Collections: Teaching Paradox, Imperator, Part IIb: Built in a Day

This is the second half of the second part of our three part (I, IIa, IIb, IIIa, IIIb) look at Paradox Interactive’s ancient grand strategy game Imperator: Rome which covers the broader Mediterranean and South Asia from the very late fourth century through to the end of the first century BC, the period of the imperial rise and political collapse of the Roman Republic and a period – relevant for today’s topic – of rapidly rising urbanism across the Mediterranean, especially the western half. Last week, we looked at how Imperator models populations and economies; some of the details from that week will be relevant here as well, particularly how pops work in the game.

This week, we’re going to turn to look at Imperator‘s view of cities, in particular: how do cities come to be and who are they for? Here too, Imperator has a ‘theory of history,’ a vision of how it understands these historical processes. The game is abundantly aware that this is a period of rising urbanism, both the creation of new cities and the expansion of some existing cities (particularly imperial capitals) into huge, sprawling metropolises,¹ and it very much wants to simulate that. Building both new cities and huge, monument-strewn cities is, after all, a core part of the ‘fantasy’ the game is presenting to the player and it wants to facilitate that.

At the same time, as we’ll see, the vision of urbanism that Imperator‘s mechanics create is one of fundamentally monumental urbanism, where cities are created by the state, for state purposes and often serve primarily as a canvas on which state power can be painted in marble and limestone. This vision of ancient city building is ubiquitous in the gaming space – at some point I will come back and tackle it in more depth looking at the Impressions city-builders and their spiritual successors – but it offers an incomplete vision of actual ancient urbanism, as we’ll see.

But first, if you like what you are reading here, please share it! And if you really like it, you can support me, my academic research (on the Roman Republic!) and this project on Patreon – behaving much like the local elites of antiquity by funding at private expense the production of public goods for the greater glory of the community (although I cannot promise to get any of you elected aedile)! If you want updates whenever a new post appears, you can click below for email updates or follow me on twitter (@BretDevereaux) for updates as to new posts as well as my occasional ancient history, foreign policy or military history musings, assuming there is still a Twitter by the time this post goes live. I am also on Bluesky (@bretdevereaux.bsky.social) and (less frequently) Mastodon (@bretdevereaux@historians.social).

Brick and Marble

Before we dive in to the mechanics of how Imperator understands urbanism, it is worth taking a look at the ‘whole package’ of how the game presents this theme. Games are, after all, multimedia: not just a bundle of mechanics, but also music, sound-design, graphical and art choices which both provide the developers an opportunity to communicate their intent but also shape the player’s experience of the game.

And Imperator goes some length to communicate its interest in cities, particularly graphically. It isn’t just splash screens like the one I’ve used above (although those certainly count), but how cities and their buildings are represented on the game map that communicate their importance. Whereas rural territories (‘settlements’) are simply represented with their terrain on the map, cities are drawn with clusters of 3D buildings in a set of different styles depending on the dominant culture of the city in question. Not only are cities marked in a way that other settlements are not, they change based on their development: as a city grows, the number of small buildings that indicate it on the map grows, from a cluster at the center of a territory for a small city to a sprawling mess of buildings filling the entire territory for a large city. The player is thus given graphical feedback to reward building big magnificent cities.

You can see the contrast in this approach fairly clearly if compared to CKIII‘s approach to the same question – how to depict settlements on the map – released one year later. Whereas Imperator marks cities but does not mark settlements, CKIII marks all three kinds of holdings; empty holdings in CKIII really are empty – waiting for someone to build something in them and unproductive until that is done. However, CKIII represents each kind of holding the same way, as a single small model on the map. Cities on its map do not sprawl out the way they, becoming clear visual landmarks for the player on the map, in Imperator – or, for that matter, as in Victoria III, another game interested in development and urbanism to a substantially greater degree than most Paradox titles. These sorts of choices are intentional: development time and art assets were sunk into displaying cities like this, visibly on the map (though the rules and system for doing so did change post-launch) and thus this is an important part of the game’s design.

Added on to this, certain key urban buildings are also represented as part of the digital city-scape: forts, ports, and particularly wonders. ‘Great Wonders’ are a feature added to Imperator in the Heirs of Alexander DLC, enabling players to build their own huge monumental constructions to rival some of the famous ‘Wonders of the Ancient World.’ Once built, these structures are large and visible on the map as part of the city-scape and the player is given the ability to customize, to a degree, the design of these structures and thus their appearance on the map. Wonders also provide a customizable set of bonuses to the city in which they are built, which (as discussed below) generally incentivizes building them in large imperial capitals.

This is thus a visual language which aims to encourage the player to build not just cities but large cities, both for the curiosity of seeing just how far the system for generating city-scapes can be pushed (how dense will it place buildings, how large will it go?) and also because the sight of one’s own massive city-scape on the map is a kind of reward in and of itself. A player can choose to found a city in a rural territory and then grow it over the course of the game, building up a huge metropolis where once there were only fields or forests on the map – that is part of the ‘fantasy’ that the game is attempting to tap into. Imperator is interested in cities and urbanism and communicates that not just through its mechanics, but also through its artwork and visual design.

But Imperator is not just interested generally in urbanism, but in a particular vision of ancient urbanism, what I’ve come to call monumental urbanism, where the city itself serves as a monument, a vast canvas for the demonstration (and creation) of state power. This is a vision of urbanism that is less about creating a space in which people live and more about creating massive, sprawling imperial capitals whose most important attribute, apart from their great size, are their imposing monuments, build by and testaments to the power of the state.

And that kind of urbanism, while it synchronizes neatly with the visual design of the game (big, monumental cities are the most impressive sort and the actual monuments – the ‘wonders’ – are the buildings that get the most focus) is more clearly expressed through the game’s mechanics.

Mechanics and Incentives

Even if the player didn’t come into Imperator already looking for monumental urbanism, the game is set up to incentivize it, both over undirected organic urban growth (which mostly doesn’t happen in Imperator at all) but also over other models of urbanism. Massive imperial capitals are not merely satisfying for many players to build, they’re encouraged by the mechanics, actively incentivized over more decentralized urbanism (like a collection of mid-sized cities strung over multiple regions).

Assuming the player is wholly uninterested in playing to the visual or ‘fantasy’ aspects of building cities in Imperator, cities still have mechanical import. As noted last week, the primary unit of production in Imperator is pops, who produce most important resource at a set base rate per the number of pops of that social class. Mechanically, then, cities allow pops to be concentrated, which in turn allows production to be concentrated for nearly every key resource (except food). This concentration of production does not suffer diminishing marginal returns: the 200th pop produces the same base amount as the first and the food consumption is likely linear.²

Concentrating those pops turns out to be a very smart idea mechanically for two reasons: bonus stacking and stability.

Resource production in every territory can be further affected by buildings, which provide various bonuses to production, population capacity and so on. Rural territories, ‘settlements,’ can generally only ever have one building on them (two with the ‘rural planning’ civic), but cities can have many buildings on them: a base of three building slots, plus one for every 10 pops in the city. This can be raised further through certain innovations, wonder bonuses, being a major holy site, or by investing political influence with the ‘make religious endowments’ province investment, and so on.

Many of the city buildings, in turn, provide percentage bonuses to the production of the city itself: libraries provide +5% research points, tax offices +10% tax, training camps +10% manpower, market places +2.5% base trade routes and so on. Crucially, these bonuses stack, so a city with three tax offices has its tax revenue increases by 30% over whatever base value its pops would imply.

A quick exercise using just tax revenue can demonstrate the incentives that creates. Posit a city with ten slave pops and ten freemen pops, producing 0.2 tax revenue per month (three quarters of which is from the slave pops), assuming 100% happiness among the freemen. If the player builds a tax office there, that becomes 0.22 per month (+10%) so the office is generating just 0.02 revenue per month. It cost 100 gold to build (base), so it takes just over four hundred years to pay for itself, assuming, of course, the city doesn’t grow. What a bad deal! But what about the same building in a very large city – say, one with 50 freemen and 50 slave pops? That city is producing 1 revenue per month – 1.1 with the tax office – and so that same tax office now pays for itself in just about 80 years. The same effect, of course, goes for all of the resource buildings. Larger cities allow building bonuses to multiplied over more pops, producing larger effects and making those buildings more efficient.

This effect, in turn, is compounding, to a degree. Most buildings also raise a territory’s ‘civilization value,’ which modestly increases both population capacity, population growth and population output, as well as raising pop happiness which also impacts production (which for all pops but slaves is multiplied by the happiness percentage). Several buildings (the academic, court of law, forum and mill) allow the player to shift the desired ratios of pops in a city, so you can create, for instance, a city that is shifted towards freemen (with forums) and loaded with training camps and thus produces a lot of manpower efficiently; these buildings also increase that pops happiness as well, increasing the effect.

Finally, the player can spend 80 ‘political influence’ to make a province investment, which provides a bonus across every territory in a given province. The available investments can increase population capacity, the number of import routes (to get those surplus goods bonuses), the number of building slots, or fort-capacity and loyalty (if the last one sounds lame, that’s because it is).

Because these are all multipliers on pop production (or, as in the case of some of the investments, ways to get a fixed amount of new multipliers, like more building lots or trade routes) with fixed costs it makes sense to try to ‘hit’ as many pops with them as possible. Which of course means the smart player is going to try to focus as much of this development as possible into a single massive urban center (later sprawling into several adjacent urban territories all benefiting from heavy provincial investments), where the concentration of very large numbers of pops and multiple overlapping bonuses produce the biggest ‘bang’ for the player’s investment ‘buck.’ Several of the available bonuses for building Wonders are also localized (notably the bonuses to trade routes and population growth and capacity), so it is going to make sense to concentrate those too. By concentrating, the bonuses to migrant attraction, population growth and population capacity ensure that the buildings giving bonuses to production are spreading that bonus across as many pops of the desired type as possible.

Which brings us to the question of where this focused, highly centralized urbanism should go. And the answer to this is actually fairly simple: it should go in the player’s country capital city. First, the country capital gets its own set of bonuses, with an extra building slot, higher population capacity and a substantial bonus to migrant attraction. It also usually gets more bonuses on top of this through mission chains and events and is the primary target for newly relocated enslaved pops.

But more than this is the impact to realm stability. We’ll explore this more next week, but in short for this, Imperator has two kinds of civil wars, one of which triggers off of low loyalty among elite characters (governors, generals and such) and the other off of low province loyalty, representing popular revolts. But the region a country’s capital is in always has full province loyalty and is always governed by the country leader (the king, consul, etc.) who always has maximum loyalty as well. In short, the manpower, production and revenue compressed into this grand capital becomes a key stabilizing force, because it can never become disloyal to the ‘country,’ which is to say to the player. Even if it wasn’t already smart to stack bonuses by focusing as much urban development as possible into one or a handful of mega-cities packed into a single province, it would still be optimal to do so simply to maximize the ‘locked in’ loyalty of as many pops (and their optimized production) as possible.

This stability effect is, I may note, somewhat ironic, as gigantic imperial capitals had quite the opposite reputation! First century Rome is, of course, the famous example of a city that, from the Late Republic through much of the Imperial period was difficult to govern – at points almost uncontrollable – because of its vast size. But it was hardly the only example: Alexandria, for instance, had much the same reputation in its heyday (e.g. Polyb. 34.14.3-4). Large imperial capitals, after all, created both a large, relatively poor urban population reliant on food imports for whom economic disruptions could quickly mean starvation and bread riots. At the same time, they also focused most of the wealthiest and most influential elites into a single place, where they could communicate rapidly, meaning that politics in these capitals could move with destabilizing speed. As we’ll get to, such imperial mega-centers were both unavoidable because of the nature of the empires that bring them into being and also necessary for their administrative function, but they were hardly stabilizing influences!

Nevertheless, Imperator‘s game mechanics are set up to encourage the player through a variety of mechanisms to build exactly this sort of outsized imperial capital. And that, to a substantial degree makes sense: the player is incentivized to build Rome, which by the end of this period was – by ancient standards – an absolutely massive urban center of probably around one million in population. And that makes good sense: the massive size of cities like Rome, Alexandria, Carthage, and so on is a famous and accurate feature of this period (one that featured rising urbanism across the Mediterranean) and so central to the ‘fantasy’ of playing a game set during the Roman Republic.

But what I want to turn to is how Imperator models the creation of such cities rather than there simple existence. In short, I want to ask…

Who Build Rome and Why?

Imperator actually has a very clear answer to this question: the state built Rome, an answer which is, unfortunately, mostly wrong (but not entirely so). Or put more accurately, Imperator presents a model of state-driven urbanization in which most of the key decisions are made by state actors – kings, consuls and senates – such that the player (who effectively plays as the state) is able to intentionally and deliberately mold and reconfigure the economic structure of the countryside, creating cities with specific purposes. Crucially, it does not create cities any other way. Nearly all urbanism in Imperator is directed by the player (or the AI for AI-controlled countries), which is to say it is directed by the state.

Instead, in Imperator, the life-cycle of city development is almost entirely player-directed. Cities are created almost exclusively by state action: cities are founded by spending 200 gold and 80 political influence on any owned territory of settlement status. This initiates a foundation process lasting two years, after which the territory type changes to a city. Political influence is a currency we haven’t discussed much here, but effectively represents the state’s political capital, usually used for various internal political actions. In this case, it presumably represents the exertion of political will necessary to reorganize the territory and establish whatever special civic rights the new citizens will have, as ancient cities (and their citizens, who were of course, not all of their residents), even within larger empires, were almost always accorded special status. Most players are going to want those new cities to ‘come up to speed’ quickly and so are likely to both queue up quite a bit of building as well as transfer population. For the latter, the game offers two options: the player can set the local governor’s policy in the province to centralization, which pushes the migrant attraction of the province capital (likely the new city) or they can directly and instantly transfer any slave pops in the province to the capital at the expense of political influence. Because that huge infusion of slave pops will put the new city well over its desired ratio, they will fairly quickly begin promoting into freemen and other pop types, far more rapidly than if the player simply waited for migration to do its work.

The player is then likely also to build out the available building slots of the new city for whatever purpose it is intended to serve: markets, ports and mills for a trade-resource exporting town, training grounds and fora for a military settlement generating manpower and so on. The ‘trick’ here, as an aside, is pairing the buildings which shift the ‘desired ratio’ of a city and the buildings that boost the production of those pops so as to create a specialized mid-sized city that is focused on more efficiently producing its core product, but that trade goods (mills, markets, ports), revenue (mills, tax offices), manpower (fora, training grounds) or research (academies, courts, libraries). Perhaps fittingly, given how many mid-sized Roman cities resembled each other, the player is going to end up with what are basically a handful of ‘templates’ for new, highly specialized cities (which is, as we’ll see, not how ancient cities formed).

While as noted, the player is likely to direct a lot of their city-building efforts to the imperial core, there are reasons to create new regional centers in non-urban, newly conquered territories as well. Cities allow the player a lot more tools to manage pop happiness, making them valuable for stabilizing the loyalty of recently acquired provinces, and they can also be useful for generating larger amounts of rare trade resources for shipment back to the capital. Finally, regional levies are exactly that, regional, so a few mid-sized cities can provide a local military force that does not need to make long marches from the imperial core to respond to problems on the frontier.

And I should note here, in fairness, that larger imperial powers did found cities for these reasons, particularly for the creation of colonial military ‘strong-points.’ From the late fourth to the very early second century, the Romans got into the habit of planting colonies of either Roman citizens or socii of the Latin right (the latter called ‘Latin Colonies’ for this reason, though many of those colonists were Romans) in militarily important or difficult to control parts of Italy, effectively as nails hammered into the countryside to hold down the tarp of Roman power. Outside of Italy, the Romans continued this habit with the occasional settlement of veterans colonies, many of which bloomed into famous cities, although it is worth noting these veterans colonies were frequently ‘planted’ on existing urban centers. Likewise, we’ve already discussed in another series the habit of Hellenistic monarchies in planting military settlements in key areas of their kingdoms in order to create pools of ethnically acceptable ‘Macedonian’ manpower for their armies.

One oddity I will note here, at least in its current form, is that Imperator doesn’t model the population movement effect of such colonies very well. Colonies of this sort are – very infrequently – created by event or cultural decision, in which case they generate a small group of the appropriate ethnicity of pop. But in fact what is happening is that a fairly fixed population is being redistributed geographically: the Seleucids, Ptolemies and Antigonids are competing for manpower that fits their ethnic preconceptions, while the Romans seem to have used their colonies as an opportunity to reduce land pressure in core Roman territories by offering poorer Romans or loyal socii the opportunity to settle on captured land. What is almost never happening is a ruler founding a city for the conquered populace and then consolidating them there: when an ancient state founds a city, it usually imports the citizens. Indeed, that is typically the point.

What is perhaps more important to observe is that such direct, intentional urban foundations were the exceptions, not the rule. That isn’t to say that state action had no impact: the spread of Roman power contributed in clear, visible ways to the emergence of a broader, more urban Mediterranean culture, but it was not the Roman state who built the cities. Of course in much of the Eastern Mediterranean, large local urban centers already existed as they had for a long time. No one needed to found a city at Ecbatana, Ephesus, Athens or Cyrene as those cities already existed. The urban patterns of the Levant, Syria, Egypt, Mesopotamia, Anatolia, Greece, Italy, Tunis and the Mediterranean coasts of Gaul and Spain were already at least partially set before this period (though Syria and of course Alexandria in Egypt see substantial state-run settlement). Apart from those Hellenistic settlements, those urban patterns were only indirectly reshaped by imperial power.

Imperator, unfortunately, models effectively none of this, leaving it up to the player to do with intent what ancient states did largely by accident. So let’s talk about the forces (mostly social and economic) which reshaped the urban fabric of the Mediterranean in this period.

Urbanizing the Provinces

We can start with the most extreme examples in Rome’s western provinces, where urbanization itself was relatively new. Now, Imperator represents these areas – we’re talking here mostly of Gaul, Spain, Britain and the Rhine frontier – as entirely rural. At start game start, the few urban centers Imperator places out in these regions are generally Greek or Punic colonial enclaves and that tends to remain true over the course of a playthrough, as the ‘tribal’ AI does not generally found cities.

But that is our first problem, because while these areas were far more predominately rural than, say, Roman Italy, that isn’t to say they had no large settlements or that the new Roman cities they were about to have were entirely new foundations.

Instead, what we might term as early urban settlements are already emerging in the western Mediterranean before the Romans arrive.³ These typically begin as fortified hilltop settlements, though their size and pattern vary, which we call oppida. Indeed, by 218 when the Romans arrive, the Mediterranean coast of Spain, inhabited by the Iberians, was already a fairly dense network of such settlements.⁴ The Celtiberians, the inhabitants of Spain’s interior, begin coalescing into oppida in the third century. In southern Gaul, oppida emerge (most on local Gallic patterns, a few, like Glanum (go visit!) imitating Greek patterns) beginning in the third or second century, while oppida start showing up in northern Gaul in the late first century. These hilltop settlements are more than just castles or forts, they are population centers and pretty clearly the first phases of urban development.

Then the Romans show up, conquer the place and convert these regions into provinces, overseen by Roman governors. Contrary to popular perception, Roman provinces in this period did not have capitals at all: governors were instead generally itinerant and what little administrative apparatus they had they carried with them. As a result, the Romans aren’t immediately building ‘provincial capitals’ because they do not require any. Indeed, outside of a few veterans colonies founded in the late second or first centuries, official Roman interest in the urbanization of these regions is basically nil, outside of a handful of sites that were state-owned or state-leased mining operations (particularly in Spain).

Instead it was local elites, the same elites who built the oppida who took the next steps. The basic pattern here varies from one region to the next a bit, but is generally fairly consistent. Where the original focal settlements were somewhat inaccessable (being often on hilltops at the edge of larger areas of flat, agricultural lands), they move down into the lowlands, remaining quite clearly the same settlement. In other cases – perhaps most famously Paris and London – the old oppidum ends up converted into the newly forming and expanding Roman city, with the ruins of the old archaeologically visible beneath the Roman buildings (which are in turn beneath the medieval buildings which are in turn beneath the modern buildings).

The first step was typically the creation of a Roman-style central open meeting and market space – a forum – modeled less on the Forum Romanum at Rome and more on the fora of those handful of veterans settlements and colonies (which tend to be regularized, ‘perfected’ versions of the far more irregular and organic Forum Romanum – if you’re in Italy and want to see this, contrast the Forum Romanum‘s somewhat chaotic layout with the neat, regular forum at Pompeii; the provincials are imitating fora that look like the latter). Within a few decades, these new fora get populated with the standard set of public buildings, first in local styles and materials (often wood, wattle and daub) and then later as the city grows in wealth, rebuilt in Roman style and materials (brick, stone, with ceramic roof tiles).

These new cities become focal points for the local elites and it is the local elites who are building them. Those local elites, in turn, have become the conduit between the local population and the imperial elite (that is to say, the state), organizing the local sub-imperial polity (what the Romans call a civitas) with its own laws and internal governance, collecting its taxes and interfacing with the Roman governor and his staff.

At the same time these settlements are ‘plugging in’ to the Roman trade network. Roman currency begins to circulate – the locals need it to pay their taxes and they get it by exporting goods into the Mediterranean trade network. Sitting on navigable rivers, major Roman roads or both, these nascient cities often become simultaneously the point of contact between the locals and this trade network and the political focal point of the local elite as the civitas (but not provincial)⁵ capital. That also means these are the places where all sorts of Roman goods are coming in and available: wine, olive oil, ceramics, metal-work and so on. At first, the local elites acquire these things as signs of status and part of culturally aligning themselves with their new imperial masters – a good idea if they want to prosper in this new world – and then, in imitation of the local elites, many of the rest of the locals who want to signal their own status begin doing the same on a smaller scale.

That flow of goods, money and labor through the cities, along with the elites themselves who are collecting rents on the agricultural lands around them and spending those rents in the cities (where they live) turn these nascent cities into local focal points of economic activity, which of course draws workers (both free and non-free), expanding the population. Now some of those workers are enslaved and thus have little choice about being moved into the new, growing cities, but many of them are free and do have a choice, but move regardless (indeed, due to disease effects, above a modest size, cities could not maintain positive population growth and so required continuous inflow of migrants) and we shouldn’t ignore their agency. These new cities offered access to economic opportunities that might not exist in the countryside (where wage labor was very scarce and thus opportunities for non-landholders few) but they also offered access to status and prestige.

In the meantime, those elites who are the leaders of these places are themselves anxious to show other elites, both local elites and the imperial elite, that they too can be ‘civilized’ (they can acquire, in Latin, humanitas, ‘culture, cultivation, education’ but also ‘generosity, kindness’). And so their cities begin to acquire those standard markers of ‘civilization’ in Roman eyes, like baths, aqueducts, stone temples, monuments and public spaces. Some of these will be motivated by the need to serve a growing population (like aqueducts), others (like amphitheaters, theaters, baths) by the need to demonstrate the status of the community.

In most cases the only direct role the Roman state may take in all of this is that those local elites might ask for a Roman military engineer to help them design the structures. But the funding and labor were almost invariably local. I should note, these processes were hardly unique to the less urbanized parts of the empire: we also see elites in the more urbanized east engaged in the same kind of projects, often resulting in a mix of Roman and Greek influences in the buildings. There’s actually a memorable bit in Pliny the Younger’s letters where he notes, as the governor of Bithynia et Pontus in 110, that the city of Nicomedia (modern Izmit) – a long-standing and major city – were trying to build an aqueduct to supply the city with water using local funds. The project had run into trouble repeatedly (Pliny and Trajan, in their letters, suspect some of the local elites ‘doing eachother favors’ with the funds) and so Pliny deploys a military architect to get the project back on track (Plin. Ep. 10.37). While Pliny is told to investigate what happened to the city’s public funds, he isn’t replacing them – the city will still have to fund its own damn aqueduct. Likewise, he writes about other public works, like a theater in Nicaea, badly behind schedule, which was to be embellished by various private individuals (now waiting for the main structure to be finished) or public baths in Claudiopolis funded by the town councilors (Plin. Ep. 10.39, note also 10.40, 10.41, 10.49, 10.70, 10.90, 10.98 (and probably some I’ve missed) which all concern public works begun and funded by various cities).

This isn’t quite bottom-up urbanism in that the direction is coming from local elites, rather than from the common folk, but it also isn’t top–down, coming from the imperial elite, much less the state itself. Instead, economic changes in this period, driven in part by Roman expansion, created the conditions for increased urbanism and the status-conscious upward mobility of local elites created the demand for it. For the state – the senate (and later emperor) and their agents – this was a most a happy unintended byproduct of something (military conquest and customs normalization) that they did for entirely unrelated reasons (military glory and the extraction of revenue).

The Least Lonely City

The growth of the largest urban centers in this period was similarly a product as much of indirect economic forces as it was direct state action, although the latter plays a more important role. Many of the largest cities of the Roman Empire – Alexandria, Antioch, Carthage – had begun as imperial capitals before being eclipsed by Rome and maintained their regional economic importance afterwards. For imperial capitals of this sort, things were a bit different because the local elite were the imperial elite. State or royal largess thus tended to be concentrated in these cities, often as a means of legitimacy building. In this sense, the game’s system of wonders – huge, monumental state-funded constructions that players are almost always going to place in and around their capitals – fit neatly into that sort of state project.

But it wasn’t monuments that made ancient capitals massive or encouraged dense networks of urban centers around them (as we see in Italy). Instead, it was the indirect forces created by imperial expansion that did so.

Fundamentally all of the major empires of this period are structured on the macro-scale as military-tributary complexes, which is to say that they use their military to extract agricultural surplus and other revenue from the subject regions of their empire (the ‘periphery’) for the benefit of the home region (the ‘core’). For the Romans, this process begins with the conquest of Sicily in the First Punic War (264-241), where instead of being incorporated into Rome’s socii system, Sicily becomes the first permanent province and eventually ends up paying a regular tax to Rome. Subsequent conquests outside of Italy are added to the same system, each establishing a stream of income flowing to Rome.

Some of that income ends up in the Roman state treasury (the aerarium Saturni) and some of it – quite a bit of it, in fact – ends up captured by Roman elites. There are actually quite a few points at which these revenues can ‘fall off the truck’ so to speak, into the pockets of elite Romans, some legal, some otherwise. Roman governors clearly profited off of provincial assignments: Cicero reportedly pulled 2.2 million sestertii out of his province of Cilicia (ad. Att. 11.1.2) evidently more or less legally; corrupt governors could do much more. Many of Rome’s taxes were collected not directly by the state, but contracted out to tax farmers called publicani, who could profit off of the business of tax collection. Back in Rome, the state also let out various public contracts for the repair of public buildings or supplies to be sent to the armies and so on, from which wealthy Romans with resources might profit. And then of course the rest of the revenue is often being spent on building projects which employ workers – or more often on military activity which employs soldiers (who are, you will remember at this point, conscripts not long-service professionals).

All of that revenue flowing in creates a very strong focal point of economic activity. Increasingly wealthy Roman elites want services, after all: they purchase slaves and hire workers and then those folks need food and housing and so on. In the second century, that attractive effect where Rome’s high degree of economic activity is drawing in workers (some of them non-voluntarily) seems to have been combined with rising population in the countryside due to falling rates of military mortality⁶ which means there is also a population surplus (over the availability of land) in the countryside and those folks have to go somewhere and Rome is where there might be work.

But of course, unlike cities in most fantasy settings, the growing city of Rome does not exist in the center of some flat, empty featureless plain. Instead, it exists in a network of both rural, agricultural areas and other smaller urban centers. But the growth of the city (true for any imperial capital of this sort, but especially so for Rome) is going to impact the human-created terrain (the ‘demographic space’ in John Landers’ phrasing)⁷ around it, because this huge center of population and wealth creates tremendous demands for goods. That process causes agricultural utilization in the immediate hinterland of the city to intensify, to make maximum use of the very high value land in close proximity to the city, but as that proves insufficient, the city’s hinterland spreads, coming to encompass other towns and indeed eventually other cities.⁸

You might expect that process to hollow out smaller urban centers, but it seems to have had the opposite effect. After all, the economic ‘pull’ is moving goods from the countryside – these are mostly agricultural economies – to the imperial metropolis in Rome (or Alexandria, or Carthage, or Antioch/Seleucia Pieria- this goes for any big imperial capital), but those goods need to be gathered, warehoused, often processed (like the making of olive oil from olives), packed and purchased by larger wholesalers in order to go to the Big City. All of those processes in turn create economic activity in the market towns supplying Rome. Indeed, Rome’s economic ‘pull’ was so vast that we see towns responding to it from across the Mediterranean, like Leptiminus, a North African town in what is today Tunisia which ends up seemingly specialized in olive oil processing and transport. Because the economic forces surging through ports or down Roman roads might not follow the same geographic pattern as before, they could often reconfigure the urbanism of whole regions, elevating new cities to prominence and causing others to become secondary.⁹

That role, the city as the focal point for trade networks, likely explains the persistence of the old imperial capitals (Alexandria, Carthage, and Antioch/Seleucia Pieria) after their empires were gone and indeed in Carthage’s case after the city had been destroyed and then refounded: they had already transformed the local economic system to place the Big City and its large markets at the hub and while their imperial purpose had gone, that economic purpose (big markets on big ports)¹⁰ survived.

Of course for the imperial capitals, there was also substantial state-funded building. Rome, of course quite rapidly acquired a spectacular suite of public buildings, many of them built by the first emperor Augustus who could justly boast that he had found Rome a city of brick and left it a city of marble (Suet. Aug. 28.3). But of course that explosion of public construction (financed substantially by the proceeds of the conquest of Egypt) comes right at the end of Imperator‘s time period, at a point when the city of Rome was already a vast, sprawling imperial capital. But not all of Rome’s public buildings were the product of emperors! Suetonius, after recollecting all of the buildings Augustus himself built during his reign is also quick to note that Augustus encouraged other Roman elites to do their own (smaller) building projects in Rome: a temple of Hercules restored by Marcus Philippus, a temple of Diana restored by Lucius Cornificius, a Atrium Libertatis (‘Hall of Liberty,’ originally an office for the census) restored by Asinius Pollio, a theater built by Cornelius Balbus, an amphitheater by Statilius Taurus, and “many magnificent structures” (to include the original Pantheon) by Marcus Agrippa (Suet. Aug. 29.5). That was the continuation of a long tradition of Roman politicians building or refurbishing public structures in the city; it was a particularly effective way for triumphing generals to make their loot politically useful. For instance, that temple of Hercules that Philippus is restoring was built my M. Fulvius Nobilior (cos. 189), who built it following his triumph of the Aetolian League in 189. Famously, Rome’s first stone theater was the Theater of Pompey, with construction beginning in 61 and being completed by 52. Like Nobilior, Pompey was capitalizing on military success, having celebrated his third triumph over Mithridates in 61.

Of course alongside this there were also buildings whose construction or maintenance were ordered by the state itself, in which case the funds came from the treasury and the construction contracts were let out by the censors during most of the period Imperator covers (the Middle and Late Republic). The Basilica Aemilia, for instance, was so let out by the censors of 179, M. Fulvius Nobilior (that same guy) and his colleague M. Aemilius Lepidus, after whom the structure is named. Though notably after that, members of the gens Aemilia tended to regularly refurbish the structure at their own private expense, so even here, not everything was at state direction.

The other major element of state infrastructure were projects responding to the large size of these cities and the need to provision food and water. Rome had five aqueducts when Augustus became sole ruler of the Roman world in 31 and two more were built during his reign. The earlier republican aqueducts were built at state expense, their construction and later repairs and maintenance contracted out by the censors. In Imperator, aqueducts function to raise the population limit of a city, and that is a pretty direct extrapolation of their actual purpose in supplying clean water (generally drinking water) for large cities that had exceeded the capacity of local wells and cisterns.¹¹

The other point of state interaction was the grain supply. With an absolutely massive population, the city of Rome obviously outstripped the food production of its local hinterland and so needed to import grain. Moreover, the large urban population meant that the natural stabilizing force of most ancient societies, that most farmers produced much of what they needed to survive and were thus very resilient, wasn’t present: these folks relied on the markets to buy food. The result was both elevated grain prices (roughly double what seems to have been something like the average)¹² but also unstable grain prices. The Roman state begins to try to stabilize those prices in 123, with a law passed by Gaius Gracchus whereby the state would guarantee the sale of grain to a small number of poorer households at a fixed relatively low price. Over the decades that followed, the distribution, called the annona (its administration the cura annonae) was understandably a matter of considerable political debate, with the roles generally expanding (they peaked around 300,000) and the price eventually being set at ‘nothing.’ Both Julius Caesar and later Augustus pared back the number of authorized recipients, with the figure fixed at 200,000 under Augustus (Res Gestae 15).

This was the ‘bread’ in ‘bread and circuses,’ a grumpy line from Juvenal (Juv. 10.81) taken far too seriously by modern authors as a sign of luxury, frivolity and decadence. Instead, the annona was a fairly essential no-fuss system. What was distributed was not bread, but raw grain, collected as a tax in kind in the provinces and shipped to Rome through a system of contractors. The annona did not, until much later, include anything else: it was grain only and the amount (60 modii per person per year) is a bit more than enough for a single person but not enough for a whole household. The function here is not as a replacement for the market (or to provide anyone with leisure) but as a price-stabilizer designed to avert catastrophic shortages or surges in prices, mostly as a means of avoiding the political instability of bread riots. That the people were thus anxious about the officials who supervised their essential food supply is hardly a sign of decadence. Even if the Romans had chosen another mechanism, some state intervention would have been necessary to ensure the reliable food supply of the city – for as much as I am often impressed by the sophistication of Roman economic structures, no ancient economic system that relied on sailboats to move grain, men on horses to move messages (like that the city was running out of grain) or modern food storage (like huge grain elevators loaded and unloaded by machine) was going to be up to the task of ensuring a stable food supply to a city of a million people under all conditions.

What I think is important to note from this context is that both sets of actions, the building of aqueducts and the provision of grain, were not anticipatory actions where the state aimed to create a super-sized capital city. Instead, they were responses to a city that had outgrown local resources and was continuing to expand regardless, driven by the tremendous nexus of economic and political power Rome’s conquests had created. Rather than planning, either for strategic reasons or simply out of pride, to have such a large city and then intentionally producing it, the Roman state was in the position of scrambling to keep up with the demands of the creature their own actions had inadvertently created.

Who Are Cities For?

So how does Imperator‘s model of ancient urbanism stack up to how we understand the expansion of cities in the Hellenistic period to have actually proceeded? On the one hand, Imperator‘s design places a substantial focus on the development of urban centers, from the graphical presentation of them on the map to a robust set of mechanics meant to allow the player to shape their development.

On the other hand, Imperator seems to fall into that overly narrow vision of monumental urbanism.

What I mean by monumental urbanism is that, as noted, the city in this vision serves not as much as a place for people to live as a canvas for the painting of state power (at some point, we’ll loop back around to ancient-themed city builders in the Impressions tradition, because they are very focused on this vision of urbanism). Sometimes that elite power is strategic, with the city built by the state as a machine for the production of armies, monuments to royal legitimacy or bulk trade goods, but it also often more narrowly monumental, with the city – its size, its monumental architecture and wealth – serving as a statement of royal power. In short then there are two components to this: that the city is produced by state power and that the city is created for state ends, either strategic or monumental.

Imperator certainly imagines cities as produced by state power. As noted, cities come into existence in Imperator almost no other way. There are no background processes whereby a territory with sufficient population will begin urbanizing without the player’s specific action and in particular without the expenditure of political influence, a currency specifically meant to communicate the burning of political capital. Certainly some cities in antiquity did come into being this way: military settlements, veterans colonies and so on. But they were comparatively few. Most cities in antiquity were not produced by state action but rather were organic creatures created at local initiative for local purposes.

And precisely because the player is given so much power to shape the development of urbanism in their state, they are going to tend to shape that development for state purposes. Nearly all public buildings in Imperator have to be built by the player, at state expense. Events where local elites decide on their own initiative to build something are both random (and so the building often makes little sense) and rare. Instead, with just a few button clicks the player can pick an existing city, demolish all of its existing public architecture if it doesn’t suit their purpose, and then completely remake the city. Notably, because buildings not only enhance pop production but also shape those ‘desired ratios,’ that means that the player can step in and reshape not only a city’s physical architecture, but also its economic role. As a result, cities are almost always built to state purposes in Imperator.

And those key purposes are more or less what you might expect: cities exist in Imperator to be efficient centers for the production of the key state resources (revenue, manpower, research) and to stabilize local provincial loyalty. And most players, pulled by the fantasy of building Rome – and encouraged by the game’s graphical presentation – are going to want to build a mega-capital just to see it sprawl, gloriously, over the map (and even if they didn’t want to, the game mechanics strongly encourage them to do so anyway). Cities thus exist to produce resources for the state and as monuments to state power.

What cities in Imperator do not really exist to do is house people or even really provide for economic purposes. Unlike Victoria III, in which the limited amount of arable land means that, as population grows, folks who cannot be farmers end up in cities as a result of economic processes, Imperator‘s map starts sufficiently empty and population grows sufficiently slowly that this never really becomes an issue. By contrast, we’re pretty sure rising population in Italy was one of the things driving Italian urbanism.¹³

Likewise, the player doesn’t face situations where local urbanism is being pushed along either by local elites looking to improve their status in this new, larger imperial polity or by the need to provide markets in order to pay taxes or move goods. Simply incorporating non-state peoples into an imperial polity that collects taxes in coin should begin pushing those people towards urbanism, because they need to sell their goods in markets to get the coins to pay the taxes. But unless the player makes this happen, in Imperator it doesn’t happen.

The result is a vision of urbanism that is much too ‘top down,’ and far too focused on the role of the state. I suspect this is a result of the tendency in games towards what Kristoph Rosencreutz calls player autocracy (a concept I have found really useful in discussing historical game design – I even snuck it into a peer reviewed chapter I hope will appear at some point)¹⁴ Player autocracy is simply the design ethos where, to expand the player’s agency and empowerment, they are given direct control over things that might only ever have been indirectly controlled by historical leaders. And it isn’t hard to see why this is a common element in design: a game in which you had as little control over your country as real leaders might not be as fun.

At the same time, should Paradox revisit this period, I’d like to see them take a bit more design DNA from the Victoria titles (a reminder that I am one of those people who really likes Victoria III) where the player is given the ability to reinforce or direct social and economic processes, but is hardly the ‘sole agent’ driving them. A system, for instance, whereby exposure to state polities and especially state taxation in coin might cause cities to begin forming organically, while the player can still found a smaller number of cities the old fashioned way would be fairly welcome. Likewise, cities should probably start to fill their own building slots organically over time, acquiring one by one the standard elements of respectable ancient urbanism.

Of course the player as the state could still influence these processes. The Roman authorities seem to have mostly welcomed the urbanization of their provinces (even as they were sometimes annoyed by the upward striving of their elites-becoming-Romans) and it is pretty clear that part of the reason provincial elites are learning Latin or Greek, beginning to dress like Romans and fill their cities with Roman-style public buildings is that in so doing you could get more respect and a better deal from the Roman authorities. Roman investment into infrastructure – roads and ports, especially (mostly for military purposes, mind you!) – also encouraged rising urbanism and could shape it, since those roads and ports were channels for the economic activity that cities served as markets. The fact that the Romans built lots of of roads and ports (again, mostly for military purposes, but these sorts of infrastructure projects also clearly served an ideological/cultural purpose for the Romans, a demonstration of one’s virtus by compelling the natural world into a shape more fitting) almost certainly intensified the process, as did the Roman provision of things like military engineers to support civilian building projects.

All of which is to say, I think these processes could be better modeled less as the player tending a ‘garden’ of plants they have deliberately and carefully arranged and more as something like managing natural spaces: the player doesn’t pick and choose every plant, but can make guiding changes with large implications over time. Because fundamentally, ancient cities weren’t for large imperial states, they were for people, who lived in them (particularly, but not exclusively, the local elites whose power shaped them).

Of course, all of this talk about elites is bringing us back around to politics, and that is where we will turn for the last part of this series: how does Imperator imagine ancient politics: who participates, how do political regimes change over time and especially why do they fail? But before we go, I do want to note if any readers are interested in reading more about pre-modern urbanism, I do have a two–part series on the structure of ancient cities you can check out!

1. I dislike this plural and would prefer it to be metropoleis, but alas, in English, it is not.
2. There are escalating costs in a few other things we’ll get to, but they do not overwhelm the effects we’re about to see.
3. For a discussion of the process of urbanization focused on pre-Roman Gaul, see G. Woolf, Becoming Roman: The Origins of Provincial Civilization In Gaul (1998).
4. For more on early Iberian urbanism, see M.C.F. Castro, Iberia in Prehistory (1995). J.S. Richardson, The Romans in Spain (1996) covers the urbanization of the Roman provinces, although not in the same depth as Woolf does for Gaul.
5. That is, the governor does not live here, the local elites do. The governor comes through once a year to look down his nose at these fellows, make sure they get the taxes collected and solve any problems they have, and then – as soon as he can manage – leave again.
6. As suggested by N. Rosenstein, Rome at War: Farms, Families, and Death in the Middle Republic (2004).
7. In The Field and the Forge: Population, Production and Power in the Pre-Industrial West (2003)
8. This process is most completely set out in N. Morley, Metrpolis and Hinterland: The City of Rome and the Italian economy: 200 B.C. – A.D. 200 (1996)
9. A notable study of this in Etruria is Ward-Perkins, “Etruscan Towns, Roman Roads and Medieval Villages” The Geographical Journal 128 (1962). Because the road to Rome became the key economic thoroughfare, urbanization in the region shifted towards it. On the broader effects of Rome’s expansion on the integration and economic conditions in Republican Italy, see also S.T. Roselaar, Italy’s Economic Revolution: Integration and Economy in Republican Italy (2019)
10. Carthage’s position on a major natural harbor in a coastline with few of them is both part of why it is refounded and part of why the site surges back into importance so quickly.
11. If you are asking, “why not use the river water?” the answer is that for such a large population center, the river is where your sewage and run-off water goes and so not great to drink from.
12. Computing an average of grain prices is difficult because there really was a lot of variation, but G. Rickman looks at this question, along with a general discussion of Rome’s grain supply in The Corn Supply of Ancient Rome (1980).
13. A point once endlessly debated because it hit at some of the key questions between the ‘high count’ and ‘low count’ of Italian population in the Roman period, however at this point even the most stalwart ‘low counters’ admit that the evidence for meaningful population growth in Italy during the Republic is too extensive to ignore or deny (while the ‘high count’ has largely fallen away entirely). On these debates, see De Ligt, Peasants, Citizens and Soldiers (2012).
14. It’s in an edited collection and the publication of those sorts of volumes can be infamously slow. Also, to be clear, by ‘snuck’ here, I mean I used it and cited it to his video, which means I also got a youtube video into the bibliography of real scholarship (and not for the first time!).