A short overview - Let me know what you think, and what needs to be added.

The goal of the diplomatic system in Clash is to provide as many of the diplomatic options that real states use as possible without adversely affecting gameplay. There will be a large array of items, issues, and treaties that can be exchanged or agreed-upon between states. There should be provision for multi-state agreements, such as mutual-defense alliances. Another important issue is when one state fully or partly controls another or its actions in some sphere.

Treaties or agreements can be formed between any number of states. To keep things simple, treaties that involve more than two states must handle all members of the treaty on equal footing. So, for instance, five civs can participate equally in a mutual-defense pact against another civ, or another alliance. Three civs could participate in a most-favored-nation trade pact.

Treaties or agreements can include an unlimited number of clauses.

Treaty components (possible clauses) include:
1. Change of basic diplomatic status (peace, war, cold-war etc. A change in status that makes the diplomatic state between two parties more hostile can be undertaken by either party if they have the internal power within their civ to do so. A change in status that is less hostile requires agreement of all parties.) The diplomatic state can be collateralized
2. Cash (either in lump-sum or in an installment plan) or loans
3. Territory transfer (either now, or at some future date)
4. Technology (although, as discussed in the tech model, trade of a technology does not necessarily result in the acquiring civ immediately having that technology available.)
5. Ceding control of military units (usually temporary)
6. Trade status (from most-favored-nation up to embargo)
7. Internal matters, such as treatment of religious or ethnic minorities (this one will be tough)

Threats And Unilateral Actions

Threats in Clash need to be every bit as nuanced as treaties. The player should be able to make a threat using all the components above. In addition, civs with sufficient power in a controlled area can unilaterally create another civ. The created civ has whatever properties the creating civ desires. However, if the foundation of this state is unrealistic it will quickly change radically from the form in which it was created.

Further notes on above

1. Diplomatic Status

Here's a sample list of diplomatic states available between civs. These are not necessarily well thought out. The values associated with each name are modifiers on a -10 to +10 scale that I have used in the existing code on the diplomatic system. The numbers generally indicate the "strength" of the relationship between the powers. I'm pressed for time right now, so I will only give brief descriptions of these.

Diplomatic State Constants

Vendetta____= -10; State Seeks To Annihilate Opponent, And Its People
Total_War___= -9; Seeks To Destroy Opposing State, But Not People
War_________= -7;
Limited_War_= -5; War Whose Aims Are Respected In Some Way
Cease_Fire__= -3;
Cold_War____= -2;
Contact_____= -1;
Peace_______= 0;
Co_Agression =1; States Cooperate To Attack A Third, But With No Other Alliance
Cooperation_=2; A Deep, Peaceful Relationship, With A Long History
Def_Alliance =4; Defensive Alliance
Off_Alliance =8; Very Strong Offensive/Defensive Alliance
Same_Ruler_=10; Case Where One Civ Is Completely Ruled By Another

Any diplomatic state that is agreed-upon between two or more parties can be collateralized for any mutually-agreed-upon period of time. For instance, a peace treaty between states might be guaranteed by both to the tune of 1000C over a period of the next 20 turns. Any party that unilaterally breaks the peace gives up the collateral. Whether the collateral becomes the possession of the offended party, or simply disappears is also determined in negotiations. Collateralization is meant to symbolize things like hostage-holding, marriage alliances, and other ways to guarantee treaties where the game does not have sufficient depth to include the particular factor. I think collateral could become standard for treaties, since it ensures that each party will take the treaty seriously. This feature seems to work better in the ancient and medieval world, than in the modern. We'll have to address it as a play balance and realism issue at some point.

6. Trade Status

The player has control over taxes on trade, and can essentially move trade from a free-trade basis to a trade embargo by changing tariffs. In my opinion in would be way too messy to let the player handle tariff levels for every special commodity with every civ. So instead, I propose to let the player just raise or lower the overall tariff level with respect to each other civ. Trade status in the economic model would simply be handled by adding a number to the average tariffs between 0% and, say, 999% to get the tariffs for a given commodity with respect to a particular country. A quick example. I have instituted tariffs of 20% on steel, and 50% on textiles. So a most-favored nation (a + 0% modifier) would get these same numbers. A country I'm having a trade war with (+ 50% modifier) would suffer tariffs of 70% on steel, and 100% on textiles.

7. Internal Matters

Negotiations on internal matters will be unique to Clash within the genre. Because we are actually modeling internal dynamics of the government and culture these issues have real meaning in Clash. Threats like "if you do not free your slaves, we will raise tariffs on your goods by 30%" can be made.

A few quick notes on other issues

Governments will be constrained in their actions by the internal politics and power relationships within their civ

Reputation will matter

Civs will know how to play power politics due to our good AI

AI civs will know the relevant military realities, and will not provide tribute to states that could not hurt them in the slightest way

Diplomatic interactions will be colored by the culture, and previous history of the states

1) If you conquer someplace that is more trouble than its worth to keep (restive population, strategically indefensible, etc.) you can turn it over to the natives.

2) After the rise of nationalism there are lots of places you Can't keep long term if you're a democracy. Eventually a referendum will be forced upon you, and you'll have to give it up. This way you can accelerate that process.

Good points. I said nothing about diplomatic AI here, but I've thought a lot about it...

All:

Kull is going to be the new Duke of Diplomacy, and will I hope, take over here. Good Luck Kull.

Good point, negotiations should include optional time limits for treaties, or time limits for collateral.

On the Cold War status... I just meant it to mean a level of hostility between general peace and a cease-fire in a hot war.

Some good ideas from the Civ3/general diplomacy thread (made by hannes I think)
Add to the list of treaty items...

8. Demilitarized zones
9. Arms reduction treaty elements (fe, my army will grow to at most 20% more powerful than yours.)

I've looked at Mark's Draft Model and reviewed a number of the comments. In no particular order, here's a few of mine:

Treaty Duration - In modern times, the duration of a treaty is typically a function of government type. Here's a couple recent examples. Iran: Ally of the US while Monarchy, Enemy under Fundamentalism; Nicaragua: Dictatorship=US Ally, Communism=Enemy. The world is rife with examples. So this should be the fundamental determinant in the modern era. (Other than being on the receiving end of a surprise attack!) The same rule should apply in the ancient and medieval eras, but we'll need something else, too. The more I think about it, Hrafnkell's time limit idea makes sense. If you look at in terms of "Ruler Lifespan", than it's historically realistic. Nothing was more likely to change a treaty than the death of one of the co-signers. This is particulary true of non-representative governments. Which suggests an even better rule of thumb for Treaty Lifespan:

1) Treaties always come up for review whenever there's a change in government type.

2) Non-representative governments experience a "treaty review" every "X" number of turns.
(The determining factor for "X" should be entertaining game play, not realism.)

3) A variety of factors will impact the treaty review process. (ie Will the treaty be extended, upgraded, canceled, etc.)

4) Unilateral cancellations are always an option, but carry their own set of risks (ie. Your reputation is shot, other treaty partners may insist on a review, etc.)

MAD - Good concept, but it should be tied in to mutual possession of an extinction level quantity of nuclear weapons. Looking to history, the possession of a few nukes on one side didn't end the world. Likewise, the earth probably could have endured a nuclear exchange between the US and the Soviets in the early 50's. It was certainly considered during the Korean war. But by the 60's, forget it. That's when MAD became apparent to all. So there is a role for MAD in Clash. (What about a madman?) The tricky part is knowing when it kicks in. I have some thoughts I need to work on, but comments are welcome.

Cold War - The Name: This is a fairly new term when considering the whole history of the world, and it comes with lots of baggage attached. For that reason, I'd prefer a different title for this level of inter-state hostility. Ideas?

Cold War - The Proxies: Should Proxy Wars be a part of Clash? You know, "my troops against your buddies" like Vietnam, Korea, and Afghanistan; "my buddies in one country against your buddies in another" like Israel-Arabs and Pakistan-India; and civil war scenarios like Cuba, Angola, and Nicaragua. Maybe this is what Mark means by "Limited War".

Collateralization - To quote Mark, "This feature seems to work better in the ancient and medieval world than the modern." One way to handle this would be different collateral options for different eras. The earlier eras featured money, marriage, hostages, and territory exchange while the modern has arms treaties and trade agreements. Ideas to help flesh this out are welcome.

"Trust but Verify" - This arms control axiom will have to be incorporated into the treaty process. Without going into extensive detail, I can see a link between treaty type/level and the amount of information a partner has about your society, military intelligence in particular. Let's look at how this might work based on different "Diplomatic States":
1)Peace: I know exactly how many units he has and vice versa, but no details.
2) Co-Agression: Same as above. Perhaps also a sharing of "War Zone" unit info. (As defined in #4 below)
3) Cooperation: We share numbers and locations and generic types.
4) Def. Alliance: Same as Cooperation but includes detailed unit specs for those stationed on a mutual border. (ie between the allies) If we are at war with the same enemy, we share unit info on those which are in the "war zone" with that enemy. This could be defined as "X" number of hexes from the "front".
5) Off. Alliance: I have complete details on his military and he on mine, except injury and preparedness data.
6) Same Ruler: I know everything about his military, he just gets total numbers on mine.

"Stand down your bombers and I'll give you the MTV tech advance."

The info available could be :
- X units of type Y
- N TaskForces of type T
- New TF created near B's borders
- Commander C recently hired by Civ A

Maybe sometimes these are made public thru the press, if the press of Civ A has sufficient freedom to act.

OF COURSE, one needs to be able to lie to others..

To have something different than CIV, I could imagine showing a kind of throne room (or the equivalent for later gouvernments) from the side and having the ruler and the emissary of the AI having a a conversation.