Trip said:
Sending 10 Missionaries to flip a city is obviously a hostile action - just as hostile as sending 10 Archers. Why worry about declaring war on the enemy civ when war has already been declared? Sending 2 Missionaries will accomplish zip except for a waste of resources.
I wish I'd stepped in sooner before things got out of hand, because I think you're looking at the idea too narrowly. What we're essentially talking about is the prisoner's dilemma. Do you know how the prisoner's dilemma works? Pardon the lecturing, but there's a very valuable lesson to be learned here:
The Prisoner's Dilemma
Two partners in crime are apprehended by the authorities. They are suspected of assault, but the best the authorities can point out is theft. Putting them away for
3 years each for theft isn't enough, and they'd like to be able to put them away for assault. The only way the authorities can do this is to try to get one to rat the other out. They seperate the partners from each other and try to make a deal:
if you rat your partner out, we'll give you a shortened 1 year sentence, and nail your partner for 10 years for assault. The partners also know, though, that
if they both rat each other out they both get 6 years for assault.
Each prisoner makes a choice: do I rat, or do I keep tight lipped?
- If I rat him out, I'll get only a 1 year sentence!
- But if he rats me out, we both get 6 year sentences
- So maybe I should keep my mouth shut and get a 3 year sentence.
- But if he rats me out, then I get 10 years in jail!
It's obvious, then, that the smart thing to do is to play defencively. You should rat your partner out -- because that minimizes the worst case (6 years) AND maximizes the best case (1 year). It's a no brainer.
Extending the Prisoner's Dilemma
However, that's if you only play the prisoner's dilemma once. Compare two pairs of partners. George and John, versus Bonnie and Clyde. They're going to play the prisoner's dilemma 10 times in a row, and we're gonna see who ends up with the lightest sentence.
George happens to be a lying, untrustworthy son of a gun. He says "I'm going to rat out John to try to get a smaller sentence right off the bat." John says "I'm gonna take a chance and hope my buddy is honest with me." In the first turn: George gets 1 year, and John gets 10 years.
John, aggravated, decides that he'll never trust George again. They both continue to rat each other out each of the next 9 times they are caught (why they work together after that and constantly get caught is beyond me). They both get 6 years each for the next 9 turns, that's 54 years.
For George, that's a total of 55 years.
For John, that's a total of 64 years.
Compare that to Bonnie and Clyde. They make a pact: I will never rat you out. Every single time, they refuse to rat each other out and both get the small charge of 3 years, over ten turns.
For Bonnie and Clyde, that's a total of 30 years each.
Bonnie and Clyde cooperated and did better than even George, who tried to cheat to get ahead. (Poor John ended up the biggest sucker in this deal.)
What does the prisoner's dilemma mean?
This is the foundation for what some people call "Game Theory". Game theory is about relationships between two people where there's no distinct winner or loser. In the prisoner's dilemma, everyone's a loser, but some people lose more and you can minimize those losses. You can translate this to anything. One example is two fishermen.
(I'll make this quick.)
A fisherman can catch 2 fish on his own on a fishing trip.
Two fishermen can handle a net, that allows them to catch 10 fish in a trip.
But one fisherman can take all 10 fish for himself and run away.
You can see that the greedy strategy of screwing your fishing partner is only going to work once (like George and John) and you're going to end up with two guys fishing alone. Whereas two guys who cooperate will split the 5 fish and both turn out to be big winners in the long run. That's only if they can maintain the long-term trust required to be fishing partners.
Missionaries as a Prisoner's Dilemma
A nation can generate a little bit of culture on its own.
A nation can generate a lot more culture by sending a missionary to a neighbour.
If the neighbours betray each other, then there's no chance for missionaries to work.
Imagine this situation: Germany sends a missionary to France, getting themselves a bonus of 20 culture. France decides to send a philosopher back to Germany, in good faith -- and Germany KILLS the philosopher. The two engage in war for the next hundred years. They manage to pump out missionaries, but can only send them around domestically, getting 10 points each.
Now imagine this situation: Britain sends a missionary to America. America sends one back. They are both received graciously. They continue to pump out missionaries for the next hundred years, getting more culture in the world, all while Germany and France are at war.
At the end of the game, Britain and America will be much better off culturally than Germany and France.
The Value of Cooperation
As a small anecdote to show that I'm not just making this up... (although it's a thought experiment. If you follow the logic, it should be self explanatory. I've tried to be brief, if you can believe that.)
An AI researcher did a competition on the prisoner's dilemma. Everyone brought in their own algorithm that was supposed to try to win. There were 200 agents who would interact with each other 200 times each, and they would weed out the agents that did really poorly. They would repeat the experiment with fewer agents -- the one that did well in the last round -- until one algorithm floated to the top.
The AI that tried to cheat all the time actually ended up losing. It quickly became "isolated", and nobody would trust it, and thus it would always get a hard sentence.
The AI that was nice all the time, I don't have to tell you how that went. They got screwed. Badly.
The winning algorithm was something called "TITFORTAT". The program was 4 lines of code:
1. Co-operate by default
2. Store in memory what this person does to you.
3. If they screwed you last time, screw them this time.
4. If they co-operated with you last time, co-operate with them this time.
The Belaboured and Tedious Point
We finally made it. What's the point?
The point is that to win the culture war, you need to be aggressive, but not too aggressive. Sending 10 missionaries in one turn is a sure fire way to make your opponent declare war on you. Another example: scoring with your missionary one turn and then rejecting your neighbour's missionary the next turn is a surefire way to lose someone's trust -- you're "cheating" your neighbour.
But the neighbours that can coexist and engage in this mutual culture-spreading will end up being better off for it. Maybe they'll both be neck at neck in terms of culture, but they'll be light years ahead of the continent where all the nations are at war.
And therein lies the REAL competition, and paradoxically, the REAL reason to get along.