Gov.model - next level (long)

[Mark_Everson]

Rodrigo:

Your judgment about the limited uses of political power in the model is probably reasonable given what you've already read. That's the fault of our spotty documentation . However, who has the political power in the civilizations should have wide-ranging repercussions on almost everything in Clash. (I'm pretty sure Hrafnkell agrees with me on this, since we have discussed it a lot, a ways back) In conjunction with the culture itself, who has the political power will affect the following factors:

What the people choose to build with their own money (not player controlled) in the economic model (if religious authorities have a high status in the government, many more religious buildings will be built by local governments and the people themselves)

What technologies or social ideas the people are the most interested in

The quality and behavior of military troops (if the military pulls the strings in the government, then higher-caliber individuals will be attracted to the military than if they were reviled)

I'm sure I haven't covered all the inter-connections that we might be able to do. But I think you can get the idea of what we are shooting for from the short list.

[roquijad]

The type of interactions you state, mark, look alright and they're reasonable too. But I don't see limits on ruler's rule. Let me ask it another way: what's going to be the difference in clash for a dictatorship and a democracy at the govt lvl? Is there any change in players actions being in some form of govt or another? A player can call for a "religious war" being in a communist govt?

[Mark_Everson]

Rodrigo:

I could make up something on the spot on exactly how it should work, but the truth is the model isn't there yet. Maybe Hrafnkell has a better idea... It could also be handled as an extension of the social model.

If you want to state your ideas on how it would work they'd be welcome.

[roquijad]

To continue my effort to change some things about the govt model, I've re-read the model (more carefuly now) at the web page and I have some Qs for the clash team (given the answers I'll make a precise critic on the model):
1) Is the principal class a class for its own? It's defined as the ruling class, so I wonder what happens if you have a fundamentalistic govt.... PC=RC ?
2) Can you give a definition for "contribution". In chapter 6 you can find how it's computed, but... what is it?
3) The type of govt you have is related to the pol.power distribution as stated in chapter 8. So, since pol.power struggle is mostly automatic, this could lead your civ to change govt type without the player's intention, because a class can gather enough power to put your govt into a new category (new govt type)... is this correct?
4) The main reason to introduce pol.power is to compute the Anti-govt effect and then chances of rioting or changes in productivity/happines/economic effects. So, a high pol.power doesn't give real POWER to the class like altering high govt decisions. Is this correct?

[Mark_Everson]

Hi Rodrigo:

(These need to be taken as provisional, since I can only give my views. Hrafnkells may be somewhat different)

1) Principal Class is a class of its own. In a true Fundamentalist govt (where religious forces Really determine policy) PC power might be 20% and RC power might be 50-60%. So the player Can't do whatever they want. They can only try to do things consistent with what the religious class will allow. In a state that's nominally Fundamentalist but is basically Totalitarian hiding behind clerical robes the power might be PC (strongman) = 60%, RC = 20-30%.

2) Contribution is an absolute measure of how much the class contributes to the power of the civ. It is then modified by cultural factors. If you haven't followed the link in the model on the web page to http://people.mw.mediaone.net/markeverson/govt_v2.htm , you should probably look over that one too. Just search for "contribution" on it, and you'll see it discussed several times.

3) Yes. Although a player can either take direct action to try and maintain a govt. form, or order an AI advisor to do it.

4) Incorrect. See section 3.3 of the model. However we don't know in detail how to handle this yet...

Mark

[roquijad]

I've been doing lots of reading on the govt model. I'm not ready yet to do a formal critic and some proposals, but there's something I'd like to mention beyond this particular model:
Reading mark's first proposal for a govt model (URL above) and the first interactions with Hfrnkell (damn, that name is tough! ) made it much easier to understand things. If you're really interested in encouraging people to participate at this forums, a sort of "How we've come here" text is needed for every model. Just a brief, concise text pointing the major decisions along the way.

Anyway, that's the sort of thing I can help with (in concrete) if all info (discussions and proposals) is available.

Rodrigo

[Mark_Everson]

That's a good suggestion. We already have an historical list of threads for each area on the web page. If they were dated (by the thread start) and in chronological order would that be enough? If so, please go ahead and suggest it on the web page thread.

If that wouldn't be enough... Well, then I guess the next person doing each voyage of discovery through each model's history could do a great service by writing up a more detailed account of that history.

[roquijad]

My research about the govt model is finished. I have lots of comments and suggestions regarding detail-level topics, but I want to concentrate 1st in general things that IMO need to be changed/included. Please read them very carefully since they're the outcome of a long meditation on the subject. As you'll see, they don't represent a "get rid of your model and take this new one" option, but a serious critic of current model aspects that lead to proposals that are supposed to be implemented ON Hrafnkell/Mark's model.

1. In the current model, pol.power relation to actual power is vaguely achieved. Some decisions the ruler takes can be over-ruled by a province or undertaken in a unefficiently way if the primary class has low pol.power in it. This is not realistic and as I said in an old post, pol.power should appear as an excercise of power, that is, a class holding pol.power should be able to have a share in govt decisions. What's the problem with the current model proposal on this? Suppose you're in a democracy (a pol.power distribution that represent a democracy, to be exact). Suppose the president has the constitutional right to close borders for immigration. Since PC pol.power is low (the pol.power rests mostly on other classes), provinces will disobey and borders will be kept open. This is largely unrealistic, since the president is granted by constitution to take those decisions. One may respond to the above case, defending the current model, saying that what the president/player is allowed to do in a province must be in line to what classes with pol.power like, and so, if closing the borders is something that classes with high pol.power want, then the president's order will be undertaken, but this is a tricky answer since the model doesn't include a list of things a class wants and they react to drawbacks of pol.power only (or mostly, at least). The way I see it, if PC has a low pol.power it should be reflected as limits to rulers decisions, but never as inefficiencies in the way decisions are undertaken. What decisions are able to be limited and how? Let me leave this for later...

2. The current model lacks of an "ideology" concept. This is very important since the arrival of democracy and communism depend on it strongly, but it exists for all forms of govt. Take fundamentalism for instance: the RC is strong and has most of pol.power. LC may have some share, but there's no way to know if they support this type of govt or not. In real (nowdays) fundamentalist countries most of the common people support their system and that means they want the RC to have a large political power. In game terms this means LC won't just look at its contribution and its traditional pol.power share to define their Anti/Pro status, but should also look at the RC share. If it's low, they'll be more Anti, because they want a fundamentalist govt. In a more general case, a class will look at the ENTIRE distribution of pol.power to determine how Pro/Anti will be. Think in democracy now. The only way it was ever possible was through extensive education on classes about its benefits (propaganda). You had to convince the rich people that giving common people political power would make society a better one. This means classes started to want a particular distribution of pol.power. A last example: How can we model spanish civil war or Vietnam war without ideologies? So, you need to define an ideology in the model which is nothing more than a desired pol.power distribution. If a class support a particular ideology, then it'll be more Anti or Pro according to how much the actual pol.power distribution reflects that ideal. An ideology should also include "ways to govern", things govt is expected to do. In the communist ideal, economy should be centrally planned, FE. This is, as you can see, a missing link between models. Mark's econ model may reflect a communist economy, but how is it connected to the govt model? This is the way! IMO, ideologies are not a simple add-on to the model, but a necessary feature that lets you simulate interesting and relevant struggles in world history.

3. I also wrote in an older post that it'd be good to include govt-level variables like "level of people's freedom". I really believe this is something missing in the govt model. Just to give an example, there have been dictatorships in southamerica supporting US-like free-market. In Clash under current govt model such a dictatorship could do better than a democracy, since you have the benefits of a free market and the benefits of a high PC pol.power (then achieving most of what you want). This is wrong and the reason is exactly because of the lacking variables telling you about how free is people or how much personal rights they posses, which are the things that make a dictatorship eventually ungovernable. For the rest of this post, let me name these type of variables "Behavior Variables" (BV).

4. Now let me link the things I've said so far. Suppose you have, for every form of govt we know, an ideology for it. It should include a distribution of pol.power among classes AND a "way to govern" (WG). The WG should be defined as desired levels for: BV, tax rate, conscription, public investment (in all areas) and type of economy. Now, for a given point in time there's a number of available ideologies. For each class, and in the same fashion you divide it into Anti/Neutral/Pro, you also divide it into those supporting each of the ideologies present. Such division (and how it moves overtime) should be determined, among other things, by a level of propaganda each ideology has. Now you can do the following:
a) Determine the class ideal. The class ideal is the simple weighted average of each portion of the class supporting each ideal. The level of anti/pro for each class is then computed according to how the real pol.power distribution reflects the class ideal, the contribution and tradition (moving average of past pol.power distribution).
b) Using the distribution of people supporting each ideal and the distribution of pol.power, you can do another weighted average to define the WG the society wants as a whole. FE, if using a) a class wants a tax rate of 20% and its pol.power is 10%, the final tax rate in the civ will be set to a rate influenced by these numbers. The player (PC) also had its participation in a dialogue box where he/she stated his preferences for the WG and it was included in the overall weighted average (So, he/she also propose a tax rate but the final one is determined using all pol.power distribution). Notice the tax rate (like any other variable in the WG) is SET to the value defined by the weighted average. The player is not allowed to define the tax rate alone since he/she doesn't have a complete power. This is a better way to reflect pol.power not held by the player instead of letting the player to set the tax rate and then cumpute it inefficiently in provinces with low PC pol.power.
c) As in b), but now the people's desires for distribution of pol.power are determined. In other words, classes ideals are merged to define the current pol.power distribution. Again, the PC (player) participate in the weighted average with its own share of pol.power. Notice that this particular calculation replaces current model mechanism to determine how pol.power evolves. In the current model pol.power changes mostly by the rule "each class can get more pol.power each game turn in a magnitude equal to 10% its current pol.power". The proposal here is, IMO, more consistent because on one side the current rule isn't able to determine in a straight forward manner what class loses power when other earn it, and in the other side the pol.power distribution reflects people's desires for the entire distribution. Furthermore, with this new system for pol.power evolution, classes actually USE their power (since it's used in the weighted average) to influence pol.power distribution in a way that's consistent with their desires.

Notice that in b) and c) above you can actually see classes exercising their power, because they have influence in govt decisions about things like people's lvl of freedom, tax rate, type of economy and also on the distribution of pol.power. The player, on the other side, has its decisions limited to his/her share of pol.power, but there's no inefficiencies on what he/she does. It's possible to imagine other types of limits like ruler's possibility to order troops to implement a martial law in a province, but I think that type of limits are of detail in nature and can be discussed later.

5. What's the benefit of the system proposed in 4? The main benefit comes from making the govt model able to simulate thing it can't now, like ideologies struggles and how they alter the civ's political picture. Also, there's a more consistent way of treating pol.power, since a class having pol.power should indeed be able to influence govt decisions, instead of making player's decisions inefficient, which is IMO closer to the idea of corruption. Now the question is how expensive in computational terms this proposal is. As far as I can see, one good thing about ideologies is they're unique. This means you only need to store the ideology attributes once per ideology and it's valid for the entire planet(and there isn't much info to store too). I think it's good to emphasize that uniqueness of ideologies doesn't mean that every civ embracing the same ideology will behave the same. That's because a civ never really embraces a single ideology because govt variables such as pol.power distribution and WG that a civ has is a mix of the same variables from those ideologies supported by its people (with different shares of support for each civ). Back to the Q about comp. cost, I also see very cheap implementing BV, specially if they are only defined at a govt level instead of having different BV levels for each province. So the only drawback comes from having to split a class into segments that support each different ideology. Then you need to store for every province and for every class in it a vector of numbers (its length is equal to the total number of ideologies in the game) meaning how much people support each ideology. This may be expensive. So, is it worthy? Is the cost lower than the gain in simulation capability? open question....

6. To finish, I see little connection between the govt model and the rest. Particularly, I feel a closer relation should exist with the social model. I plan to give it a thought later but 1st I want to see your comments on this post.

Mark: It'd be good to have Hrfankell involved in the discussion about ideologies and the rest topics I've mentioned. Can you email him....?

Rodrigo

[Mark_Everson]

Rodrigo:

Your suggestions are very good in general IMO. I have in fact thought about them a fairly large amount, although I don't have time to write much about my thoughts. I'm currently racing to get the demo out ASAP, so I'll delay detailed comments until after that is out. I know I also owe you some information on the Econ model, that will come then also. Now my comments...

First, and I believe most important, we Can't just layer your suggestions on top of the other model. IMO the old model was Already way too complicated. So I believe you should think in terms of potentially changing the model, and Simplifying the model. I think we need to have at Most three or four Big principles in the model. If this were just a game about the internal governing of a Civ, we could afford to make the model fairly complex IMO. But given that government is just one small part of the game, the model needs to be Both compact and relatively easy to understand. This process is going to end up necessarily in us stepping on each other's toes a little bit. But the alternative to having honest disagreements, and making hard decisions, has to be balanced against the chance that we will come up with something that is completely unfathomable to many players of Clash. So with that in mind, I'll address your points.

I think your point 1, making the Civ's responses to the ruler more homogeneous, is the right way to go. I don't agree with all your arguments, but I think for the sake of simplicity and player understanding it is clearly better for the Civ to respond in a monolithic fashion. We will have to use other ways to simulate the diversity present in things like feudal civs and civs composed of a number of city-states.

Your Ideology idea is Really interesting. If you have the time to work up a couple of detailed examples about how it would work, especially for extremes in forms of government, I'd like to see them. I think you are right that with a limited number of ideologies, may be 15 to 20, we could cover most of the big movements in history. It also occurred to me that ideologies like "self-determination" could be what precipitated things like the American Revolution.

I'd dispute your point number three at length, but I don't have time. Essentially, in my original model I lumped political and economic power together to simplify the system. A free-market dictatorship would, I believe, seriously under-perform because of the large penalty of corruption in the model, which I have described in several places. But this isn't a major point, so I'll continue on with the rest of your ideas.

I think the rest of what you discuss is the core of a Really good model. I urge you to come up with some examples of how it might work, just to make sure we are on the same wavelength. Having read your outline, the things that come to mind that I think you need to illustrate are:

How does a huge multi-ethnic empire work? If all the provinces respond the same to the ruler's decrees, then what makes it unstable? (As most have been throughout history)

How you capture the potential for innovation that's present in a feudal society? IMO it is Because the ruler is not very powerful that it becomes much harder for the ruler to Screw Things Up too badly in such a society.

How you handle a Civ that is essentially a collection of mostly-independent city states like the Greeks were during the classical age? Do they have to be 15 little civs?

What power does the player have to guide the evolution of the government? What parts of this are the most fun for the player?

Like I said, you have a lot of intriguing ideas. I'd like to see some details put on them.

Thanks for the great contribution to Clash!

Mark

[roquijad]

Mark:

I'm a little bit concerned about something you wrote: "we Can't just layer your suggestions on top of the other model. IMO the old model was Already way too complicated. So I believe you should think in terms of potentially changing the model, and Simplifying the model".
It'd be fun and challenging to me to actually make major changes in the current model, but I'm affraid that'd be unproductive. I mean it's maybe too risky to make major changes when we already have something that may be complex but it's solid enough to work with. As I said in my last post, I was thinking of introducing my proposals in the current model instead of changing the model. I believe it'be good if you explain your position about this more extensively. Something tells me you had doubts about current model even before my last post... did you?

About ideologies, I'm glad you like the general idea of it. I really think it's an important feature to include.

I can get to numbers and show how feudal and other cases may work with the introduction of "ideology". It's pretty clear in my mind at least. However, in order to avoid an unuseful effort on this, I better wait for an answer to the above question. That is, should I create examples thinking in ideologies ON the current model or, if you're more in a position of making major changes, should I think beyond the simple introduction of ideologies....

One more question: How involved is Hrfankell at this time?

Rodrigo

[Lord God Jinnai]

One point i'd like to dispute... A dictatorship wuth a free market economy can do quite well...look at how China was before the Asian Crisis, and prob would be if it weren't for Indonesia. Anyway IMO unless the government puts direct input into the economy through taxes, price controls, gov. run bussinesses/controled monopolies, etc. It shouldn't matter what type of government u have to determine the economy, the eception being an anarchy.

[Mark_Everson]

Rodrigo:

I stated my personal opinion on the model's complexity and where we should go from here. It is of course a Group decision what the model should look like. Perhaps by the nature of the changes you propose, and the parts of the old model that would Automatically be superseded it would be ok. I'm just getting uneasy about a "mega-model".

Why don't you give examples in whatever way you are comfortable with, and we'll take it from there.

LGJ:

Dictatorships are frequently good at catching up in technology for limited periods of time. But overall IMO, the corrupting influences of complete rule by a small number of people usually manages to screw things up royally. I think the long-term historical record speaks very clearly on this. So, perhaps an enlightened despot with a market economy could do well, but the rulers never stay enlightened enough for very long.

[roquijad]

Mark:
I'll work some examples and maybe include some ways to simplify current model. Give me some time.

On the dictatorship thread:
I had the unpleasant experience of living under a dictatorship. During 1973-1989 Chile was ruled by general Pinochet, an extreme-right character. There are a couple of things that may help us about modeling a dictatorship from my experience:
1) It's very important to notice that a dictatorship is in no way a rule of the military for the military. Pinochet hated communism and very soon in his reign he put a number of economists (most of them educated at the US) to implement a solid free-market economy. This was helped and supported (that support also came at the moment of the military coup) by the right conservative parties (the rich people, landowners, company owners) so his rule was actually a co-rule with the "upper class". One may see it in the other way: it was a rule by the upper class using the militay as a tool to preserve their power. My point is a dictatorship isn't necessarily ideology-free. So, if one just see a dictatorship as a govt made to serve a military elite, we may lose ideology-related dictators like Pinochet or Mao-Tse-Tung with his communist ideal. Furthermore, a dictatorship always serves an elite. The elite may take the form of a class (religious or upper class) or a party (nazis/communists). This link between the military and some other part of society is something to keep in mind when modeling a dictatorship.

2) Since 1985 aprox. Chile has had an excellent performance in economy with very high econ growth, low inflation and low unemployment, mostly achieved by Pinochet's implementation of free market system. Corruption never had an important role, since Pinochet's freemarket ideal made State size small and also its participation in economy was strongly reduced.

3) In 1989 Pinochet asked the people "should I stay for another 8 yeras?" and he had a huge "NO!" for answer. The main reason: A nation tired of killings/kidnapings/torture/abuse and general low level of personal freedom. The only influence of economy in people's frustration was the low public investments on poorer classes and this was a govt policy instead of a corruption effect.

I agree corruption plays a role in dictatorships since the people and their organizations are limited in controlling public funds, but I also believe corruption is highly related to State size and that's why we see it much clearly in socialist dictatorships, since socialism NEEDS a big State. Dictatorship on its own doesn't come with built-in high levels of corrpution, IMO.

Rodrigo

[Lord God Jinnai]

Furthermore, a dictatorship always serves an elite. The elite may take the form of a class (religious or upper class) or a party (nazis/communists).
-----
Prob the only prob i have with that whole paragraph is the above sentance. Although its generally true, its not always true. There is a concept known as "Popular Monarchy/Dictatorship" in which the ruler goes by what most of the common people want since they are much more numerous. Marx was this kind of ruler and Lenin was to very much extent.

I agree corruption plays a role in dictatorships since the people and their organizations are limited in controlling public funds, but I also believe corruption is highly related to State size and that's why we see it much clearly in socialist dictatorships, since socialism NEEDS a big State. Dictatorship on its own doesn't come with built-in high levels of corrpution, IMO.
-----
Corruption plays a role in any gov. model, be it dictatorship, republic, democracy (although in a pure democracy such as ancient greece, this may be quite hard to corrupt a majority of the pop), etc. The only one that may have almost none-existant would be a fundimentalism where everyone basically worshiped the ruler and his word was considered absolute and perhaps divine.

[Starfighter08]

quote: Originally posted by Lord God Jinnai on 12-16-1999 11:12 PM Corruption plays a role in any gov. model, be it dictatorship, republic, democracy (although in a pure democracy such as ancient greece, this may be quite hard to corrupt a majority of the pop), etc. The only one that may have almost none-existant would be a fundimentalism where everyone basically worshiped the ruler and his word was considered absolute and perhaps divine.

Even a theocracy is not without corruption, political murders etc. If you happen to read "I discorsi" by Niccolo Machiavelli you will see that the different "prarties" actually noble dynasties used every possible mean to have the one chosen as the new pope that they supported. And then there was "Simony" (German: Simonie. The Bible says that a certain Simon tried to buy the capability to heal people from Jesus) which is to pay a superior in the church to be chosen as the new bishop.

BTW Some gov'ts especially republic are quit Civ2 like (If you are the speaker of the parliament, who is the administration?)

My two cents on gov'ts (I'm quite "write happy" today)

ORGANISATIONAL FORMS

Centralized: decisions are all taken in the capital city

Provincial: hu? like the Romans?

Federal: Local gov'ts have power to take their own decisions (for local issues) and levie their own taxes. Federal gov't takes decisions for the whole civ and levies taxes too.

Feudal: Power depends on landownership. The King, who owns all land lends a part of it to say dukes in excange for military services and of course governing their territories. The dukes lend some of their land to say counts also for military services. If the king issues a call to arms, king, dukes and counts raise their armies (whereby the count's armies are part of the dukes' which are part of the king's. Later on vassals did serve less in the military and paid taxes instead.

City States: It's actually an alliance of politically and economically independent city states (like the one Athens led for some time, I think Athens led the alliance's Navy while Sparta led the land based troops during the persian war).

Tribal: I think because of the small size of such "states" the chieftain and the highest priest led the civ directly.

ECONOMIC STRUCTURES

Barter: changing goods for other goods or services (services were not so common because it was a self sufficient economy).

Free market: Gov't is not involved much. It levies taxes and tariffs and spends money for defense and police, roads and few other things.

Manoralism: A manor had 9 families working on fields and 1 family for "production and services". The economy was self sufficient but surpluses were traded for things needed with other manors. Growing markets/cities led to a specialisation of the products of the manors ->more trade. The landlord also held soldiers for protection and to serve his lord (FE count).

Mercantilism: Foreign trade is strictly controlled by the gov't. To ensure a large influx of money worth of the exports must be higher than that of the imports. Ergo resources are imported (export of them and of food is forbidden or has prohibitive high tariffs) and finished products are exported (import of those is forbidden or has prohibitive high tariffs).

Socialism: The state owns all the production sites and also controls foreign trade. You could say that all the citizens are employees of the state.

Keynsianism: Welfare state which in times of crisis (FE high unemployment) raises gov't expenditures to fill in the lack of demand on the markets (usually with infrastructure building projects).

FORMS OF GOVERNMENT

TRIBALISM (actually Tyranny or Despotism) is merely a rule by brute force.
Organisation: Tribal (surprise surprise)
Economy: Barter

MONARCHY greek: rule of a single one, ofte with "divine" justification
Organisation: Centralized (absolute monarchy Louis XIV style)
Provincial (antique, if I take your meaning)
Feudal (medieval times)
Federal? (for (modern) constitutional monarchies?)
Economy: Barter, free market, manoralism, mercantilism

REPUBLIC latin: something like "public issue" elected gov't
Organisation: Provincial (FE Romans; was actually an Aristocracy)
Federal (FE BTW are there some modern republics comparable with the roman one? )
City States (FE Athens and its allies during presian war?, medieval Switzerland)
Economy: free market, mercantilism?

THEOCRACY greek: something like rule of the clergy, of cours by divine justification
Organisation: Provincial (antique?)
Feudal (church in middle ages)
Centralized (can I say Iran?)
Economy: free market (again Iran?), manoralism (middle ages), mercantilism?

DEMOCRACY greek: rule of the people
Organisation: Centralized (FE modern France)
Federal (FE USA, modern Switzerland)
City States (again antique Athens?)
Economy: free market, mercantilism? (modern Japan? I'm sarcastic I know),keynesianism

DICTATORSHIP modern version of Tyranny/Despotism, one single person has all power (could also be the military)
Organisation: Centralized (dictators don't allow people to do something on their own)
Economy: free market, keynesianism (like fascist Germany/Italy), socialism (Stalin's USSR)

OLIGARCHY greek: something like rule of a few people
Organisation: Centralized (Post-Stalin USSR)
Provincial (antique form?)
City States (antique form?)
Economy: free market (antique version or a state with military junta->could also be keynesianism), socialism (Post-Stalin USSR)

Guess that was it. BTW for a communist state see dictatorship and oligarchy gov'ts.
[This message has been edited by Starfighter08 (edited March 18, 2001).]