# Empire as core and periphery

The working definition from Bret Devereaux ([[no-empires-in-age-of-empires]]): an empire is a state where a core ruling population exercises control over, and extracts resources from, a periphery composed of people who are linguistically, culturally, ethnically or religiously distinct from that core.

The etymology does real work here. *Imperium* is a sphere of command, and you cannot hold *imperium* over yourself — so an *imperium populi Romani* is necessarily command over someone else. Otherness isn't incidental to empire; it's in the definition.

## What the definition rules out

**Monarchy is not required.** Republican Rome held an empire. France kept one after abandoning monarchy. Athens ran the *arche* over the former Delian League while remaining internally democratic. The republic/empire distinction historians use for Rome and France describes internal constitutional arrangements, not whether an empire existed.

**Homogeneous expansion is not empire.** If conquest replaces a population with your own, you have a nation-state expanding by settlement or extermination, not an empire. The distinction matters because the two have opposite logics: an empire wants the periphery's labor and taxes intact, so killing the periphery destroys the asset.

## Three consequences

**Extraction is the purpose.** Everything else is instrumental. Rome demanded military service from the Italian *socii* and revenue from provinces that didn't supply troops; that revenue increasingly paid for the army that held the whole arrangement together. Once you see extraction as the goal, imperial tolerance stops looking like virtue: empires largely didn't care about local customs so long as the taxes arrived, which is why forced religious or cultural conversion is unusual imperial behavior rather than typical.

**Diversity is an asset, not a cost.** This is the counterintuitive part, and the best-evidenced. Managing culturally varied subject populations is a core imperial competence — "empires that are bad at handling a wide array of different cultures/religions/languages do not long remain empires" — and the payoff is access to capabilities no single culture supplies. Mongol subsistence produced superb cavalry and weak infantry and siege engineering, so Mongol armies fielded Chinese siege technology and locally raised infantry alongside their horsemen. Roman armies at their height combined Italic infantry, Sarmatian or Gallic heavy cavalry, Numidian or Arab light cavalry, eastern composite bowmen and light infantry from mountainous provinces. The *socii* were always more than half the mid-Republican army, and Roman armies were essentially never majority Roman in any period.

**Hierarchy is permanent, not transitional.** Empire means a small core and a much larger subject population, and the arrangement is not on its way to becoming something else. Until the *Constitutio Antoniniana* of 212 A.D., most inhabitants of the Roman Empire held no Roman citizenship; the *populus Romanus* was a minority inside its own empire. The dominant human experience of empire is therefore being its subject — taxed and extracted from by a state that may inflict violence on you and care little for your needs — not being its conqueror.

## Why the historical frame keeps getting lost

Two mechanisms flatten it. First, the modern world is unusual: nation-states and voluntary federations are recent, and many present-day nation-states are consolidated empires whose long application of state power *manufactured* the shared culture that now reads as a pre-existing nation. Their imperial origin is hidden by their success. Second, the divisions that mattered enormously at the time — between Etruscan and Latin, between Iberian and Celt-Iberian — stopped mattering to us, so a map that looks homogeneous today looked nothing like that to the people on it.

Strategy games inherit and reinforce both flattenings. Titles that erase subject populations model expansion as growth of the core; titles that model them at all (*Europa Universalis*, *Stellaris*, *Imperator*) usually reduce them to a flat production or stability penalty — exactly inverting the historical case that varied subjects were what made an empire capable.

## Scope

This describes pre-modern **tributary** empires. Modern **mercantilist** empires worked differently, though Devereaux notes they were no less diverse and were on the whole more brutal and less tolerant. Whether contemporary states like the United States are best read as empire or hegemony is contested and he declines to settle it, while noting the answer looks different depending on whether you ask from Guam, Puerto Rico or a Navajo nation.

## Cross-references

- [[no-empires-in-age-of-empires]] — the source essay
- [[acoup-blog]] — Devereaux's blog, where the tributary/mercantilist distinction gets developed further