than before the conflict started.

As a national player, I expected more coordination problems to crop up. The only glaring issue we ran into was if I didn’t speak to NATO about where they were sending their military forces. If we didn’t ensure that every province had capable forces, then Taliban attacks could occur without the chance of fighting back. I started to pay attention to who was getting what NATO forces, though it seemed that my counterpart could care less who was getting the Afghan army or police. That proved irritating as well. I do wish that more coordination problems could have occurred to prove the importance of national level players coordinating with the government. Not once did the national NGO player ask what my priorities were either, but instead went straight to her provincial level players. I wish I had a voice in that as well but I’m not sure how that could be worked into the game.

National events, the way they played out, had a tendency to severely limit my resources at times, leading to discontent from my national players. In the initial phase of the game, I and my NATO counterpart requested that two provinces with warlords get rid of them and gave them the extra resources to do so. When they didn’t, I threatened to limit resources and I did – giving one cash chip instead of three. My NATO counterpart’s threat didn’t hold and his players didn’t care. Unfortunately, getting rid of the warlord was relatively useless and I didn’t have the foresight to realize that every other provincial event card (or so it seemed) had another one popping up. Still, the idea of corruption and having my resources given by my provincial players to the warlord every turn was frustrating.

But on the topic of national events, the cards seemed powerful but not as powerful as provincial events which could wipe 6 projects off the board. I wished there were different sets of events for each province and, like many other players have mentioned, that the provinces themselves were more differentiated in their dynamics, their projects and their needs. I had a tendency to treat all three provinces the same, if I gave more to one turn, another got extra the next and so on. I wish I had been forced to consider trade-offs of different projects in different places. Allocation on my side was always very fast and efficient. By the time national resource allocation rolled around I had almost always decided what I was going to do and I just doled out to my players. The NATO player on the other hand had a lot more bargaining going on and begging from players saying what they were going to use it for. I tried to be more demanding and give my players resources for specific things – though they didn’t always listen. Other than limited resources the next turn, I had no other way of rebuking them which kind of sucked.

Regardless of the troubles and mild coordination problems, I very much enjoyed the game and the pizza.

The pizza—from Angela’s Pizzeria—is always good at McGill game nights!

The overall NATO commander, TS, had this to say about the game:

Although the odds were in their favour, player-controlled Taliban actors were an important (and more realistic) addition to the game. Their ability to survey our actions and conduct theirs in private was a good part of that—the disparity in our information let them react more appropriately to developments than we were able to react to their actions, much as I’d expect.

At the same time, I was never really as worried about the Taliban as I was about the dreadful national and provincial cards. LK and I looked through the deck of national cards after the game, and positive developments were sparse, while negative ones could be terrible. With the Taliban actions, at least, we got some level of control with the rock-paper-scissors games. There was also no practical way to prepare for national cards, and mitigating the effects of provincial cards was hampered by reliance on my memory (and that of NGO players when the mechanic arose) and imperfect relaying of the contents. As Prof. Brynen mentioned near the start of the sim, there needs to be a way to look at national-level cards so that it actually matters for gameplay—possibly by looking at the second, rather than the first, in the pile. The addition of a fourth level for projects was also important. Without it, I think it would have been almost impossible to meet the alternative province or faction win conditions after the final round, with it, there was at least a chance. Sarah seems to have similar feelings.

As MM reminds me, Kandahar province was the recipient of most of the attention in real-world Afghanistan, and I’m tempted to say that an in-game strategy which mirrored that would be more successful. The problem is that the attacks from Taliban and the provincial event cards were all very good at removing projects when they were built up minimally, however, when many were in place it seemed much easier to weather the assaults. In another iteration, it might be best for the national-level NATO actor (and likely the other two national-level actors) to feed all of their resources and chips to a single province for rapid development, then repeat the process for the other two provinces while maintaining the developed province(s). Without needing to split up protection, the province being built up would be much less prone to attacks, as well.

My strategy was not complex; it consisted of giving each provincial NATO actor what they needed according to their situation as they presented it to me, punishing them for changing their plans if we agreed to a particular use for what I gave them, rewarding them for keeping their word, and generally assuming that they each knew what the most important things to them were. When it became clear we couldn’t win with the index, I switched to pursuing the national-level win condition for NATO. (Focusing exclusively on that condition from the beginning might also be a good component in a revised strategy.) I’m glad TM noticed the withholding of funds, I was worried the NATO player in her province might have just ignored punishment in his quest to be Rambo.

One of the bigger problems (although definitely a ‘pro’ in terms of creating a more realistic game) I saw was communication. LK discusses this, too, and I think the focus of her post might highlight one of the biggest barriers to communication between the national-level players (I was seated in-between her and the NGO player, and that seems to have made the difference in who everyone interacted with). While I think a strategy which relies on the provincial players to assess their own needs and bargain for what they want is a more effective form of allocation, neither it nor the more nationally-directed strategy LK employed (of deciding her allocations ahead of time) were going to be successful on their own, something I didn’t realize until near the end of the simulation when I switched to a more selfish strategy. I had wrongly assumed that everyone individually working towards their own goals at the provincial level would naturally lead to a heightened NSI. It might happen with a better overall system for the NSI, but a better communication system would still be helpful.

A probable model for more effective allocation would be one where the nine provincial-level players came collectively to us three national-level players (likely grouped as their provinces) and plead their cases for monetary tokens and chips which we would allocate together. This would allow us to get past the problem LK observes where protection wasn’t evenly distributed, increase accountability from the provincial actors, and to better implement an overall strategy where each player knew what the others were doing.

The NSI index formula would need to be changed (FSJ’s choice of “wonky” to describe it is spot on). The suggestion of using two tracks (1s and 10s) would be great, allowing the movement of the index to be more representative of the investments in each region. I think it’s possible that may be the only change necessary, though, if other changes are implemented to make it a bit easier for provincial players to build up their provinces (because they can then garner a higher increase in the NSI). You may want to decrease the impact of some of the more negative cards, too. One, for example, dropped the index by four stages without a possibility of mitigation, more than the Taliban managed in a few turns. Earlier in the game, it seemed possible that we could succeed via the common NSI win condition, but that proved unrealistic near the end.

(It was also interesting that warlords were seen as a form of protection/mitigation worth the recurring cost—I was staunchly against them but the provincial players seemed happy to have them around at times.)

In the original game rules, Warlords are solely a bad thing, draining your resources and undermining your projects. In the revised rules, however, they also adversely affect Taliban operations too, making their impact rather more ambiguous. As in the real Afghanistan, many local players found it more convenient to tolerate than to confront them, despite their costs.

KJ, one of the NGO players, offers his thoughts too:

All in all, I thought the game was a tremendous success. For me, the most evident lesson was the necessary interdependency of all the provincial actors. As an NGO representative, I obviously started out prioritizing my own projects. I soon realized, however, that my projects were very easily destroyed and made worthless without cooperating with the security efforts of the Afghan and NATO forces. Similarly, the Afghan and NATO forces really benefitted from the fact that, as an NGO, I was “close to the people” and often had intelligence on what provincial events were on the horizon. It became very evident why, in real life Afghanistan, a lack of coordination among provincial actors can be so detrimental.

For me, like most of the others, the biggest issues were the fact that potentially project-destroying provincial events occurred AFTER provincial resource allocation, and also that one could not ‘over-build’ one’s projects. This made it nearly impossible to possess fully completed projects at the end of a turn, which really crippled our ability to boost the NSI. I also agree with Tiphaine’s point that, most of the time, I was far more scared of the random provincial events than the actual Taliban. These need to either be toned down, or interspersed with more positive cards.

Like Mina, I was uncomfortable with the idea that securing influence with the tribal elders primarily amounted to paying them off. It would be more interesting, I think, if the influence->project completion relationship was more of a two way relationship: one needs the initial influence of elders to complete projects successfully, but successfully completing these projects and keeping them in operation for a certain amount of time also earns additional influence by gaining the elders’ trust. There are some provincial cards that achieve this, but I think there should be more. A provincial card, for example, that reads “the Mullah walked by a collapsing school today, and shows increasing concerns about provincial education projects. Maintain a fully completed “Education” track at the end of the next turn to earn 2 Mullah influence points” makes considerations of ‘influence’ seem far more realistic and reciprocal.

Also, there are necessarily a lot of rules to the game, and I think there could be additional visual reminders on the board to help keep the players on track (although maybe the problem was simply the cluelessness of the players, ie. me.) For example, the completion of “top-row” projects was supposed to yield one additional money chip each turn for the corresponding provincial actor. However, almost every player forgot this. On the Taliban/Warlord square it reminds the player that they must PAY a money chip per turn, so I think it would make sense to visually remind players that with completed ‘top-level’ projects they also RECIEVE an additional coin per turn. This, to me, seemed like an obvious omission.

I agree that the rock paper scissors game was a great way of making the game more exciting and involved. I did, however, find it hard to believe that the Afghan National Police was basically just as likely to defeat a coordinated Taliban strike as the US Army Stryker Brigade. Perhaps there should be a way to randomly factor corruption variables into the rock paper scissors games? Perhaps in 2 out of 3 rock paper scissors games the Afghan Army automatically “flees” out of cowardice?

But like most respondents, I would have to agree that one of the game’s best aspects was the complementary Angela’s Pizza. Thanks again, professor!

Based on this year’s game—and thanks to the players—I now have plenty of modifications in mind for next’s year’s POLI 450 class.


The victorious Taliban team pose triumphantly with a “Peace on Earth” poster while the Afghan, NATO, and NGO players look on glumly

One response to “Reconstructing Afghanistan (2013)

  1. Rex Brynen 20/03/2013 at 6:20 pm

    Some more comments on the game, this time from JM. they arrived too late for me to include them in the original post.

    * * *

    I really enjoyed playing this game, although it had some faults I think that it was a great success overall. What I realized during the game was that my province took a very different approach to playing the game than any other province. We decided from the start to strategize together and prioritize our needs and wants as a team rather than each organization for themselves. I believe that this worked to our benefit in the long-run as it became obvious that almost every turn would ruin someone’s project­­—thus the best possible outcome each turn was to ally together and play out the best possible outcome for one member of the group in the hopes that it would benefit the group in the long-run.

    On that note, one of the major flaws of the game is that influence is really all you need to win, and thus project loose importance. In fact, we actually played the game wrong in the beginning thinking that you needed to always be purchasing influence and as a result increased the country’s stability each round while other provinces only focused on themselves. Since we did this unknowingly and found out about half way through we had to time left to focus on our provinces projects, but during that time no other provinces really contributed to the overall countries stability. In fact we managed to have doubled all the provinces even after we found out. This is problematic because there is should a little more incentive for other provinces to increase national stability since we all gain from it in the long-run.

    This was brought up at the end of the game, but I think that it should be forgotten how much effect the provincial cards had on the game and in determining the outcome of the final round which near impossible to win nationally.

    The most significant aspect of the game that determined our fate was intelligence. We had many opportunities to gain intel and took them—however, each time without fail the intel was ill-communicated. This led to our failure ultimately. I think that this is a very valuable lesson learned, as we often take communication to be an easy task that just anyone can do, and that is not true. Just like the game of “telephone” it is skill to be able to rely messages accurately. This is a crucial asset to have in the game and one that determines the winner and the loser.

    Overall, I think the most important lesson learned from the game is cooperation and coordination. If your team can master that and negotiate with national leaders I think that it is possible to win the game.

    One last thing, I really enjoyed how the Taliban was able to interact with the rest of the members I think that this was very effective and should be continued in the years to come. Additionally, the “rock paper scissors game”, was surprisingly very effective in getting members to interact, focus and allowed for a more realistic twist to the game, confrontation and uncertainty.

    Thank you, again for inviting us to play the game and for the pizza!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: