PAXsims

Conflict simulation, peacebuilding, and development

Discussion welcome: (war)gaming the US as ally and adversary

eagle

I’ll be giving some thought over the next week to “(war)gaming the US as ally and adversary,” for a piece I hope to write soon. I have always been interested in how we model actors with murky or complex decision-making processes, as well as actors who may at times appear irrational (North Korea for example, or Qaddafi’s Libya). How much of this is simply different worldviews and interests, and how much of it is truly non-rational? How can pol-mil wargames best generate policy responses that reflect ideology, confirmation bias, pride, narcissism, bureaucratic infighting, and other non-realist determinants of strategic or operational behaviour?

In particular, in the coming months and years those of us outside the US who do national security gaming may need to consider:

  • How best to model unpredictable US behaviour (say, wavering alliance commitments) or behaviours that veer between supportive and threatening.
  • How best to model the US as a partial adversary or threat to national interests (for example, on trade policy, or liberal democracy).
enet9tmt71_tvhsae-1.png

No this isn’t a real tweet.

Any ideas are welcome in the comments section.

14 responses to “Discussion welcome: (war)gaming the US as ally and adversary

  1. brtrain 23/02/2017 at 5:54 pm

    I’ve been waiting weeks for you to make such a post/ announcement, Rex!
    I too will be thinking on this…

  2. gersonvmjr 23/02/2017 at 5:54 pm

    Well… What can I say? As a first point, I must say that at whole American continent, US don’t have enemy countries… What may exist from times to times are enemy or crazy political leaders, what maybe has been the exactly situation during the last years… This mad guys may conduct their countries through irrational behavior…
    Irrational behavior from the US I guess is a difficult thing to imagine, because of a solid democratic system and the US Congress supervisioning… And even this happening, I suppose this will be soon corrected…
    Outside the American continent, the situation is completely diferent…

  3. Joe Saur 23/02/2017 at 6:00 pm

    I’m glad you only ask easy questions… One option goes back to the way NWC depicted the Japanese Navy prior to WWII: in each game, the behavior, weapons, etc., were different. Thus, for your issue, multiple versions of each country: as it seems to see itself; as the current opposition sees those in charge; how their competitors/allies/coalition partners see them, etc.. It won’t be a single “US”, for example, but 5-6 different ones, and the results would help define the US that some other entity would like to see emerge (and which they might want to nudge along…)

  4. Rex Brynen 23/02/2017 at 6:14 pm

    My initial thoughts are it depends on how much fidelity one needs, and how important US policy is to overall game dynamics. One could reduce it to a random card draw, which sets the (changing) strategic parameters for the US team each turn. Alternatively, as Joe suggests, one could design a game-within-the-game, wherein various US factions (foreign policy conventionalists, radical ideologues, self-aggrandizers, etc) all compete to shape policy, and these players also have to balance how much effort is devoted to shaping vs implementing strategic decisions.

  5. John Carter McKnight 23/02/2017 at 7:37 pm

    We just had this discussion in my undergraduate course, Simulating Global Crises. We’re working on response to an earthquake in a fictional country in the Caucasus, and students asked why we hadn’t played out US involvement yet. It sparked a discussion on discontinuity and potential Russian influence in US policy, “spheres of influence” thinking, and an unpredictable policy process.

    I’d initially wanted to run the scenario in a past year in order to present a consistent US humanitarian/military policy, but got talked out of it. So we’re going to play out the uncertainty in real time.

  6. Mark Wallace (@markwallace) 24/02/2017 at 1:42 am

    Love this idea. Can’t wait to see what it generates. It sparks some interesting ideas in my mind from a game-mechanical point of view.

  7. Robert (Bob) Cordery 24/02/2017 at 4:39 am

    The most obvious way in which to model what you are proposing is to ensure that the ‘humans in the loop’ to whom you give the roles of representing the US in a game are capable of thinking and acting in accordance with the current administration’s agenda … whatever that might be. This could be partially covered by writing detailed and relatively constrictive player briefings, but then the problem arises if the game is more fluid in its form and is likely to stray from an expected path. Getting people to ‘act’ in the right way requires a degree of casting of player to role, using their temperament, knowledge, and understanding – and perhaps even their psychological profile – to ensure that they will perform as closely to role as is reasonable to expect.

    No necessarily an answer to your questions, but hopefully it might spark a few comments that will be.

  8. Anja van der Hulst 24/02/2017 at 4:44 am

    At the moment, imo the best way to explore the new situation with the Trump administration in office seems to be to do a fair bit of Matrix gaming- but simultaneously I’am not sure whether it will give more insight. This week we Matrix gamed the Black sea- and the Baltics situation and observed indeed that the US role was fairly hard to play as it was so totally unclear how US would respond to Russian provocations, whether it would back up Nato, whether it would totally lash out, or e.g. just say ‘fake news’…

  9. texarkana23 24/02/2017 at 5:57 am

    For initial unpredictability but with an emergent pattern once you get going, perhaps something based on drawing options from a “Polya’s urn”? I have been thinking of experimenting with this after reading: https://www.technologyreview.com/s/603366/mathematical-model-reveals-the-patterns-of-how-innovations-arise/ (I was thinking of using it to generate technology advances, as tech trees are a bit too predictable, but the concept could be adapted).

  10. Nicholas Murray 24/02/2017 at 9:02 am

    For my Po Valley in 1796 game I have a very simple system that uses drawn cards “Special Happenings” with inputs on them. They deal with everything from political interference, conflicting demands from other theaters, etc. All are also based upon actual things that came up during the campaigns. I have used this system for other games, and have gone with a simple list and dice roll (with 100 “Special Happenings” decided with two d10). That served to throw curve balls (a googly for the non baseball inclined). It took a while to set up, but once I had a set of things to use, it became a problem merely of choice. This fits nicely with the three states involved, as Piedmont, Austria, and France all have their own conditions for success. The cards allowed for the players to choose what to do with their goals, actions, policies, etc., and how that might change the relationship to the other players.

  11. Nick LaLone 24/02/2017 at 1:28 pm

    Been thinking a lot about this article:

    https://motherboard.vice.com/en_us/article/how-our-likes-helped-trump-win

    And the various ways algorithmic society was blamed and scapegoated in the past election. From there, I have been trying to figure out if you can boil these data down to a gameable set of variables to make some sort of war/game based on the proliferation of certain kinds of rhetoric.

    It seems as though you should be able to do it relatively simply. However, you’re stuck (at least within a gaming sense) with a “game” that is politics. Gamers tend to hide when these things appear.

    Also, hi Anja!

  12. John Curry 24/02/2017 at 3:54 pm

    When faced by external threats, the tribe practically always unites against the threat. This is well established. So America, despite all its internal differences, will be almost certain to pull together when threatened. However, will America see its tribe concept extending to include Europe, Japan, Philippines, Taiwan and South Korea? Europe will stand more chance of being defended by reflex (as European is seen as filled with similar people to America), the other nations less so.

    How to model changing policies? The old game NATO by SPI used simple dice rolls to determine if individual member states would react to Russian invasion. Decision games continue this with political options determined by a six sided dice roll e.g. by changing the amount of victory points awarded for each objective. Activation rolls are used to model command and control, so some games add +1 to these rolls if there is political uncertainty.

    Another method is Confrontation Analysis (the handbook of this method will be going to print in a few months).

    The problem with including varying America intervention is it changes the game completely so a different game system is needed; 1 for American intervention, 1 for not. e.g. The Baltic Republics are invaded: with American support for NATO, there is a limited war ending in Russia’s defeat- a kinnetic wargame. If America does not support NATO, then Europe would prepare for a long war and mobilise the greatest concentration of soft power the world has ever seen in order to bring Russia to the negotiating table. This would be best modelled by a seminar, matrix game, confrontation analysis or other pol-mil methods.

    Hope that helps a bit.

  13. Klemens Köhler 26/02/2017 at 4:28 am

    I am a big fan of emergent behavior. Having said that, I prefer one of two options: Either modell and game out the complex decision making processes, or set more or less random goals (draw cards if necessary). Of course, you would have to do multiple runs to identify the important aspects, but using a Matrix-Game-Frame this shouldn’t be a problem.

    The goal of all this gaming would be to identify the gamechangers and not rely on one outcome or game. If every game results in the US using limited military resources to achieve unlimited goals (i.e. regime change), the you might have it. But if, on the other hand, you have a game, where this does not happen, maybe you are onto something. You can infuse the unique aspects of this game to wargame specific scenarios with more fidelity.

    In order to throw an even wider net, you could use sideshows to look at what distracts administrations the most. I think just putting the migrant cards on the table will have a Trump-Administration ignore most of the other stuff that’s going on.

    I experiment with costs of decision making and restructurinng. I.e., if you use a Matrix-Game framework, you could make the players fill their organizationsal chart, assign responsibilities to staffers, but everytime they restructure, because the administration wants more direct control (or any control at all), they have to give up political capital or their gameturn.

    My final point is that for this kind of process, you do not need experts or player castings. Just make sure that the players follow their rules. As economics teach (see Vernon Smith), give them real monetary rewards for achieving their goals. This changes player behaviour under experimental conditions a lot. Additionally, to model democratic processes, you could let players, who do well participate in another game.

    As an economist I would stand against the idea of putting irrational actors in the game. Everybody should act rational according to his goals. From a limited IR-perspective, this may look irrational, but analytically, it will make more sense.

    I believe that most of these techniques work for the US and a potential adversary. Of course, the sides should not have the same starting conditions and goals.

  14. Lorenzo Nannetti (@LorenzoNannetti) 01/03/2017 at 3:25 am

    Some aspects that are key, in addition to others: 1) use non-US people to evaluate if they feel the US has a significant leverage/influence in the issue to be analyzed. It may be too simplistic at first glance, but US people tend to have an overbloated consideration about their country’s ability to influence events wordlwide, and while this ability is certainly there in some aspects (it’s still the US after all), it’s much less so in the current and future world in several others. Sometimes having non-US people evaluate the limits may make sure you don’t overrepresent the US ability to effect events and situations. At the same time, be careful not to fall into the opposite problem. 2) use non-US players to play non-US actors (may look straightforward, but one problem in US foreign policy from 2001 was thinking everyone would use the same worldwiew) – they will more likely have a different perception of how others may act in respect to the US than US people, which means their perception of the US as “ally” or “enemy” will be more realistic. This works better if at the same time you have US people playing US so the difference in worldwiews is enhanced – non-US people will help show where the US is perceived as a threat or competitor more than an ally (even for Europeans). They will show you what they fear, what could be considered problematic, etc… You could even get from them ideas on US events/decisions they really consider problematic, then you could work these as “starting points” or “events” during a game. However, regarding “erratic”, be careful about considering it “casual” – even the Trump admin is not acting so erratically in foreign policy than most thought beforehand, thanks to the actual people doing the job (Mattis, etc…) which means that solid knowledge of existing US positions among leadership groups is key to understand the debate – and therefore possible options.

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: