Coups and Power Struggles

In the Cold War, governments rarely fell to elections alone. Officers moved against presidents, revolutionary movements stormed capitals, and foreign intelligence services paid for both. Atomic Dawn models this with a single rule: every seizure of power goes through a flashpoint. A faction can't grab the state on its own one day. A plot has to ripen, a power struggle has to come to a head, or a rising has to boil over, and you can see the danger coming if you're watching, or if your spies are.

There are two main forms. A coup plot is secret: one faction plans a strike at the top. A power struggle is public: a failing government faces several factions competing to replace it. Revolutions and uprisings are a third route and are covered under Flashpoints and Crises.

Coup plots

A plot is a hidden flashpoint with the plotting faction on one side and the government on the other. Its tension is how ready the plotters are. It rises with the plot's own pace, with the plotters' agitation, with the government's weakness (whichever is lower of its Legitimacy and its popular support), and with every foreign backer behind it. At 100 the plotters strike.

Where plots come from

• At home. An armed faction that is highly agitated, prefers another party to the one in power, and serves a weak government may start plotting without any foreign help. In the 1946 scenario that means agitation of at least 60% and a government strength of 40% or less. Even then, a country in that state has only about a 3% chance a month of a plot starting.
• From abroad. A foreign power with a stake in a discontented faction can use Back a Coup. Only the faction's leading sponsor can do this, meaning the backer with the largest stake. See Foreign Backing for how that stake is built and how it lapses.

Plots are rare by design. A country can have only one live plot at a time. A faction that has plotted can't plot again for ten years. A country that has survived a coup attempt becomes Coup-Wary for five years: it gains Counterintelligence (+15%), and plots against it ripen at half speed.

Secrecy and discovery

Only the plot's backer, anyone who later joins it, and anyone who uncovers it can see it. Everyone else sees nothing, on the map or in any list. Each month, countries that don't know about a plot roll to uncover it. Their chances are better the deeper their Espionage penetration of the target or of the backer. A foreign country that uncovers a plot can join it, back the government, send arms to either side, or Warn the Government.

The target government also rolls each month, using its counterintelligence. It is much better at catching its own officers when its Internal Security policy is harsh. A regime with secret police or mass surveillance catches most military plots. A democracy with civilian policing is often taken by surprise.

An exposed plot loses most of its momentum. Tension drops sharply, the plotters lose much of their agitation, the backer loses prestige, and the government's opinion of the backer collapses. The plot can carry on in the open, but it now strikes a government that is ready for it, and the government's defence counts half again as much. The host may also expel the sponsor's ambassador, which leaves its covert work in the country without an embassy (see Diplomatic Missions).

The strike: the armed forces decide

Every country has at least one armed-forces faction. Some have several, such as a regular army next to a party militia. When a plot strikes, each armed faction takes a position somewhere between the government and the plotters. Its position depends on which party it prefers and on the country's Military Control policy:

• A party army always defends the ruling party, and so does an army that is itself the government.
• A politicized military follows its own preference strongly.
• A professional army under civilian control defends a legitimate government, and only acts on its own preference as legitimacy fails.
• An agitated army that doesn't support the government leans further toward the plotters.

Each armed faction adds its position, weighted by its share of the national forces, to the side it chooses. The plotters add the strength of every faction that would support the coup, their own trained Brigades, and the weight of their backers. The government adds its legitimacy and popular support, which count for far less than the army does. The capital garrison doesn't count at all.

If you play a country with more than one armed faction, you choose which faction each newly trained unit belongs to. The AI coup-proofs by training for the faction most loyal to the ruling party.

Outcomes

The strike is a gamble weighted by the two sides' shares of the combined strength.

• Success. The plotters' party takes power. The plotters and the armed factions that sided with them become Victorious: +15% influence and calmer for five years. The ousted ruling party is Purged: half its influence gone, more agitated, and unable to plot for ten years.
• Failure. The plotters are Purged. Every armed faction that sided with them is Disgraced for three years: national Brigade Training Speed falls by 80% and legitimacy by 1% while that faction governs or backs the government. The backer loses prestige.
• A split. When the two sides are close, the country may break apart instead. The plotters proclaim a rival government in their strongholds, and the brigades there go over to them. The rival government is a military junta if an army leads it, otherwise a provisional government, and it is drawn on the map in a darker shade of its parent's colour. A civil war follows (see Wars), and peace terms can end it in partition.

Power struggles

A power struggle opens when a government loses its grip, meaning the lower of its legitimacy and popular support falls very low, and several factions are strong enough to replace it. A country can have only one struggle at a time. Factions that grow strong enough while it runs join as contenders.

Tension rises with instability, illegitimacy and Unrest, with the contenders' agitation, and with foreign backing. Foreign powers back contenders with the same Sympathetic, Committed and Guarantor levels as any flashpoint stance, and each level adds 15% to the contender's strength. A legitimate government steadies the struggle: for every point of legitimacy above half, tension falls faster, which buys a shaken but legitimate government time to recover. A government that is stable but illegitimate holds on only by force. Unless it represses, its order breeds unrest, and that unrest feeds the struggle.

The government has three ways to reassert itself:

• Crack Down on a contender, which then fights the struggle at half strength.
• Call early elections, where the constitution has them. This costs 10 political capital, lowers tension by 10, and brings the vote forward to 90 days later. Any election held during the struggle ends it. A government that wins comes out with its legitimacy raised to at least 60%. If the voters choose someone else, power changes hands peacefully.
• Martial law (below).

At 100 tension, the government falls to the strongest contender. If the runner-up commands armed forces of its own and is within 15% of the winner, or if unrest has reached half, the runner-up may reject the result and rise against the new government. The remaining contenders then join the side whose party they clearly prefer, or sit the war out. A successful coup during a struggle ends the struggle. A failed one leaves it running, minus the purged plotters.

Martial law

Martial law is a policy in its own State of Emergency policy set, so declaring it doesn't cost you your Internal Security policy. It takes about a week to put in force and requires military support of at least 50%. While it lasts, it gives +25% Political Capital Gain, much faster policy reforms, and +50% Unrest Suppression.

Its price is Legitimacy:

• Declared while unrest is below 30%, it is Unjustified: −10% legitimacy for as long as it stays in force.
• Declared amid real unrest, it turns Prolonged after six months: −5% legitimacy plus a steady monthly drain.

Returning to Normal Order removes both penalties. A new ruling party never inherits martial law: a change of government restores Normal Order automatically.

The new regime

Whoever seizes power, by coup, revolution or power struggle, starts with low Stability and low legitimacy, and receives a regime modifier for five years:

How the AI uses coups

AI powers back coups only for their own camp. Communist powers work through communist factions, and the West never does. They mostly target governments with low legitimacy and never a great power. An AI government facing a struggle readily cracks down. It declares martial law as unrest climbs, and it calls early elections when tension reaches 50 and its party leads the polls.