Civil War

[Xyth]

Moderator and modder The_J has just created a Civil War event that causes one of your cities to secede from your civilization and become the capital of a entirely new civilization under a new leader. It's something I can easily adapt and I plan to start working on it straight away for 1.18 as it's an awesome feature that I and many others have hoped to see added to HR some day.

By default the event will trigger whenever a player has more than 1 city and a city that isn't the capital has 3 angry citizens. The city will become the capital of a new civilization under a new leader, all units in that city will switch allegiance to the new civilization, a couple of conscript units will be granted to the city for free, and war will be declared between the two civilizations.

Note that this event will only cause one city to secede, not a group of cities. I can change this later if desired, or add additional mechanics or events that might encourage other cities to join the rebel city, but there's time to discuss that later. For the moment I want to focus just on the triggers that could cause that single city to rebel, and the effects of such.

Let me know your thoughts on this. What other conditions should exist for a civil war to happen? What else should be considered? Due to the extreme complexity of doing so, I don't wish to add 'stability' measure like in the Revolutions mod has. Lets stick with existing measures and mechanics.

 

[inasnapp]

civ's biggest failure was not having civil war built into it, and when it did like in civ2, was still limited.
I would like to see it as developed as possible, it would not only be more realistic but also make for more interesting games if a civil war in the largest civ can throw the balance of power upside down.

 

[Nightstar]

What if there are already a full 18 civs in a game?

 

[Xyth]

civ's biggest failure was not having civil war built into it, and when it did like in civ2, was still limited.
I would like to see it as developed as possible, it would not only be more realistic but also make for more interesting games if a civil war in the largest civ can throw the balance of power upside down.

Best we start relatively simple and develop the system further over time.

What if there are already a full 18 civs in a game?

The 18 civ limit can never be bypassed (on Mac) so if 18 civs are alive in a game, civil wars will never trigger. For Windows folk using a custom DLL, the limit will be 40 or whatever else the DLL is set to.

 

[Howard Mahler]

I hate this idea.
While it makes sense historically, it is exactly the type of thing I hate in a game I am playing. Particularly obnoxious is the potential loss of a gigantic stack of military units.
I am more than happy to compete against opposing civilizations but not against something like this.

If it occurs in a game I am playing I would either reload or quit.

There is always a potential tension between simulating history and having a game that is fun to play.

As always different people react differently to different aspects of a game.

 

[Xyth]

I hate this idea.
While it makes sense historically, it is exactly the type of thing I hate in a game I am playing. Particularly obnoxious is the potential loss of a gigantic stack of military units.
I am more than happy to compete against opposing civilizations but not against something like this.

If it occurs in a game I am playing I would either reload or quit.

There is always a potential tension between simulating history and having a game that is fun to play.

As always different people react differently to different aspects of a game.

The key is to make it trigger from circumstances of poor management, and not just from bad luck. The player should be aware that it could happen have a chance to avert it before it does. That's why I'm asking for feedback and suggestions on what said triggers could be, dismissing it outright isn't helpful.

 

[Azoth]

I must admit, my first thoughts were similar to Howard's.
Randomly losing a city - its gold and science output, its resources, its garrison, the hammers invested in its current build, its buildings on recapture - would be very frustrating.

However, I think a compromise is possible.
So long as the mechanic is well designed, the threat of civil war could be quite engaging.
I think it's best to give players plenty of options for dealing with the issue. So I suggest an event chain:

I agree that the triggers should be kept simple.
A city that is severely unhappy (3 or worse) or starving (-2/turn or worse) has a chance to rebel.

If the rebellion triggers, the player would receive the following event:
"Rumors of civil war are swirling in [insert city name]. What should we do?
Option 1: The city is of no significance. Grant the fools independence, and spread tales of my generosity far and wide. The city becomes the capital of a new civilization, as a vassal state to the player. +1 relations with all civilizations.
Option 2: I will not tolerate treason. Declare martial law, enforce a curfew, arrest all suspected rebels, and have them shot. The city revolts for 15 turns, scaled by game speed. -1 relations with all civilizations.
Option 3: I will deal with the situation myself. 0-5 improvements near the city are destroyed."

A player who picks Option 3 has 5 turns to "deal with the situation." After 5 turns, if the city is happy and has not lost population, nothing happens. Otherwise, the city becomes the capital of a new civilization, at war with the player. (The rule against lost population not only requires that the player address food shortages, but also prohibits slavery and conscription as solutions to the unhappiness. Nothing inspires rebellion like the whip and the draft.)

So what do you all think?

 

[Howard Mahler]

Pirate temporarily cuts trade routes.
You lose access to trade goods.
-3 happiness, civil war.

Civ declares war on you.
You lose access to trade goods.
-3 happiness, civil war.

Civ declares war on you.
Blockades city.
-3 happiness, civil war.

Privateers blockade city.
-3 happiness, civil war.

You conquer city.
It comes out of revolt while you are still at war.
-3 happiness, civil war.

I think the AI will have even a tougher time avoiding these types of things than a player who after being hit a few times will start to pay attention.

I think the effects of unhappiness are enough as is in the game.
No need to pile on with more.

Sorry if that is not helpful.

 

[Azoth]

Pirate temporarily cuts trade routes.
You lose access to trade goods.
-3 happiness, civil war.

Civ declares war on you.
You lose access to trade goods.
-3 happiness, civil war.

Civ declares war on you.
Blockades city.
-3 happiness, civil war.

Privateers blockade city.
-3 happiness, civil war.

You conquer city.
It comes out of revolt while you are still at war.
-3 happiness, civil war.

All true.

That said, I think the 5 turns grace I suggested for Option 3 would minimize the damage.
You have 5 turns to end the blockade, reconnect your resources, rush-build happiness buildings, or find another resource supplier.
If all else fails, you can simply bump up the culture slider for a few turns.

I do think some more limits on the event trigger for civil war would be wise.

• The civil war event should only trigger once per civilization, or even once per game. No player or AI should lose multiple cities to multiple new empires over the course of a single game.
• The civil war event should only trigger in cities in which your culture is the majority. Civil war should not be a risk for foreign or newly conquered cities; they already have a chance to revolt and join their mother civilization. (Besides, a conquest victory would otherwise become a nightmare.)

I think the AI will have even a tougher time avoiding these types of things than a player who after being hit a few times will start to pay attention.

Also true. However, if civil war triggers only once per civilization or once per game, I think the AI can handle it.

 

[Xyth]

Triggering at 3 angry citizens is simply how the event happens by default, before I've made any changes to it! I'm not at all proposing or intending that it remain this way.

Similar to Azoth's suggestion, I think civil war would be best as the possible end result of an event chain. One advantage of the event system is that events can be set to require another event to have triggered previously. So the first event triggered is very minor in consequence, the next a little sterner, but if the same problems keep happening to the same city then the civil war could trigger. So, perhaps, Civil Unrest --> Civil Rebellion --> Civil War.

It's also possible to reset and delay events, so this gives us plenty of possibility for actions or quests for the player to reduce the threat of civil war.

Another idea I have was that rather than having a flat threshold trigger, it could be instead be incremental. For example, the chance of one of these events triggering is 10% for every angry citizen or food shortage. The sum effect of all this is that civil wars don't just happen randomly or instantly; the player (AI or human) is aware that one is brewing, and has the opportunity to do something to prevent it.

I agree that only cities in which you have a clear cultural majority should be eligible. Civics and religion could play a part in the civil war process too, though I think it's best to leave that till later and work out the fundamentals first.

 

[Howard Mahler]

Making hunger more important is useful for game balance; not so much for unhappiness which is already relatively more important.

An escalating series is much better.

Time to react, should this vary by game speed?

Having a small chance of something every turn means it is very likely happen.
For example a 10% chance per turn even if just one possible city can be effected, means the event takes place within 10 turns 65% of the time.
(The meaning of 10 turns varies by game speed.)

I am not a fan of adding more randomness.
(People vary in this regard.)

Are you sure that the AI can be taught to react properly given the extra potential danger.

Triggering at 3 angry citizens is simply how the event happens by default, before I've made any changes to it! I'm not at all proposing or intending that it remain this way.

Similar to Azoth's suggestion, I think civil war would be best as the possible end result of an event chain. One advantage of the event system is that events can be set to require another event to have triggered previously. So the first event triggered is very minor in consequence, the next a little sterner, but if the same problems keep happening to the same city then the civil war could trigger. So, perhaps, Civil Unrest --> Civil Rebellion --> Civil War.

It's also possible to reset and delay events, so this gives us plenty of possibility for actions or quests for the player to reduce the threat of civil war.

Another idea I have was that rather than having a flat threshold trigger, it could be instead be incremental. For example, the chance of one of these events triggering is 10% for every angry citizen or food shortage. The sum effect of all this is that civil wars don't just happen randomly or instantly; the player (AI or human) is aware that one is brewing, and has the opportunity to do something to prevent it.

I agree that only cities in which you have a clear cultural majority should be eligible. Civics and religion could play a part in the civil war process too, though I think it's best to leave that till later and work out the fundamentals first.

 

[Xyth]

Making hunger more important is useful for game balance; not so much for unhappiness which is already relatively more important.

Agreed. I don't think its realistic to exclude unhappiness from the equation - however, perhaps it could be, say, 10% for each starving citizen and 5% for each angry one. After all, people get pissed off at government decisions all the time, but if things are so bad that they're starving then they're fighting for their very lives.

An escalating series is much better.

Time to react, should this vary by game speed?

If there is any sort of time component it will most probably need to scale with gamespeed.

Having a small chance of something every turn means it is very likely happen.
For example a 10% chance per turn even if just one possible city can be effected, means the event takes place within 10 turns 65% of the time.
(The meaning of 10 turns varies by game speed.)

It wouldn't be checked every turn because the chance of it occurring would be zero for any city without any angry or starving citizens. However, if you leave a city angry or starving for 10 turns then, yeah, the likelihood of it triggering becomes less remote. But that makes sense and gives the player motivation to not let their cities suffer too long at a time.

Are you sure that the AI can be taught to react properly given the extra potential danger.

I think so, working on it.

 

[inasnapp]

how well could civics be incorporated into this system? I see a society that has a fundamentalist (with a large nonreligious minority) authoritarian regime that relies on slave labor to be more prone to a civil war than a democratic society that preserves its people's civil rights. Realistically this is probably more important than just the people are unhappy (even for a long period of time)
is it possible to have some (or all) civics make civil unrest/war more or less likely?

 

[Azoth]

Similar to Azoth's suggestion, I think civil war would be best as the possible end result of an event chain.

It seems we're thinking along much the same lines.

If there is any sort of time component it will most probably need to scale with gamespeed.

I don't disagree; I just want to point out that scaling reaction times to game speed is less important than usual. Most of the responses to chronic unhappiness: raising the culture slider, rush buying happiness buildings, trading for excess resources, breaking blockades, sending extra military police under Authoritarianism, switching civics, and spreading the state religion are time-insensitive. Only reconnecting pillaged resources and constructing happiness buildings with ordinary hammers are time-sensitive.

how well could civics be incorporated into this system? I see a society that has a fundamentalist (with a large nonreligious minority) authoritarian regime that relies on slave labor to be more prone to a civil war than a democratic society that preserves its people's civil rights. Realistically this is probably more important than just the people are unhappy (even for a long period of time)
is it possible to have some (or all) civics make civil unrest/war more or less likely?

Again, I think we should be careful that our biases don't get the best of us. Some of the most famous civil wars in Western history, the American Revolution and the American Civil War, were instigated by wealthy, educated landowners, many of whom owned slaves themselves. In many ways, they had a higher standard of living and greater personal liberty than any other people in their time period. They simply knew enough to assert their rights and had the resources at hand to secure victory in war.

I suggest that civics should instead limit the options available to players when civil war events are triggered. A democracy might be flexible enough to broker a compromise, as in Option 1 above. An autocracy might be ruthless enough to crush the rebellion before it starts, as in Option 2 above.

 

[Howard Mahler]

Building buildings is time sensitive.
(Random event destroyed one or more of your key buildings.)
Changing civics is time sensitive.
(Some civs for some parts of games are not in Authoritarianism, Slavery, etc.)
Building improvements is time sensitive.
(Random event destroyed one or more of your key improvements.)
Building units is time sensitive.
(To break a blockade or repulse enemy units sitting on key tiles.)

Also on huge maps one usually plays at slow speeds.
On huge maps it does take longer to rush units to a city, break a blockaade.

It seems we're thinking along much the same lines.



I don't disagree; I just want to point out that scaling reaction times to game speed is less important than usual. Most of the responses to chronic unhappiness: raising the culture slider, rush buying happiness buildings, trading for excess resources, breaking blockades, sending extra military police under Authoritarianism, switching civics, and spreading the state religion are time-insensitive. Only reconnecting pillaged resources and constructing happiness buildings with ordinary hammers are time-sensitive.



Again, I think we should be careful that our biases don't get the best of us. Some of the most famous civil wars in Western history, the American Revolution and the American Civil War, were instigated by wealthy, educated landowners, many of whom owned slaves themselves. In many ways, they had a higher standard of living and greater personal liberty than any other people in their time period. They simply knew enough to assert their rights and had the resources at hand to secure victory in war.

I suggest that civics should instead limit the options available to players when civil war events are triggered. A democracy might be flexible enough to broker a compromise, as in Option 1 above. An autocracy might be ruthless enough to crush the rebellion before it starts, as in Option 2 above.

 

[atirtanadi]

If you put in civil war, I think it would be wise to also include new diplomatic options: when a renegade city secedes, it should immediately begin asking other civs for help (like donating military units or declaring war). By the same token, a seceding city in a rival civ might come to you for help.
To further the diplomatic gameplay, I think that the seceding city should have different civics or religions from the parent civ. For example a seceding city might adopt democracy where its parent civ has monarchy. Or it might adopt Islam where its parent civ is Christian. Thus, the seceding city would have different diplomatic inclinations than its parent.

 

[Xyth]

This is turning into much more work than I originally anticipated so I'm going to put it on hold until after 1.18 is released. Keep the feedback coming though.

 

[Absolution]

This is a great idea, but it might cause a too crowded world.
Along with this "birth of civs" mechanic, I think a fall of civs system will have to be included to, to make it more balanced, and to make the domestic side more important in winnin the game.
It can be made in two ways:
1. These civil wars can bring a state down, and turn it into a developed babarian area.
2. The new civilization will start as Furious towards the original civ (the opposit of a colony), and will be gifted with some military units which will be able to weaken the original civ and maby capute many of itsery cities.
A bith of a new civilization in the middle of a global war can be very interesting.
The problem is that the new civilization might stay a bit behind, having relatively smaller territories, have a lack of diplomatical friendships.

I don't know if it was already mentioned, but specific technologies can be a trigger as well.
For example, after discovering Communism, some cities may demand State Property. If you refuse, chances for a civil war started in that city are increased. Same with Nationalism and Natiohood, Liberalism and Free Speech, Democracy and Universal Sufferage.. Every civic can be related to a specific trigger-technology.

I have once written a post in the Revolution Modpack forums which can fit here in someway:

I think that a great idea for the revolution mod would be converting citizens' happiness and conditions into demands for a revolution within borders.

For example, hungry cities will demand Environmentalism, a city with many religions will demand Free Religion, cities with national minorities (%) will damand Emancipation...
If the demands for revolutions are not fulfilled it will cause unhappiness and instability.
When it reaches high levels, it can cause a city to create a new civilization, and if another city will have the same demand, it will join that civilization.
Another cool idea is that when a city demands a revolution and the window with the options is shown, it will have names of famous revolutionaries in the description of the event. For example: "Thousands of slaves from the city of Hamburg are gathered under the leadership of Vladimir Lenin in a demand for Serfdom." - a demand to move from slavey to serfdom..
A bunch of names for every civics category is possible.

And if you combine the civil wars with BarbCivs it will be perfect.

 

[BBQ man]

A great idea one way or another. Randomness can be frustrating, but predictability can be just as bad.

 

[inasnapp]

Agreed. I don't think its realistic to exclude unhappiness from the equation - however, perhaps it could be, say, 10% for each starving citizen and 5% for each angry one. After all, people get pissed off at government decisions all the time, but if things are so bad that they're starving then they're fighting for their very lives.

Since so many revolutions/civil wars began because of starvation, it should be included. I am concerned though that in my most recent game, my starting location was very dry (primarily plains, and a good sized desert-which does have some oil and incense-) and rocky, overall it was a really nice location; but the areas I was able to build larger cities, (near what grass land and flood plains I had, I still managed to only be able to produce odd amounts of food in those cities. a 10% modifier for starvation is probably fine, but I'm concerned that when a civ's food production is limited, once you end up with a city with 1 left over food, it will perpetually starve and regrow, typically this should be avoided anyway, but sometimes that's not realistic. How would you balance this? When you have 10 or more turns of a slow leak, civil war in that city starts to become a very real possibility, even with the low chance it happens on any single turn.