This is a medieval European (fantasy) setting, with the following (highly tentative) parameters:

  • Each town is vaguely autonomous, and have their own municipal government system.
  • Towns do trade with each other, but are not exactly dependent on said trade.
  • Towns vary in size from the rural village (pop. ~400) to a more prosperous city (pop. ~6000).
  • Town-vs-town warfare is historically rare and often not very substantial; any border control of towns would mostly focus on rogue bandits and the like.

Given these parameters, what would be a feasible range of distances for towns to be separated by?

11 Answers 11

up vote 106 down vote accepted

Distances between Medieval towns varied quite widely, as did the population and size.

I think the estimate, and this is a low estimate, for England was that there were over 3,000 deserted Medieval villages that they hadn't discovered.

Now, there are also going to be communities in between smaller than your 400 person population--groupings of families and trading posts along major roads.

Also, not that many people would have lived IN some of these small places. Instead, there were a number of shop keepers, a tavern/inn, maybe a single guard and people who lived and worked inside the village, but for the most part, the population that you might see there would come from those coming in for the day to trade/buy/sell goods, who will later be going back to farmsteads within a day's walk.

A town of, say 3,000 people actually would need about 10 villages and their surrounding farmsteads to support it. SEE LINK

This isn't just a straight line on a road with miles between. A large population place, such as a town or city with like 1,000 people, will actually be surrounded by "satellites" villages adjacent to these larger places. This happened in London--but as London grew, these communities were "eaten up" by the city itself by around the 1700 and 1800s. As to closeness, this rule about "a day's walk away" that people seem to be spouting all over the internet,--that goes right out the window. Many of these satellite villages would only be a mile or two away from their towns--or even less, some would be a scant 1/2 mile. They were not always on a main road, but they would be next to resources (such as water).

Consider as well that some of these towns would be built around a particular thing--a mill for instance--a particular town might be a place where everyone from the surrounding communities come to get their wheat ground, or their leather cured, or something specific to that township. Not every town is going to have the same amenities, and that's important to remember when building them. Not every village is going to have an inn (though it might have people willing to rent a bed).

A town of 3000 would have these 10 or so communities orbiting it--sort of the Medieval version of the suburbs, with a slice of country in between them. So when you were coming up on a real township, you might see more of these small communities, near the larger one--they'd be anywhere from 20 minutes away to 3 hours walk away (12 miles), to a few on the edges that might be as far away as a day's walk. An inn in a town such as this is going to be busy, because travellers just along the road, and traders--people also paid to bed up in barns and sheds nearby--or for a place to park their wagon for the eve.

So your model actually needs to be built with the larger cities placed on the map first.

Then, around each of them, place their satellite villages and settlements. Your max place of 6000--that will need maybe 20 little villages scattered around it. (Some of these won't have more than 50 people in them).

Density and distance between can get weird and really really close in some places, especially when the "orbit" of two cities cross.

In places where there aren't communities/cities of 1000 people to influence the placement of the smaller communities, the density gets lighter. So you might not see any villages for an entire day's walk or ride--this is why some enterprising fellows would set up an inn at a crossroads in the middle of nowhere. They knew people would be travelling, tired, and that there was no chance of anywhere (along the main road anyway) where one could rest, and feel somewhat secure (highwaymen are less likely to visit where there are people).

These less dense places sort of make up for the areas where, every few miles, there's a village...

Even your 400 person village is going to have smaller communities supporting it in this satellite configuration. These will be made up of substantially less people.

As one of the other posters points out, 400 is pretty large.

But let's take a mega-city of the time--London in the 1000s--population estimates are about 10,000-12,000. That means there are going to be about 40 communities surrounding it and supporting it, likely all within a 10 minute walk to about 3 days walk, because it's so large.

By 1337, the London population had grown considerably. Here's an estimate of the population of largest cities from wikipedia, which I am adding because you've said your numbers are arbitrary, and it's good to have a basis:

  1. London 23,314
  2. York 7,248
  3. Bristol 6,345
  4. Coventry 4,817
  5. Norwich 3,952
  6. Lincoln 3,569
  7. Salisbury 3,226
  8. King's Lynn 3,217
  9. Colchester 2,955
  10. Boston 2,871
  11. Beverley 2,663
  12. Newcastle 2,647
  13. Canterbury 2,574
  14. Bury St Edmunds 2,445
  15. Oxford 2,357
  16. Gloucester 2,239
  17. Leicester 2,101
  18. Shrewsbury 2,083
  19. Great Yarmouth 1,941
  20. Hereford 1,903
  21. Cambridge 1,902
  22. Ely 1,772
  23. Plymouth 1,700
  24. Exeter 1,560
  25. Hull 1,557
  26. Worcester 1,557
  27. Ipswich 1,507
  28. Northampton 1,477
  29. Nottingham 1447
  30. Winchester 1,440

Keep in mind that there are lots of smaller communities surrounding each of these. A city of 6,000 would not be at all common by the standards of Medieval times. Just throwing out these numbers and specific cities to give you a base. Add to this that the low estimate of the number of Medieval villages that we HAVEN'T discovered yet is 3,000--and you'll see that the landscape was likely teeming with tiny villages along the way, with some less populated areas in between.

As other posters have pointed out, the more farmable the land, the closer settlements are to one another. In the densest places it's going to be about one every mile or two (and there will be some very close to large cities). These communities are going to be tiny, but EVERYWHERE, with farmland in between.

Please see the fantastic video in Giant Cow's answer--this distance is a pretty good yardstick for world-building (about 10 miles between each sizable town, which means that there's a town is within 5 miles for everyone) for places that are not as richly farmable, and less influenced by the larger cities. Medieval towns were different because of the factors I have outlined, often closer, much smaller, and with more distance in between settlements where farming was impossible or there was woodland.

  • 2
    @ChrisBKreme I haven't added that some of these villages are going to be monasteries/tiny towns supporting monasteries. Monks often produced specific things (such as beer, preserves and other products) and it could work like a factory with a religious bent. The bigger places would sometimes spout real towns. – Erin Thursby Apr 9 '17 at 21:03
  • 7
    When you write "there were over 3,000 deserted Medieval villages that they hadn't discovered", what are you referring to? Who exactly hadn't discovered which villages? When? – Dronz Apr 10 '17 at 3:35
  • 2
    @Dronz that's the estimate of the number of deserted Medieval villages that we haven't unearthed--that's based on references from larger towns and the distances between the ones that have been recorded in England. This number comes from research and reading that I've done concerning it, so I don't have an internet link to it. – Erin Thursby Apr 10 '17 at 4:13
  • 2
    @Dronz That's an estimate, but it's a low one. Considering that a Medieval village could consist of just 10 houses, or less, that's not beyond the pale. lots of these places were totally abandoned during the plague, and most were converted to farmland. Farms were smaller then. – Erin Thursby Apr 10 '17 at 4:15
  • 6
    +1 for the last paragraph. I come from a region which is praised for the good quality of soil it has. Around here, villages and even bigger towns are closer than in the rest of the country and almost all villages and all the towns can trace their origin back to the medieval ages. The big towns here are 15-25 miles apart, while 30-60 miles is more common in the rest of the country. Around towns and along the roads connecting them there is a dense net of villages 1-2 miles apart from each other, while 2-5 miles is more common elsewhere. And in woodland areas the distances are even larger. – Sumyrda Apr 10 '17 at 5:33

This video explains why cities are were they are. You can apply this logic to the distance between towns

Its worth pointing out that there is no standard distance between towns necessarily. Generally speaking, chains of towns can be seen along rivers at surprisingly even spacing (see the video), but these are all relative to one another. It is possible to have a city and surrounding towns be highly isolated. The trick is to make a map and work backwards from it. It will be obvious where to place cities.

As the video explains, the expected distance for towns on a route is usually the distance you can travel there and back in a day.

  • 4
    This is an excellent resource for world building! The spheres of influence are a great way to build a world! You'd have to rate common services vs. less common and what larger cities would have...+1 – Erin Thursby Apr 10 '17 at 3:18
  • 2
    This video is what I came here to answer with. It's an excellent explanation of the logical process you can work through. – Werrf Apr 10 '17 at 19:35

Half a day's walk, little less. Perhaps a day's walk for the more remote villages. Given the average of 5km/h (3.1 mph) I'd say 25 kilometers, maybe 30 kilometers.

Having half a day's walk allows you to go to the market of the city. Buy some things, sell some things and then still get home. So radiate villages outwards from the cities. Not neatly of course, the land dictates their exact location. This also gives you a good distance for inns. Set them about half a day from the village.

Rivers might change this a bit, as going downstream might speed up your journey quite a bit. Depending on the location, if your people are blessed the average wind will blow against the current. That way you get either the current or the wind on most days. If I recall the Nile was such a river.

I'd like to make a side note. Your estimate of a rural village being 400 seems a bit high to me.

  • 3
    Yes, the 400 person estimate seemed high to me as well. A village has a range from about 100 people to about 1,000. Anything above 1,000 is a town. There are actually going to be lots of micro-communities smaller than 400 everywhere... – Erin Thursby Apr 9 '17 at 16:25
  • 13
    You're proposing that people would walk eight hours to the market, trade there, then walk eight hours home again in the same day? That seems very unlikely, not least because it requires 17+ hours of daylight, which is only available for a small fraction of the year. – David Richerby Apr 9 '17 at 19:38
  • 9
    Even four hours walking distance (20 km or thereabouts) would seem far, because in order to go to trade, you would then need to walk for four hours (carrying your goods, if any), do whatever trading you came to do (and you wouldn't exactly go there to buy just a few eggs and some salt!), and then walk back home again for four hours (again carrying goods). Someone would come up with the bright idea of establishing a marketplace closer. And of course as David Richerby pointed out, we are rather spoiled with artificial lighting these days. Darkness harbored bad things. – a CVn Apr 9 '17 at 20:13
  • 8
    @MichaelKjörling Yes, four hours is too far. One or two hours is far more realistic for this kind of trade. Villages were far closer to each other in some areas during Medieval times than most people realize. Lots of them were only a few miles from each other. – Erin Thursby Apr 9 '17 at 22:00
  • 8
    @DavidRicherby : Actually people would really do this. My greatgrandfather was a farmer in Schüpfheim (Canton Lucerne, Switzerland) and would travel with his farm servants every Wednesday (traditionally market day in Switzerland) to Lucerne - on foot! The distance of roughly 37km/23mi took almost precisely the 8h you mentioned. They'd start around 4am, arrive around lunch, do business till 4pm and then return. You don't have daylight all the way but you're familiar with the route and the times it's really dark is the regions you're familiar with. – Patric Hartmann Apr 10 '17 at 18:24

There are reasonably accurate maps from the 1570s This one shows the county of Hampshire and each settlement. Most are clearly identifiable with modern villages towns and cities. On this map, it is apparent that few villages are more than about a mile or two from their nearest neighbour. In regions of better farmland, the villages are closer. They are more separated in the New Forest region, which is less suitable for agriculture.

You can research further, as many of these locations can be found in the Domesday book. The population of England had changed between 1080 and 1570, but not by a very great amount (it had perhaps doubled), partly due to the effects of plague. Doubtless, some of these settlements are little more than farms, others are more substantial. However, each is a named location with an identity and a social structure.

In New England, especially rural Vermont, many towns are roughly 7 miles across with a town center centered. Driving Route 22a that is particularly obvious. Many of those towns were chartered in the 1600's, well before cars were in use, though they are certainly not medieval.

I was told that the distance facilitated courting on Sunday afternoons. After church a young man would have the time and energy to walk to the next town, spend some time with a woman and her family and walk home.

Also having the nearest town center no more than 3.5 to four miles away, i.e only about an hours' walk, makes going into town, doing something, and returning the same day a reasonable choice for regular events. I've tested this extensively.

Outside of towns centers, houses tended to group into blocks about a mile apart. Off the main roads, this is still a common pattern. Open fields and pasture for about a mile, then a group of 4-12 houses and barns within the span of about 1/16-1/4 mile, then repeat. Of these groups, 4-5 per town seem to include the remains of an old school-house (in addition to the currently operating schoolhouse in the town center and a union high-school in a larger, central town or city). That could be enough to consider at least the larger groupings villages.

Rural towns in that area seemed to peak in population in the mid-1800's, at a time when sheep farming was big. Then, the population could be close to 2000 people in towns where it it now closer to 1000. For example:,_Vermont.

In towns of that size in that area, town government is part-time, often volunteer, and town meetings predominate.

Basically rural New England is one of many possible examples of places that were substantially populated before motorized vehicles, have not grown significantly since, and have little town-vs-town warfare.

There was actually a lot of interesting research done about this, for example the Theory of Central Places by Christaller (see For typical towns, he came up with 10-15 miles; but he establish different 'size' categories, where each size has its own hexagonal grid.

The maximum distance between any two adjacent towns should be such that an average person can travel from one to the other during daylight hours. Traveling in the dark is dangerous, so people generally won't make the journey unless they can be reasonably certain to arrive before sundown.

J Bergen is partially right about U.S. towns for this reason. While Colonial America was much less dangerous than medieval Europe, the paradigm remained that towns should be situated about a day's journey apart. Even up to the 19th century, this was generally considered to be 12 to 15 miles. If you look on a map of the eastern U.S. (Pennsylvania for example), and follow one of the highways (U.S. route 11 for example) you'll find that the towns are generally 12 to 15 miles apart.

  • Inns were pretty common along major transit routes. Spaced about that far apart. – Sobrique Apr 11 '17 at 13:59

About 20 miles sounds right. For example, I live between the ancient towns of Dover and Canterbury, which are indeed about 20 miles apart.

  • Are you sure there were no settlements between Dover and Canterbury? You might want to check on the history of a few of the villages between the two. – Jack Aidley Apr 11 '17 at 15:02
  • @JackAidley Nothing you would call a town. Bridge and Whitfield were and are suburbs of Canterbury and Dover respectively. Elham would be the largest standalone settlement, and it did historically have a market, but I think it's always been on the village side of the village/town divide. I tend to think of a village (in England) as turning into a town when it's big enough to have more than one Church of England church. – Mike Scott Apr 11 '17 at 15:39
  • 1
    @MikeScott: from an English POV, I think the questioner's definition of "town" does indeed include things smaller than a town. The actual definition in England was political (a town has a charter, a village doesn't), hence we inherit these rules under which the oft-claimed largest "village" in England (Kidlington) is more than ten times the size of the oft-claimed smallest "town" (Manningtree)! – Steve Jessop Apr 11 '17 at 17:58

15 miles U.S. between towns. This gave farmers time 1 day a week to do the chores. Walk to town. Shop & walk home. Feed the animals. Go to bed or a 7.5 mile walk to town. That would be a 4 hour walk. Less with a ox or horse cart.

  • Probably more with an ox cart. This is why the horse yoke was an important invention. – mattdm Apr 10 '17 at 7:30
  • Do you have citation for any of this? You seem to be suggesting the towns were spaced like this to suit farmers' weekly shopping schedules (which are also citation needed, and seems to not consider the idea that maybe farmers just made sure they were within practical distance of a nearby town), and that farmers only fed their animals once a week (or is that unrelated?). – doppelgreener Apr 10 '17 at 8:26

I would start with the map and characters' needs and go from there.

  1. Geography. You wouldn't build a town with no access to water or bountiful natural resources (such as forests and suitable land for feeding livestock) would you? Seaside towns, towns by rivers, forests, and forts built on strategic high ground would be good spots for towns. After you mark these spots on a map then you can draw lines between them and note their distances apart.
  2. Characters. Are your characters going to be traveling between towns? Do you need the story to go at a faster pace and they only need to travel half a day or two to the next town? Or perhaps it's more of a Lord of the Rings type of travel time - where it's more of the adventure along the way that counts and not so much the destination.

This was kind of standard till the 1900's in town spacing. The 15 miles. Do to time to get to town & back home for farmers. Even the railroad when built in America used this formula in town spacing. As were to put towns. For max profit. Bigger towns went were rivers joined or were roads crossed. It was kind of a world wide spacing till good roads & fast transportation was here.

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.