The Empire After Alexander is a historical simulation with exactly one permission to diverge. In this world Alexander granted Egypt to Aeson of Thinis about a year before his death, displacing the allocation history later gave to Ptolemy. That grant is canon, and it is the only established divergence. Everything else — every border, garrison, treasury figure, and succession — has to earn its place through evidence and through play. The file types exist to make that earning honest.
Two histories, kept apart
The project keeps two records of itself, side by side, on purpose.
The historical baseline is recorded history: the dated timeline from 323 to 273 BCE, the satrapal allocations, the Wars of the Successors. It is the causal reference the world runs on — what would happen if the campaign never happened. It is explicitly not a script. The Babylon settlement of 323 BCE is a baseline the campaign is heading toward, not an outcome it has reached; no record should ever assert it as campaign fact.
The campaign history is what actually occurs at the table. Divergences get their own dated entries with causes, actors, and consequences, recorded alongside the baseline they replaced rather than over it.
Every political record carries both. Territory files hold historicalBaseline and campaignDivergence fields next to each other, so a reader can always see which claims come from history and which from the campaign. Snapshot fields are dated and labeled — "post-death, post-Babylon settlement, 323 BCE" — because the world before Alexander's death and the world after the settlement are different political states and must never be conflated. The live campaign opening is 324 BCE; later snapshots are milestones, not the opening state.
Evidence discipline is the core feature
A number in this world is worthless without its provenance. Every figure in the military and economic records carries an evidence class — H for ancient report, S for scholarly estimate, C for campaign record, GM for GM estimate — plus a stated basis, a confidence, and a citation where one exists. The honest answer to "how many troops does Egypt have?" is often "unknown," and the schema is built so that unknown is a legitimate stored value rather than a gap to fill with a guess. The next chapter covers the discipline in full; for now, understand that this is the feature the whole design serves. When a record is empty, that emptiness is an assertion about the evidence, not a missing file.
What each file type is for
| Type | Job in the simulation | What it carries |
|---|
| Character | The people: <@aeson-of-thinis.character> and every officer, ally, rival, and heir who acts on the world | Stats, ancestry, background, personality; the two campaign modes (immortal Aeson, heir succession) live here |
| Location | A settlement, fortress, port, landmark, or map region — what physically exists at a date | Permanent ID, coordinates with source and uncertainty, parent territory, and the split between formal authority, de facto control, claimants, and political status |
| Territory | The political wrapper: a satrapy, region, or kingdom that administers the land | Baseline vs. divergence, snapshot, formal authority vs. effective control, geographical scope, evidence notes and open questions |
| Political Entity | The actor: kingdom, regional authority, power bloc, or individual power that claims and rules | Authority (holder, office, basis), claims and control, military position, territorial basis |
| Military Record | Dated force balance sheets, battle snapshots, and deployment reports | Mutually exclusive force categories (standing establishment, field army, garrisons, mercenaries…), an evidence class per entry, never a double-counted unit |
| Economic and Infrastructure Record | Dated ledgers for money, resources, trade, and works | Resource flows and financial entries with explicit units, period, confidence, evidence class, and citation; stocks kept distinct from annual flows, cash from grain |
| Reference | The free-form markdown layer where conventions and baselines live | The historical timeline, the military strength database, the atlas build process and geographic framework |
The shape of the whole
The layers stack. A political entity (an actor) holds authority over a territory (a political wrapper), which contains locations (the physical places). Military and economic records hang off a subject and date them: forces, treasuries, harvests, harbors, roads, and fleets are always recorded as of a stated date, for a stated subject. References sit outside the stack, holding the conventions and baselines that keep everything else consistent.
The division of labor is deliberate. A location records what exists and who controls it; it does not assert why it matters historically. A territory records administration and claims; it does not hold the ledger. A record quantifies; it does not narrate. Each file has one job, and the fields were chosen so that one file's fact never needs to be copied into another — copies are where contradictions breed.
Before you add a file, ask what job it is doing. If it is a fact about what physically exists, it belongs in a location. If it is a claim or an administration, a territory or political entity. If it is a number with a date, a record. A file that does not know its job is how the system breaks.