October 19, 2022

Detail XV: Farms in the Hinterland

What lies between the settled areas and the "true" wilderness? And how can we determine this from the wilderness maps?

Let's begin with an example region featuring two settlements. The settlement to the left is about 6000 persons, and the right much smaller, about 130. These numbers refer only to the urban population of the settlement proper. This population number is the source of infrastructure numbers, and for that reason is fixed.

lighter "wilderness" hexes can be considered cultivated

Of course, not everyone lives in the settlement, and so we consider that 85% of the 20hex's total population is rural. This comes out to about 15000, and then increases to 18000 due to market spread. Already we have a problem: how are all these people fed?

As a note: the intent is not to micromanage the result such that every individual of hundreds of thousands of people each have their determined job, place, and caloric intake. However, we do want to ensure that our results are consistent and coherent. The players are not interested in the type of detail produced by slavery to the numbers. They want to know (usually) how far outside the town is the dungeon? It is unlikely that a typical dungeon (as understood by most parties) would remain uncleared within several miles of the farming hinterland. We must make the numbers serve us instead.

For some reason, the online worldbuilding community is convinced that a square mile of farmland can support 180 people. I'm not sure what the origin of this number is, and I can't exactly say that it's wrong, but it certainly ignores the incredible variability of arable land and agricultural technology. But let's pretend it's a viable average at least, so we can see where we land.

Our total population of ~24000 people, then, will require some 153 1hexes of farmland outside the city (each 1hex is 0.866 sq mi). The 20hex itself is only 400 1hexes. That certainly reduces the tight feel of the map above into something essentially cleared of all trees and wilderness.

Alternatively, we can take the 6000 as the total population. That requires about 41 1hexes to feed. There are 26 1hexes on the above map which are adjacent to settled hexes. So let us assume that the heavy cultivation of those hexes (including farming, orchards, timber management, and hunting) added to any small-plot cultivation of the denser settled hexes yields the caloric requirements of the 20hex.

Quite a lot of work to end up back at the same square. However, there are some key takeaways from this experiment nonetheless.

  1. The adjacent wilderness hexes are not pure wilderness, and in fact are probably heavily farmed. No orc lairs here.
  2. For this map in particular, the smaller settlement, although tiny in comparison to the larger, controls a slight majority of the productive land (which it likely sends a few miles down the road to the larger settlement). This is probably a collection of holdings of 1-3 knights who are tied in some way to the larger settlement.
  3. As a corollary, the smaller settlement will have about the same number of agricultural class with a much smaller merchant/elite class. They will depend heavily on the larger settlement for military protection and most market services.
  4. There are about 75% fewer people in these hexes than I'd previously thought. That will have significant implications if questions arise about hex wealth or the mustering of an army.


  1. I've done a bit of research into this as I found the popular "medieval demographics made easy" to be highly suspect where the actual numbers are concerned. I have yet to form an actual generation system out this yet, but here's some numbers to consider:

    A town of 3000 people requires the land of 10 villages or 8.5 square Km.

    3 acres of good land, or 4 acres of average land will support 1 person

    Land cultivated for agriculture will feed 10 to 20 times as many people as land cultivated for livestock

    In a survey of England with undulating land unmarred by natural barriers, settlements (villages) were about a mile apart, and the location of settlements from the time of the domesday book to present was relatively stable

    1. That matches well with my own calculations. 4 acres/person for 6000 people would be 37.5 sq mi (43 1-hexes).

      As mentioned above, I want to be consistent and believable, but I'm also not sure how much benefit is to be gained by nitpicking the perfect number. I know that I would *enjoy* finding that number, but there are many other things to occupy my attention for now.

    2. Yeah so much is dependent on the quality of the soil, what crops exactly are cultivated, techniques used to till and fertilize, etc etc

      It's best to find an average that can accommodate for differences on a wide scale

  2. Not correcting you, just throwing in data that's prescient. Present day ratio of seed to yield is 40:1, with a hectare producing an average of 4,079 kg. of wheat. Medieval yields were regularly reported as 3:1, sometimes 4:1. If we take the bottom end, that's 305 kg of wheat per hectare. A poor man's fields were 15 acres; 30 was more common. In feudal times, half this was given to the local lord as tax; if a farmer only had 15 acres, it usually meant squatting on land that had no lord, which was a fairly large part of Europe. 15 acres = 6.07028 hectares.

    In wheat, then, a farmer produced about 1,857 kg. that his family was allowed to keep (assuming the farmer either started with 30 acres and was taxed, or had 15 acres without a tax). 1 kg. of wheat provides 3,640 calories. We can quibble or not on whether wheat in medieval times was as rich as it is presently; we'll never actually know. Arguably, medieval wheat was MORE rich, as it was raised in a world without nearly so much toxins in the atmosphere. Using the 3,640 number, then ...

    15 acres without tax, at the low end, produces 6,759,663 calories annually. At 2,500 calories per day, this is enough to feed 7.4 less-than-plump adults; it's enough to give them energy to work. At 640 acres per square mile, this means that a square mile of grown wheat will feed 158 persons. At yields of 4:1, 210. Divide this number in half and you get 180 persons per square mile.

    I honestly did not know that was the result I'd get.

    Of course, this ignores that every farmer grew a vegetable patch in the lee of their lawn, owned one goat and chickens, shared with others the occasional slaughter of pigs, hunted, gathered mushrooms and other wild fare, fished and so on, as Shelby rightly points out. So the number, based purely on staple farming, is purely wrong. Chances are, at least 40% of a farmer's diet was not provided by the actual wheat in the field.