Imperial Scale to Manorial Scale in Mapping and Rulership

A few days ago I talked about Fractal Game Design and mentioned how it relates to Demesne-Level play. Yesterday I dug into Hex Population Density.

Three-Level Hex Map
Three-Level Hex Map

Today I’m going to try to bring these things together.


I am using a very simple model for the relationships between demesnes. Real feudalism is horribly complex and hard to understand, let alone explain.

manor is a home of a member of the gentry, and refers also to the surrounding lands he is responsible for. A 1.5-mile hex with an area of two square miles. Lowest practical level of detail as far as management is concerned, but may be subdivided into sixteen 3/8-mile (80 acre) hexes for allocating terrain and the like, if desired. Often owned by a knight, dame, or other gentry (usually called just ‘lord of the manor’, but ‘baronet’ is another possibility), often managed by a steward when the owner had duties elsewhere (such as a higher title). Nominal population maximum, without support, is about 200 people.

barony is a collection of manors under a single ruler, a baron or baroness (higher gentry, not quite nobility). Nominally a six-mile hex with an area of 32 (actually 31.18) square miles. Nominal population maximum, without support, is about 3,200 people.

An earldom or county is a collection of baronies under a single ruler, an earl, count, or countess (depending on naming scheme used; either way the lower end of the nobility). Nominally a twenty-four mile hex with an area quite close to 500 square miles (498.8, close enough). Nominal population maximum, without support, is about 50,000 people.

A duchy is a collection of earldoms or counties under a single ruler, a duke or duchess. Nominally a 96-mile hex with an area of about 8000 (7981.2) square miles. Nominal population maximum, without support, is about 800,000 people.

kingdom is a collection of duchies under a single ruler, a king or queen. Nominally a 384-mile hex with an area of about 128,000 (127,700) square miles. Nominal population maximum, without support, is about 12,800,000 people.

An empire is a collection of kingdoms under a single rule, an emperor or empress. Nominally a 1,536-mile hex with an area of about 2,000,000 (2,043,210) square miles. Nominal population maximum, without support, is about 200,000,000 people.

This is a gross simplification to illustrate the relative sizes of things. In principle each demesne is on a scale sixteen times the size (area and population) of the one below it.

These are all nominal values, and in practice may vary quite a bit. It is entirely possible that a barony contains only a smaller (or larger) number of manors than is shown here, and it is possible that a kingdom may consist of a mix of duchies and earldoms (and possibly directly baronies). In any case a demesne larger than a manor is unlikely to be hex-shaped, instead being a contiguous collection of smaller demesnes. Historically this is not required, but I think I will consider a noncontiguous collection of smaller demesnes as separate demesnes.

Notice the population figures. They’re pretty big, and assume the entire region is arable and settled.

Demesne-Play – Settlements


Determining the size of settlements possible, given the population, can be a bit of a challenge. I don’t want to be bothered to try to do demographic analysis and try to figure out how to transport food around. However, I observe that larger cities require both more support, and more infrastructure. Both of these might be represented after a fashion by the size of the demesne, and what is around it.

I grabbed the Pathfinder settlement definitions and realized they’re probably kind of silly, so I’m going to largely ignore the populations there and steal the names.

  • A manor is nominally two square miles, and can support on the order of 100-200 people pretty easily in most reasonably arable places. This is one or two villages, but no more.
  • A barony, a collection of manors, might be able to support a couple of larger villages (200-250 people or so) or perhaps a small town of 400-500. I’m nudging things up a bit, but I have a reason.
  • A earldom/county, a collection of baronies, might be able to support a couple of small towns, or a larger one.
  • A duchy, a collection of earldoms, can probably support a couple larger towns or a small city.
  • A kingdom, a collection of duchies, can support some smaller cities or a larger city.
  • An empire, a collection of kingdoms, can support several larger cities, or a metropolis.

In real-world history, there have been cities (such as Paris and Rome) with populations in the hundreds of thousands, and even over a million. I don’t think I need to get quite this big here, but the sheer landmass possible in this model is pretty big and should be able to produce a lot of food.

With the larger demesne designations comes more infrastructure. A manor likely has simple dirt paths, a barony might have dirt roads, and so on. The Imperial Highway is an incredible work and spans the empire, with focus on the capital.

Which brings me to an idea of how to manage settlement placement.

Let’s say that any settlement requires at least four (relatively) nearby settlements the next size down. A settlement may only support one larger settlement (though I suppose if needed a single settlement could count for half toward each of two larger settlements.

I was reviewing this post with regard to another one I’m writing, and I started to correct this one to replace ‘settlement’ (in the context of supporting a larger settlement) with ‘hex’… when I realized that if a hex has more than one largest-allowed settlement in it, perhaps it can actually support more neighboring hexes. This might be a foolish thought, since a hex with two or three of the largest-allowed settlements is probably pushing the boundary of what can be supported in the immediate area anyway. I’d like to remember this anyway, though.

Baronies, market villages
Baronies, market villages

A small town (250-300 people) would not exist in a manor without support. If a manor has at four neighbors with villages (that are not otherwise claimed for this purpose) it’s village could be ‘upgraded’ to a ‘small town’ (grow to 250-300 people). It is not necessary for all the manors to be part of the same barony. In the image to the right, assume that this is fairly arable land (with some forest and the like, sure, and some streams and small lakes); all manors on the map have villages. The stronger hex colors are manors with small towns, the paler colors are their supporting villages.

This puts a small town within a few miles of any village identified, and not really that much farther apart (it looks like 2-3 hexes, 4-6 miles, between nearby small towns). In addition to the regular market, you might find a service not present in all villages, such as a grain mill or other — it depends a lot on what the manor lords and baron allow, really.

The same mechanism might be applied the next level up. A small town in a barony could be upgraded to a large town if there are four adjacent ‘unclaimed’ baronies that have small towns. Assuming these are centrally located in the barony (they probably won’t be) and each hex is sixteen times the area (that is, four times the radius), this will place the large towns some 16-24 miles apart from their neighbors.

An earldom or county could have a large town upgraded to a small city if there are four adjacent ‘unclaimed’ earldoms with large towns. At four more times the distance, this will be about 64-96 miles — two to five days of travel.

A duchy could have a small city upgraded to a large city if there are four adjacent ‘unclaimed’ duchies with small cities. These will be nominally 250-400 miles or so apart, perhaps two to three weeks distant.

A kingdom could have a large city upgraded to a metropolis if there are four adjacent ‘unclaimed’ kingdoms with large cities. These will nominally be 1000-1600 miles or so apart (this is getting uncomfortably big now — this could be a couple _months_ apart!).

And finally, an empire could have the infrastructure, possibly, to upgrade a metropolis to something crazy… though I’m not sure quite how big that might be.

Hmm. Perhaps instead, a settlement maximum size could be increased by drawing from adjacent demesnes. Given two manors, one could have a bigger population. Given three, one could be a bit bigger yet than the other two. Given four neighbors you can draw on you might be able to get to the next larger settlement size, and given six such neighbors go a bit bigger (but not yet the size after that).

Probably too much trouble.

I see, though, that this comes kind of close to my original description. A manor has a village or two depending on what makes sense for the people there. A barony might have some small towns (I see one up there with two plus two halves, one with two plus one half, one with one plus one half — and I might have hexes assigned either to one larger hex or the other by adjusting them slightly so they aren’t quite hexes any more, as described in Fractal Game Design the other day). The same patterns can apply, with earldoms/counties having 1-3 (more or less) large towns, duchies have 1-3 small cities, kingdoms 1-3 large cities, and so on.

Closing Comments

The structure I am devising here looks like it will come somewhat close to modeling the relationship between ‘farmers’ and ‘city folk’. Depending on the size of the settlements at each level it might be a little rich (historically ‘no more than 10% of the population could be not-food production’ is a pretty commonly-accepted value), but it looks reasonably credible if you don’t know that :)

Of course, ‘reasonably credible’ will depend in large degree on how big the settlements are. Spreadsheet time! No, no, in the morning it can be spreadsheet time. If I want to end up with something like 10% of the population “not farmers”, and I can divide townfolk up into equally-sized groups (for instance, every four settlements can support one that is four times the size) I could end up with the ‘Empire-sized city’ about a thousand times the size of the ‘Manor-size city’ (large village/small town).

This does not yet feel polished, but I think I’m headed in a good direction. I suspect I slipped somewhere in here, to be honest, but it’s after midnight and I’m up at silly o’clock, so I’m going to leave it here and revise it after it’s had some time to percolate in my brain.

Suggestions are welcome, of course, they help the percolation.

No related content found.


  1. GreyKnight

    There are two “levels” of baron: lesser barons are what you seem to be describing here (actually rarely called barons since a fair while back, more usually “lords of the manor”). Greater barons (just “baron” really) are considered nobility, of which they’re the least rank. Interestingly I think (IIRC) the noble status of “lesser” barons hasn’t been formally confirmed or denied in the courts ever, so they’re in a sort of grey zone.

    You could apply a concept of “levelling up” if a player character enters the landholding system. A “lesser baron” as listed above could be promoted to the nobility and become a “greater baron”; from there he could be promoted to viscount. Advancing further to Earl/Count should come with an appropriate increase in territory; presumably he takes over the earldom his erstwhile barony/viscountcy was contained in, and the latter becomes the seat of the earldom. Similarly, an Earl could be promoted to a Marquis (female Marchioness) and his earldom becomes a march. An advancement from there to Duke comes with an increase in territory again.

    There is precedent for what you describe when you mention a kingdom containing baronies directly (a barony held “in-chief”). AFAIK subdividing a manor holding is legally disallowed (IRL anyway). Also, the female equivalent to a knight would be a dame.

    • It has been so long since I’ve looked at this in any depth… I did say I was grossly simplifying. :)

      I honestly forget what reference I looked at for these titles. I don’t think I was foolish enough to look to a gaming supplement for it.

      ‘Baronet’ is a term I’ve seen, before ‘Baron’. I’d assumed the manor lords had a title as simple as ‘lord of the manor’, and many of them are absent most of the time anyway.

      I am using the labels primarily to give a point of reference for later. Manor < Barony < Earldom < Duchy < Kingdom < Empire. In practice I suspect I'll treat it as a matter of feudalism -- as I recall each increasingly higher rank below king happened because they needed a word for "a bigger one of them". A baron was a powerful lord, an earl was a powerful baron that had other barons under him, a duke was a powerful earl with other earls under him, and so on. I understand that 'king' was in practice basically "most powerful duke the others couldn't quite cross... yet", and might have the dukes 'under him' in name only, really, something closer to an alliance rather than outright subordination. The manor wouldn't truly be subdivided, in that the land does not get split. For the purpose of mapping and element placement I figured it might be worth treating it as if it were constructed of smaller hexes that could be marked by terrain and cover (hill, forest, field, etc.) -- 80 acres is actually a pretty nice size to work with, you don't really need to get down to five-acre lots, let alone quarter-acre city lots. Huh. That actually comes kind of close to how land is around here. You can find half- and full-acre lots, but normally city lots are a quarter-acre, and once you start muttering 'acreage' you're usually around five. Creepy how these more or less arbitrary numbers keep mapping to things I actually see.quite

      • GreyKnight

        A duke is subordinate to a king, if that boundary gets blurred it would be during the more “intrigue-y” portions of history. Also often the king and his heirs might be dukes too, but that’s another matter. I discuss multiplicity of rank in the article I linked before.

        I assumed you were skipping a few ranks to match things to your desired territorial scales; I was just mentioning a way you could incorporate the others without changing the scale.

        A baronetcy isn’t a peerage; they rank below barons but above knights (except certain special knightly orders). I usually just ignore them in an RPG context to be honest.

        Two other titles which might be of interest: an archduke ranks above a duke (but still below a king), and a grand duke is similar. I’m not sure where they rank relative to each other. A grand duke can be a sovereign in his own right (AFAIK Luxembourg is the only Grand Duchy remaining nowadays), not sure if the same applies to an Archduke but I think not.

        • Duke-almost-peer-to-King was, as I recall, a thing that could happen from time to time when you had a weak king, and is basically not worth tracking here, agreed. As I said, feudalism is complicated and hard.

          It occurs to me that if I really wanted I probably have enough slots to have hex scales double at each step rather than quadruple, but I liked how the quadruple let the hex scaling be consistent from level to level, and minimized the number of halved hexes… and within the limitations of my data, appears to work.

          That it comes close to aligning with the Empire (Alderac Entertainment Group supplement for D&D 3e) 20:1 scaling is barely relevant at all… but that is where I got the idea of scaling the land and management rules like this.

          Honestly, I used the set of titles I did because they are the best known, I think. Lord, Baron, Earl or Count, Duke, King. ‘Baronet’ is hardly know, Marquis is hardly a common title, and so on, so I didn’t bother. However, I think I might keep them around as intermediate values for powerful examples of the rank below, or weak examples of the rank above.

          Lord < [Baronet] < Baron < Viscount < Earl/Count < Marquis < Duke < Archduke/Grand Duke < King < [something?] < Emperor That looks like I could indeed offer an additional rank between each of the ones I've already got. I may want to consider that. OTOH, in Echelon terms I could even have 'Nobility' as a talent (as it were): Lord is 'Basic Noble' (you rule a manor), Baron is Expert, Earl/Count is Heroic, Duke is Master, King is Champion, Emperor is Legendary. I have no idea what they do, though. However, that could be an interesting thing to try. If it comes to a struggle, odds are someone at the appropriate tier, if such a person exists, could end up with the crown. Almost anyone can become a Lord through ability, but it takes a pretty major figure to earn an empire. Just a thought.

          • To clarify my earlier comment, a “lesser baron” is basically a manorial lord; it’s not the same thing as a baronet. The latter are basically a historical accident, which is why I wouldn’t bother putting them in a fictional work set in a different universe (basically James I wanted to raise some extra money, so he invented this rank as something to sell!). I don’t know how a manorial lord and a baronet rank relative to each other, actually; I’m not 100% sure that they have a well-defined relationship.

            On the subject of well-knownness, I imagine the most famous Marquis is the Marquis of Queensbury. :-)

            The Echelon talent is interesting. I made something of a 3.5e prestige class for nobility/royalty some time ago, I could look it out if you’re interested. As I recall it included things such as divine visions of threats to the land, and the ability to put quests on people. I don’t think I included anything about pulling swords out of stones, however.

            I agree with the nice consistent scaling behaviour of the current setup; my intent was something like what you mentioned here, using the other titles as “intermediate values”.

            Other ways you could advance through nobility besides a power struggle include simple vacancy by death/abdication (a chain of people “move up one” to fill the gap) or expansion (we just conquered Hrajel, we’ll need to appoint an Earl, some Viscounts, and a bunch of Barons to govern the place). There’s also transfer (the old Marquis of Aqafi was found to be corrupt and was stripped of his lands by the king; you were considered worthy of taking the reins) and inheritance (the old Duke was fond of you and had no children of his own). Also, a title which falls vacant can be reabsorbed into the Crown. The king (possibly a later monarch) could then reinstate this title anew if he wants to grant you an honour.

            • When I went to google, ‘marquis ‘ turned up immediately ‘marquis de sade’ and ‘marquis de lafayette’ first.

              I explored a bit introducing intermediate steps (lengths doubled, areas quadrupled) and it did some decent things with settlement sizes, but makes for an icky map. 16:1 is a nice order of magnitude (heh, ‘hex map’ -> ‘hexidecimal order of magnitude’… I keep turning up weird coincidences here) for what I’m doing.

              I think I’ll keep the intermediate ranks as discussed, they’re kind of like a pip marking a slightly higher designation without necessarily full advancement to the next rank. “I’m bigger than you” sort of thing.

              Oh, nice. I like what you’re describing with the prestige class you mention, and the powers could indeed be appropriate. I imagine this would be a capstone talent rather than cornerstone or common, and is not itself a prerequisite for rank. Anyone (well, with the right parents…) could inherit a title, only the proven right-wise ruler would have this talent.

              • Yes, the prestige class had it that way around too (you must have the rank in-world as a prerequisite for gaining each level of the class). Echelon’s capstone talents are probably a better model for this sort of thing, it felt mechanically a bit awkward doing it as a prestige class.

                • Maybe something like the Dragonlance Knights of Solamnia classes? Each rank might be a three-level class that has prerequisites to get into (that don’t include previous ranks of the class — if you’re badass enough you could grab the throne itself, like Conan did). I might create it such that perhaps the first two (or all three) levels of each such class is enough to get you the prerequisites for the next class.

                  Thus you might go: Fighter 1-5; Baron 1-2; Earl 1-2; Duke 1-2; King 1-3; Emperor 1-5 (5+2+2+2+3+5 = 19 levels). Or go to Barbarian 11, kill the king and cow his followers, and carry on at King 1.

                  • David Lamb

                    This (and the similar idea of using Echelon tiers) directly associate in-game authority with level, which doesn’t interact well with the notion of a hereditary nobility. An 18-year-old can be King (or younger, but then there’s the issue of regency). It works better with something like GURPS Rank and Status, which aren’t associated with power level (other than having enough character points to buy them in the first place).

                    • I wasn’t intending to conflate the two. I was looking more at the establishment of such rank, rather than inheritance.

                      That is, a character with the Duke tier of Noble did not inherit his rank, but through his own ability carved it out of the wilderness, or the bodies of his enemies. The character who simply inherited his duchy may have equivalent rank in court, and may be a sufficiently capable administrator, but likely is not the same figure of power that the Noble(Duke) is. Or maybe he is similarly powerful, but he is a powerful wizard or the like instead — feared for his power or respected and admired for his power, but not noble in the same way.

            • Oh, hey, I just realized — the intermediate values do have a use.

              Let’s say that in order to achieve higher rank, you must have as vassals some number of members of the rank below (three or five seems like a good place to start). This should give you enough land to support you in your new role.

              For example, if we use ‘five’ as the required number, a duke must have at least five members of the lower rank (earl, count, or marquis) sworn directly to him (barons and viscounts don’t count). He may have more, though a truly large number will start to become unmanageable.

              If he has at least one such vassal, but fewer than five, or doesn’t meet other requirements (perhaps he needs a large enough city in his demesne) he cannot be a duke. However, he is notably more powerful than most other earls, so he might be elevated to ‘marquis’. He has a mix of earls/counts and barons (since he had to have at least five barons as vassals to become an earl himself), so he’s got a growing power base.

              So… attack a neighboring earl and force him to submit? Elevate a baron to earl (assuming there is a baron that qualifies)? Sponsor activity (such as settlement and development) to help his vassals grow enough to become eligible candidates for advancing him?

              This feels pretty good, GreyKnight, quite good indeed.

              • I edited my above comment to include a note about various ways an opening can come up for you to increase through noble ranks, you may have missed the edit during your comment. It’s relevant to your last point. :-)

                I feel I should point out that you don’t necessarily have to have “support” from vassals to advance (unless you’re using some sort of democratic nobility, but I have no idea how that should work). Honours (where not just inherited) are ultimately handed down by the king, so if he decides one earl gets promoted to marquis over his peers, then the rest of them can like it or lump it. :-) I’m not clear on your last example of “help his vassals grow enough to become eligible candidates for advancing him”.

                • True enough, elevation was at the whim of the liege. I was thinking of gamability, how to make it so you ‘earn’ your rank.

                  So, if you need five vassals of the rank below in order to advance, and you only have four, you would have to either beat another one into submission, or get one of your own elevated — who deserves the elevation. Archduke gets one of his vassals earls (marquis) elevated to duke, he might now meet the requirements to be king. Except there already is one.

                  Right, now I remember why I didn’t involve the king.

                  This needs some thought.

                  • The way I see it, if you’ve attained the status of archduke, you’re either happy with your position (out of genuine loyalty, or because you don’t want to be king — uneasy lies the head and all that), or you’re plotting a way to seize the crown by unscrupulous means. In neither case do you need the king to promote you. ;-)

                    If the method of succession permits it, perhaps you’ve been adopted as heir (in which case you’ll take the throne in due time).

                    Erfworld had (I can’t find the exact episode right now) a mention of a Royal side using a strategy of “forking” itself, where the heir took some of the territory to create a new side which then allied itself with the original “parent” side. The heir thus becomes a King in his own right. Mind you, this was based on some exploitation of Erfworld physics (basically, certain game rules are natural laws there, so things like a “side”, “hexes”, and “turns” are actual concrete concepts with measurable effects).

                    • At this point, intermediate ranks might serve in a couple of ways.

                      In a society where a higher ruler exists that can elevate you, that you are special within your rank. You have precedence over others of your nominal rank, but are clearly a step below your betters (your liege and his peers).

                      In a society without such a higher ruler, and where the ranks actually have specific effect mechanically (such as “rulership over a hex |this| big, with the infrastructure that implies, instead of a big pile of smaller hexes that are a headache because I haven’t delegated as I probably should”), the intermediate ranks might indicate “partway to the next rank”. The marquis may or may not use that title in practice (the Celts had many, many kings that do not fit the definition being used here), but having a marquis means you have an earl who is a little hungrier and ambitious, and has taken steps to advance further than his nominal peers. When he builds the infrastructure needed to advance (has enough vassal earls and whatever else is required) he can be advanced to ‘duke’ — though as above, it likely wouldn’t be called that.

                      Echelon makes this a little easier, perhaps, by divorcing the effects from the real-world rank labels.

            • I was thinking ‘High King’.

              I think in part I was thinking along the lines of starting a new demesne, so there isn’t anyone above you to elevate you to power (or a high council that could it, such as when no longer political unit even exists).

              • David Lamb

                I think I’ve only heard “high king” in a context where the “kings” he was king over were more like dukes or even lesser nobles — Arthur as High King of Britain, with lesser kings (sometimes) beholden to him.

                • I’m open to other suggestions. The High King having precedence (and in this context, institutional power) over ‘normal kings’ fits the role definitions, though I agree, historically it applied to situations that would be modeled here with rather lower ranks if based on demesne size. I think I mentioned — google plus, or here in a post or comment — that the ranks identified are intended to be suggestive of the relative demesne size and development, rather than how they were used historically. That is, depending where you look you could easily expect to find many ‘kings’ that in these rules might be more accurately considered ‘earls’.

                  • David Lamb

                    Aside: Cyrus the Great took on the title Shahan Shah, King of Kings, when he became emperor of Persia. “Imperator” came from an honourary Roman military title; King of Kings makes more sense IMHO.

  2. You mentioned that there may be different infrastructures (e.g. an earldom has better roads than a barony). However, the roads which are part of the earldom will be part of at least some of the baronies!

    I imagine road quality will be based more on what’s being connected; a road between a city and another city will be large and well-maintained, roads between a city and nearby towns may be lesser (and might just be branches off the inter-city road), and so forth.

    • Indeed some of the baronies will get upgraded roads. Not all will, of course, but ultimately the Northern Imperial Highway passes through many manors.

      And indeed, the more accurate indicator of road quality is the endpoints of the road segment. It’s not as simple as “all roads built by the kingdom are big roads”. However, as described in this post, getting the bigger cities requires a level of infrastructure implied by the demesne size. You must have big cities to have a big road, you must have a duchy or kingdom to have big cities, therefore big roads exist when duchies and kingdoms do.

      However, no matter how many manors or baronies you have, you won’t have the big roads and/or big cities until you get the governmental structure to organize it. This might not be completely realistic, but I’m trying to keep playability in mind and this makes it pretty easily. Somewhat simplistic, and I’m okay with that.

  3. Pingback: Kingdom Example: Random Kingdom | Keith Davies — In My Campaign - Keith's thoughts on RPG design and play.

  4. Pingback: A-Z April Challenge 2013 | Keith Davies — In My Campaign - Keith's thoughts on RPG design and play.

  5. Pingback: A-Z Challenge Forfeit | Keith Davies — In My Campaign - Keith's thoughts on RPG design and play.

  6. Pingback: Founding of a New Sandbox | Keith Davies — In My Campaign - Keith's thoughts on RPG design and play.

  7. Pingback: High-Level View of the Sandbox | Keith Davies — In My Campaign - Keith's thoughts on RPG design and play.

Leave a Reply

Your email address will not be published. Required fields are marked *