We've had a number of long weekends in December, which is very convenient if you’re the sort who enjoys large, complex board games and can get enough friends over to play them.
The granddaddy of large, complex board games that can take the best part of a day to finish is Diplomacy, a war game that is like Risk if you didn’t need to throw any dice. You win battles through negotiation, and it’s probably a failure if at least one player doesn’t want to kill another at the end of four hours.
It’s a very appropriate game to play right now because it’s about deal-making and politics. Armies are moved around on a map of Europe, and whether a battle is won or lost depends on whether your army has enough support to win it. The game is designed in such a way that you cannot win on your own and need the support of other players to really get anywhere.
It’s a game of quiet discussion punctuated by intense emotion. I once played a game as Austria, a country stuck in the middle of the board so it has to ally with somebody (anybody) if it wants to make progress. In the end, I struck a deal with Italy, who three moves later “accidentally” wrote down a move that sent his army into my Trieste (for those who care about history, in real life Trieste used to belong to Austria before it became an Italian city after World War I). The move after that, he reinforced his beachhead and exposed my weak western border.
With my back exposed, Russia and Turkey decided to carve me up between themselves. I then made it my life’s mission to bring Italy down, whatever the cost to myself. A few moves later, both of us had mutually destroyed each other. No, that wasn’t necessarily the best thing to do either tactically or strategically.
Much has been written about how to win the game: Understand the moves and how the game works; see the big picture and go for the long-term advantage; and (please take note, Italy) players who backstab a lot don’t do well in the long run. This is the best advice given by people who have had hundreds, even thousands, of hours of experience playing the game – but as we know, humans are susceptible to emotions.
Which made it very interesting when I found out that software engineers have written an artificial intelligence-based software that plays Diplomacy.
Cicero is an AI program that has performed better than most of the human players it has played against, eventually ranking in the top 10% of participants. It has a component designed for strategic reasoning that can evaluate what the best possible moves are, for both itself and its opponents. In this way, it’s similar to current top chess engines.
When it plays, Cicero tries to find a match among the best moves it can make and the best moves another player might be considering, and comes up with a set of mutually beneficial policies.
These are communicated by a second component that handles natural language processing. This component takes the policies and translates them into natural English as a proposal that will hopefully be accepted.
Say Cicero is playing Austria and, like me, it knows that Italy needs to be dealt with. It identifies that Italy might be thinking of attacking Turkey via the Mediterranean, but may be wary of Austrian aggression. So Cicero comes up with a proposal that says if both Austria and Italy can commit to keeping a gap between them, then they can both safely attack Turkey.
It then sends a message to Italy saying, “Hey Italy, any interest in keeping Tyrolia as a DMZ (demilitarised zone)?” and then following that with, “Then you could attack Syria while I go for the Aegean”.
In making proposals like these, it is important that Cicero resists the temptation to undermine allies for short-term gains. In fact, what the developers found out was that when the AI was still learning in its early days, it would try to renege on deals and even outright lie about its intentions. However, over the long-term the program discovered that being truthful is actually more beneficial.
The reason is simple: Nobody is going to make deals with you if they think you’re untrustworthy. So even your greatest enemy across the table would be willing to work with you if the offer is mutually beneficial and – even more importantly – if you have a reputation for keeping your word.
Which is ironic when it comes to politics, of course. Politicians have a reputation for lying a lot of the time and not keeping their word. Yet, so much of politics only works because of deal-making, even between ministries in the same government.
This is even more crucial in this era of a fragile political coalition like Malaysia’s unity government. Can the “players” trust each other? The only way to begin earning trust is to make small deals, deliver on them, and then work your way up to bigger ones. Cicero additionally says, talk to everybody (even those you disagree with), and remember who keeps their promises (as well as, don’t lie!).
The largest such promise to deliver on will be the vote of confidence that the Prime Minister is planning to hold tomorrow in Parliament. Current indications are that it will pass, and it will represent another milestone towards a stable government.
And, hopefully, all will bear in mind the words of the real Marcus Tullius Cicero, Roman statesman and philosopher: “A liar is not believed even though he tells the truth.”
In his fortnightly column, Contradictheory, mathematician-turned-scriptwriter Dzof Azmi explores the theory that logic is the antithesis of emotion but people need both to make sense of life’s vagaries and contradictions. Write to Dzof at lifestyle@thestar.com.my. The views expressed here are entirely the writer's own.
Already a subscriber? Log in
Get 20% OFF The Star Digital Access
Cancel anytime. Ad-free. Unlimited access with perks.
