the reason i spent all afternoon fixing my linux partition is that @youarenotthewalrus linked me to a really neat worldbuilding blog last night; they run through a bunch of different climate variations using ExoPlaSim playing with stuff like obliquity and rotational period, and it made me realize my climates for Sogant Raha are all wrong--indeed, the lack of axial tilt and seasonal variation has the exact opposite consequence of what i thought it did: very harsh temperature differences, almost no temperate zones.

i wanted to see if i could get ExoPlaSim running on my machine, but alas after spending another couple of hours fiddling with it, I don't think I can. I think Sogant Raha's climate is salvageable if i just reverse the axial tilt, and make it greater than Earth's, but I really wanted to try to play around with different parameters more to get a better idea of what its climate zones should be like. very frustrating.

I'm on a discord with the guy who writes that blog, and he takes commissions to run climate explorations, so its possible you you could find out that way.

I've been thinking about mules, and how fucked up the Discourse™ would be if humans and some other sapient hominid were in the same situation as horses and donkeys. The chromosomal mismatch (horses have 64, donkeys 62, mules 63) renders their offspring infertile and bring out some "primitive" traits not commonly expressed by their parents-- imagine what might happen in the development of a sapient creature.

Can you imagine how fucked up the fight for equality in marriage, sex rights, etc. would be? Would it be justifiable to have a child with your partner, knowing said child will suffer the consequences for your actions? Full scale eugenics is probably off the table for ethical reasons, but would it not also be unethical to bring into the word a child guaranteed to be infertile, developmentally delayed, or unable to live longer than their parents? The fight for LGBTQ+ rights has been ridiculously difficult, and there's literally no downsides. The debate over the rights for Human-46s and Human-48s to get it on would be terrifying.

It only matters in cases where pregnancy is a concern, but we all know how fucked everything about normal birth control is. How bad would your offspring need to have it for mandatory sterilization to be on the table? The dating scene for straight people would be so fucking dystopian. Sure, you like this guy, but do you like him enough to get a hysterectomy? Are you sure he's the one? That you won't fall apart in three years? That you won't find yourself with someone of the same species, craving children of your own, but unable because of a decision you made as drunk college kid? It matters less if you're gay/straight but one of you is trans/otherwise unable to produce a child via raw-dogging, but I trying to envisage the discourse around "The government forced me to tie my tubes even though my partner tops" makes me feel ill.

Growing up as a Human-47 would also have to suck -- knowing your parents cared more about a five second orgasm than the fifty years you get to spend knowing you're inferior to both of them. "Was it really worth it, Mom?" you ask. "Did he make you cum well enough to offset my fifth-grade reading level?"

The flip side is also bad: what if Human-47s were better than their parents, but still infertile. Are you so committed to the survival of your species that you're willing to forgo that hybrid vigor? Is it more important to you that your child produce sperm/eggs than be capable of true, total self-actualization? Do you you love your spouse enough to (relatively) stunt your children? On the flip side, is your child's success more important than letting them raise a family? Growing up Human-46/48, especially if you don't even want kinds, can you forgive your parents for their choice? Can a wanting Human-47 mother or father forgive their parents?

Not trying to draw explicit parallels with real-world issues here, and I think most of these questions aren't too hard to find good solutions to-- but I've been turning this idea over in my head like a rotisserie chicken for months and I think it might kill me if I don't get it out.

I don't think hybrid sterility by itself would be a big problem. Lots of people don't have kids. Hybrids would just be in about the same position as lots of IRL infertile humans, or gay couples; if they want to raise a family, they can adopt. I guess it'd be more of an issue in societies where lineage is a big deal though; I suspect in societies like that the two species would mostly just avoid intermarriage (a lot of them might not consider interspecies marriages legitimate). And I could see some philosophical and religious traditions being opposed to marriages involving hybrids for basically sex-negativity reasons, because they'd be sexual relationships without any possible fig leaf of it being about procreation.

I could actually see there being social roles for hybrids of this type where they're valued precisely for their sterility, similar to eunuchs in some societies IRL. I think they'd often be employed as bodyguards and servants for aristocratic women and guards for royal harems, for instance (a common niche for eunuchs IRL IIRC). Similarly, I think before modern birth control male hybrids of this type might have a potentially quite lucrative niche as sex workers who cater to straight and bi women; imagine what the existence of a class of men who can be conventionally attractive by straight woman standards and can have PIV sex with you but can't get you pregnant might mean to women basically any time before birth control pills were a thing. This might be a bit trickier if the hybrids can sometimes make babies, it's just really, really rare (I vaguely remember something about how mules are actually like this, but I'm too lazy to look it up now). Incidentally, if there was a sapient and human-ish looking species that just couldn't reproduce with us, I think they'd similarly often end up in "valued precisely for the fact they can't make babies with us" niches.

Incidentally, IIRC, sometimes hybrid sterility is sex-asymmetric, with male hybrids being sterile but female hybrids being fertile. I think this might be the case with lions and tigers? So there's another potential scenario.

As for if the hybrids had viability issues/health problems or had traits that made them poorly compatible with the society or one of both of their parent species or had traits we'd be inclined to see as tragic like a short lifespan... it'd really depend on the exact traits the hybrids have. To pick just two of many possibilities, "hybrids tend to be kind of frail and sickly but their childhood and young adulthood survival rates aren't much worse than parent species individuals and if they get modern medical care or are just lucky enough to not get taken out by a childhood disease they'll usually live to old age" and "hybrids usually die before they're five years old and they're usually very sick in the few years they have" are very different scenarios that will cause very different responses.

the kinetic energy of a 610 kg spacecraft traveling at ~6600 m/s is (by my math) ~1300 MJ, and the gravitational binding energy of a 5x10^9 kg asteroid that’s ~170 meters in diameter is (again, by my math) ~12 MJ, so why wouldn’t DART just blow Dimorphos apart, or at least knock a big chunk out of it? Most of the coverage I’ve seen indicates that they expect Dimorphos to be deformed by the impact, but not, like, totally destroyed.

I tried seeing what Google turned up and according to this Van der Waals forces may play an important role in holding at least some rubble pile asteroids together, though that particular asteroid is apparently an exception to a generally observed trend, so it may be atypical. Van der Waals forces are apparently also what gives gecko feet their grip. It also mentions that "until now, it was assumed that" rubble pile asteroids are held together by gravity and "friction."

That book I mentioned in an earlier post, Money, Markets and Trade in Late Medieval Europe, has a section that might be really useful to speculative and historical fiction authors! I’ll just post a copy of it:

——-

Each person who lives in the central place requires the labour of people producing food within the hexagon around the consumption center, that is around the central place. The surplus created by the farmer will determine the land area required to support each resident of the central place. That in turn means that the amount of land it took to feed a town depended on how much land it took to feed a person. For the late Middle Ages, using figured generated by Christopher Dyer from the end of the thirteenth century, with gross yields of around 1,000 litres per hectare for many food grains and after deductions for seed, feed for animals, food for the farmer’s family, and for tithes, the net figure for marketable grain was between 160 and 220 litres per hectare. Another estimate for the early fourteenth century by James Galloway and Margaret Murphy puts the average marketable surplus at 360 litres per hectare for wheat. There is a wide variation of course but something like 300 litres net per hectare for wheat and oats and 400 litres net for barley seem reasonable for late medieval England. If per caput annual consumption of food grains was about 600 litres, an estimate which is not unreasonable and reflects neither a poor nor a prosperous standard of living, then it took around two hectares of farmland, more or less, to support a town dweller. That assumes that the entire rural population was involved solely in production of food for market. That was of course not true and became less true in the late fourteenth and fifteenth century with expansion of rural industry. If for every farmer in the countryside there was one dependent person who consumed rather than produced food then the land area needed to feed one urban dweller was around four hectares, and that estimate probably is a high one.

As to fuel, though in some parts of England coal was known and used in the late Middle Ages, for the overwhelming majority of the population heat came from firewood. Around 1300 coal cost four times as much in London as it did in Durham so even though coal had higher calorific value - more than double that of wood - other than for industrial uses, firewood was the logical source of home heat. Von Thunen expected that forested areas near cities would be retained and managed to produce heating supplies and that certainly did happen in southeast England, in regions near London. Attempts to estimate the land area needed to supply people in towns with fuel are even more plagued by problems than the attempts to estimate the land area needed to supply food grains. Firewood consumption varied with the climate. Firewood production varied with the character of the land and the level of organization and management in exploiting the land. One authority offers a figure for Sweden and Finland of eight kilograms of firewood per person per day as the consumption norm, including industrial uses, but in northern France, Germany, the Netherlands, and England the estimate is about four kilograms per day. Four kilograms per day translates into about 1.5 tonnes annually and assuming a level of the average productivity of forests of between 0.5 and 1.5 tonnes per annum it would have taken from 0.5 to 1.0 hectares of managed woodland to produce that much firewood and so meet the needs of the average inhabitant of a town. Presumably the demand for firewood per caput was on the rise in late medieval England because of the spread of the use of the chimney. Production figures reported by Galloway, Keene, and Murphy from the woodlands supplying late medieval London suggest output of 2.25 tonnes per hectare, an estimate that is generous. They also estimated per caput wood consumption for the fourteenth century at 0.8 tonnes. At that pace and at the high rate of production they use it took an area of at least 0.35 hectares to supply each Londoner, but that figure errs to the low side. At that level it took about 28,000 hectares in 1300 to meet the needs of the population of 80,000 and only 18,000 hectares in 1400 when the population was considerably reduced. The estimate of 0.35 hectares per Londoner seems low in general but especially because a contemporary in 1700 per the land are required to supply each town dweller with firewood at 0.5 hectares, and that in an era when coal already was making a considerable contribution to the thermal energy needs of the capital. So despite the extensive and directed work of Galloway, Keene, and Murphy an estimate of 0.5 hectares as a per caput land requirement for firewood supplies is preferred. People in the countryside would have required firewood as well. The ratios employed for net grain production can be and should be applied to estimating rates of firewood output by country dwellers for themselves and city dwellers. Assuming 16 rural people for each urbanite, that is two farm families of four each producing grain for the town dweller and two farm families producing grain for dependents in the countryside, then to supply them would have taken eight hectares of woodland at 0.5 hectares as the land requirement just to supply the 16. it would have taken an additional 0.5 hectares then to meet the needs of the single town dweller. An estimate of something on the order of eight hectares needed to supply enough wood for people in the countryside as well as a single person in town is probably not wide off the mark.

Many factors make the figures suspect and the task of lending precision to the estimates for land requirements to sustain townspeople is far from complete. […] No matter the efforts at estimation it appears that England’s largest urban centre in the late Middle Ages, London, had its fuel needs easily met by producers in the region, and there was still a surplus for export. That along with other indications of regular and consistent access to adequate heating fuel in the Low Countries and England suggests that most of the time most towns did not have trouble with their energy requirements.

If it took four hectares of land in total to produce the food needed for a town dweller and eight hectares in total to produce the firewood, for a notional town of 1,000 people - assuming that none of the people in the town did anything to produce any food or fuel for themselves - the total rural area required was 12,000 hectares. If a distribution of production of food grains and fire-wood for the town of 1,000 conformed to a pattern of regular hexagons - a limiting assumption though as indicated theoretically a reasonable first approximation - then the distance from one side of the hexagon to the other would have been about 12 kilometres. The shortest distance to any of the six sides of the notional hexagon from the centre would have been under six kilometers. It was a distance a farmer could walk in an hour with little difficulty. For a town of 10,000 the distance along any of the six sides of the regular hexagon would been about 21.5 kilometres. For a city the size of Ghent or Bruges, that is around 40,000, the length of the walk to the outer extremity of the idealized supply zone would have been just short of 43 kilometres, an easy trip in a day on foot. For London at its maximum late medieval population around 13000 of 80,000, and the largest town in the region outside of Paris, the theoretical distance was just short of 61 kilometres. Because of the geometry that distance was not eighty times the distance for a town of 1,000 but only something more than five times the distance.

A careful examination of sources of food for London, the second largest and possibly at some times the largest city in the region, indicates that virtually all the food needed, and so excluding fuel, came from 100,000 hectares, an area in total about twice the size of the small county of Middlesex. It may be that because of the nature of surviving documents research has missed some sources Londoners used for food, for example nearby on the Continent, but whether that is true or not the fact remains that food for the largest city in England came almost entirely from the southeast of England. In around 1500, farmers in the coastal Low Countries could supply something on the order of twice as much grain as was needed to feed the existing urban population. Even if the estimates to arrive at that conclusion are somewhat suspect and even if the final result is off by as much as 100% still the weight of the evidence strongly supports the impression that for most of the fourteenth and fifteenth centuries most grain to feed towns came from close by and did not have to brought from far afield, that is except in unusual years. Such calculations depend on all the land within the notional polygon being productive and able to supply either food or fuel. But even assuming that only half the land was productive for a town of 10,000 that would have increased the length of the side of the hexagon only to a bit more than 30 kilometres from the centre to the outer edge of the hexagon, again something that could have easily been covered in half a day by a farmer on foot and with some time to spare.

——

Notes by me: A hectare is 10,000 square meters (it’s a square 100 meters on a side); a square kilometer is 100 hectares.

One note of caution I have about these figures is in theory they should make it very easy and convenient to calculate a maximum sustainable population density, but when I try that the result strikes me as wildly optimistic and inconsistent with real history.

These figures suggest you need something like 1.1 to 1.6 hectares of land per person; .6 hectares for food production and .5-1 hectares for firewood production. Let’s be conservative and say 2 hectares.

A hectare is 10,000 square meters, a square kilometer is 100 hectares. So two hectares per person gives a maximum population density of 50 people per square kilometer.

That is actually quite dense. At 50 people per square kilometer, Europe (with a land area of 10 million square kilometers) would have 500 million people, not much less than its modern population.

Let’s sanity-check this with real history. As I recall from reading stuff about this, the European population seems to have grown to the point of straining the carrying capacity of the land in the fourteenth century (note: the century of the Great Famine), and then again in the sixteenth century. The fourteenth century resource crisis was ended by the depopulation caused by the Black Death, the sixteenth century resource crisis was solved by agricultural and technological improvements like the agricultural revolution, the masonry heater, the widespread adoption of coal as a fuel, and ultimately the industrial revolution. The fourteenth century population high point was 70-80 million people. In the sixteenth and seventeenth centuries population exceeded 100 million people and kept going up.

So 2 hectares per person suggests a maximum sustainable average population density of 50 people per square kilometer and a maximum sustainable population of 500 million for Europe with Medieval technology, but in actual history the resources of fourteenth century Europe were getting strained supporting less than one fifth that many people. We have a problem!

Possible sources of the discrepancy:

- Some land is not suitable for agriculture. But I’m pretty sure Europe is more than 10% arable land.

- I think the grain yield figures here are for relatively good land, like England? The amount of land needed to grow food for one person is going to be higher in places like Scandanavia, and probably most of Russia (which is a pretty big chunk of the European land area). But would this explain a more than five-fold difference in the average?

- I got .6 hectares per person for food from the estimates above for 1000 liters of grain per hectare and 600 liters of grain to sustain a person for a year. But this ignores the raising of livestock, fruit, vegetables, etc., which would probably have mostly been less land-efficient than grain farming but would have provided vital nutrients (and also made the diet a less monotonous). Still, IIRC pre-modern peasants did tend to live mostly on grain, often to the point of suffering nutrient deficiencies (see: pellagra), so I doubt this would make a five-fold difference either (the rich would have consumed larger quantities of meat, fruit, etc., but they were only a small percent of the population, so land use probably at least loosely reflected the peasant diet).

Possibly the difference is a combination of all these factors? Or something I’ve overlooked?

Alternately, maybe the fourteenth century population crisis was a local crisis in the most densely populated regions like Britain and France, and most of the land area of Europe was still well below carrying capacity at the time? One data point that’s suggestive of that: China in 1900 had 415 million people, fed using what at the time would probably have still been basically pre-industrial farming methods. So that suggests 500 million people for a Europe-sized land area with pre-industrial farming isn’t crazy, though notably China relied heavily on rice agriculture.

Another noteworthy figure for area of agricultural land needed to support a single person with Medieval farming I found here:

“In regards to area of the village itself, plowed fields were measured in “hides”, roughly 120 acres each; Elton with its 500-600 people was assessed at 10 hides (including a church and two mills, for a tax of 14 or 16 pounds). 9 This area of some 1,200 acres is about 2 square miles; on average we see one “hide” per 50 people or so.”

An acre is .4047 hectares (4,047 square meters) - or, to reverse that, a hectare is about 2.47 acres. So a hide would be a little less than 50 hectares - or a little less than half a square kilometer. That gives about 100 people for every square kilometer of plowed fields - or about 1 hectare of plowed fields for every person (so, envision a square of plowed fields 100 X 100 meters on a side - that’s about what you’d need for one adult person). That’s roughly in the same ballpark as the .6 hectares to grow the grain one adult person needs in a year figure.

Notably that PDF also gives a figure for size and density of villages: the average village might have around 350 people, and there would be around 1 village per 10 square miles. If one person needs 1 hectare of plowed fields, a village might be expected to have around 350 hectares of plowed fields - 3.5 square kilometers of plowed fields. A mile is about 1,609 meters, so a square mile would be 1609 meters X 1609 meters = 2.59 square kilometers. 10 square miles would be 25.9 square kilometers. This suggests most of the land was not plowed land (at least, at a population density of 13.5 people per square kilometer - which is a somewhat less than twice what the average population density of Europe would have been at the fourteenth century high point of population). People who know more about the Middle Ages or agriculture than me can maybe answer whether that passes the smell test.

It’s also interesting to compare these figures to human walking speed, which averages a bit less than 3 miles per hour. 3.5 square kilometers is a circle with a radius of a little less than 1100 meters - about a 14 minute walk. 10 square miles is a circle with a radius of about 1.8 miles, or maybe around a 40 minute walk. Huh, I’m surprised at what short walks these turn out to be - it’s that radius^2 term that does it, exponents are wild! This suggests that at population density of Medieval England (so probably on the dense side for populations with non-industrial agriculture), farmers in a central village would have experienced no great hardship walking to their fields, even if their fields were quite extensive and accounted for most of the land area. This makes sense: physical accessibility of the fields on foot would have been a major limiting factor for preindustrial peasants, so it would have shaped patterns of population dispersal. If the fields necessary to support a village-size population were too big to conveniently walk to them all on foot from a central location, then probably people would not live in villages - they would instead live scattered over the landscape in isolated homesteads.

10 Lessons on Realistic Worldbuilding and Mapmaking I Learned Working With a Professional Cartographer and Geodesist

Hi, fellow writers and worldbuilders,

It’s been over a year since my post on realistic swordfighting, and I figured it’s time for another one. I’m guessing the topic is a little less “sexy”, but I’d find this useful as a writer, so here goes: 10 things I learned about realistic worldbuilding and mapmaking while writing my novel.

I’ve always been a sucker for pretty maps, so when I started on my novel, I hired an artist quite early to create a map for me. It was beautiful, but a few things always bothered me, even though I couldn’t put a finger on it. A year later, I met an old friend of mine, who currently does his Ph.D. in cartography and geodesy, the science of measuring the earth. When the conversation shifted to the novel, I showed him the map and asked for his opinion, and he (respectfully) pointed out that it has an awful lot of issues from a realism perspective.

First off, I’m aware that fiction is fiction, and it’s not always about realism; there are plenty of beautiful maps out there (and my old one was one of them) that are a bit fantastical and unrealistic, and that’s all right. Still, considering the lengths I went to ensure realism for other aspects of my worldbuilding, it felt weird to me to simply ignore these discrepancies. With a heavy heart, I scrapped the old map and started over, this time working in tandem with a professional artist, my cartographer friend, and a linguist. Six months later, I’m not only very happy with the new map, but I also learned a lot of things about geography and coherent worldbuilding, which made my universe a lot more realistic.

1)  Realism Has an Effect: While there’s absolutely nothing wrong with creating an unrealistic world, realism does affect the plausibility of a world. Even if the vast majority of us probably know little about geography, our brains subconsciously notice discrepancies; we simply get this sense that something isn’t quite right, even if we don’t notice or can’t put our finger on it. In other words, if, for some miraculous reason, an evergreen forest borders on a desert in your novel, it will probably help immersion if you at least explain why this is, no matter how simple.

2)  Climate Zones: According to my friend, a cardinal sin in fantasy maps are nonsensical climate zones. A single continent contains hot deserts, forests, and glaciers, and you can get through it all in a single day. This is particularly noticeable in video games, where this is often done to offer visual variety (Enderal, the game I wrote, is very guilty of this). If you aim for realism, run your worldbuilding by someone with a basic grasp of geography and geology, or at least try to match it to real-life examples.

3)  Avoid Island Continent Worlds: Another issue that is quite common in fictional worlds is what I would call the “island continents”: a world that is made up of island-like continents surrounded by vast bodies of water. As lovely and romantic as the idea of those distant and secluded worlds may be, it’s deeply unrealistic. Unless your world was shaped by geological forces that differ substantially from Earth’s, it was probably at one point a single landmass that split up into fragmented landmasses separated by waters. Take a look at a proper map of our world: the vast majority of continents could theoretically be reached by foot and relatively manageable sea passages. If it weren’t so, countries such as Australia could have never been colonized – you can’t cross an entire ocean on a raft.

4)  Logical City Placement: My novel is set in a Polynesian-inspired tropical archipelago; in the early drafts of the book and on my first map, Uunili, the nation’s capital, stretched along the entire western coast of the main island. This is absurd. Not only because this city would have been laughably big, but also because building a settlement along an unprotected coastline is the dumbest thing you could do considering it directly exposes it to storms, floods, and, in my case, monsoons. Unless there’s a logical reason to do otherwise, always place your coastal settlements in bays or fjords.

 Naturally, this extends to city placement in general. If you want realism and coherence, don’t place a city in the middle of a godforsaken wasteland or a swamp just because it’s cool. There needs to be a reason. For example, the wasteland city could have started out as a mining town around a vast mineral deposit, and the swamp town might have a trading post along a vital trade route connecting two nations.

 5)  Realistic Settlement Sizes: As I’ve mentioned before, my capital Uunili originally extended across the entire western coast. Considering Uunili is roughly two thirds the size of Hawaii  the old visuals would have made it twice the size of Mexico City. An easy way to avoid this is to draw the map using a scale and stick to it religiously. For my map, we decided to represent cities and townships with symbols alone.

 6)  Realistic Megacities: Uunili has a population of about 450,000 people. For a city in a Middle Ages-inspired era, this is humongous. While this isn’t an issue, per se (at its height, ancient Alexandria had a population of about 300,000), a city of that size creates its own set of challenges: you’ll need a complex sewage system (to minimize disease spreading like wildfire) and strong agriculture in the surrounding areas to keep the population fed. Also, only a small part of such a megacity would be enclosed within fantasy’s ever-so-present colossal city walls; the majority of citizens would probably concentrate in an enormous urban sprawl in the surrounding areas. To give you a pointer, with a population of about 50,000, Cologne was Germany’s biggest metropolis for most of the Middle Ages. I’ll say it again: it’s fine to disregard realism for coolness in this case, but at least taking these things into consideration will not only give your world more texture but might even provide you with some interesting plot points.

 7)  World Origin: This point can be summed up in a single question: why is your world the way it is? If your novel is set in an archipelago like mine is, are the islands of volcanic origin? Did they use to be a single landmass that got flooded with the years? Do the inhabitants of your country know about this? Were there any natural disasters to speak of? Yes, not all of this may be relevant to the story, and the story should take priority over lore, but just like with my previous point, it will make your world more immersive.

 8)  Maps: Think Purpose! Every map in history had a purpose. Before you start on your map, think about what yours might have been. Was it a map people actually used for navigation? If so, clarity should be paramount. This means little to no distracting ornamentation, a legible font, and a strict focus on relevant information. For example, a map used chiefly for military purposes would naturally highlight different information than a trade map. For my novel, we ultimately decided on a “show-off map” drawn for the Blue Island Coalition, a powerful political entity in the archipelago (depending on your world’s technology level, maps were actually scarce and valuable). Also, think about which technique your in-universe cartographer used to draw your in-universe map. Has copperplate engraving already been invented in your fictional universe? If not, your map shouldn’t use that aesthetic.

9)  Maps: Less Is More. If a spot or an area on a map contains no relevant information, it can (and should) stay blank so that the reader’s attention naturally shifts to the critical information. Think of it this way: if your nav system tells you to follow a highway for 500 miles, that’s the information you’ll get, and not “in 100 meters, you’ll drive past a little petrol station on the left, and, oh, did I tell you about that accident that took place here ten years ago?” Traditional maps follow the same principle: if there’s a road leading a two day’s march through a desolate desert, a black line over a blank white ground is entirely sufficient to convey that information.

10) Settlement and Landmark Names: This point will be a bit of a tangent, but it’s still relevant. I worked with a linguist to create a fully functional language for my novel, and one of the things he criticized about my early drafts were the names of my cities. It’s embarrassing when I think about it now, but I really didn’t pay that much attention to how I named my cities; I wanted it to sound good, and that was it. Again: if realism is your goal, that’s a big mistake. Like Point 5, we went back to the drawing board and dove into the archipelago’s history and established naming conventions. In my novel, for example, the islands were inhabited by indigenes called the Makehu before the colonization four hundred years before the events of the story; as it’s usually the case, all settlements and islands had purely descriptive names back then. For example, the main island was called Uni e Li, which translates as “Mighty Hill,” a reference to the vast mountain ranges in the south and north; townships followed the same example (e.g., Tamakaha meaning “Coarse Sands”). When the colonizers arrived, they adopted the Makehu names and adapted them into their own language, changing the accented, long vowels to double vowels: Uni e Li became “Uunili,” Lehō e Āhe became “Lehowai.” Makehu townships kept their names; colonial cities got “English” monikers named after their geographical location, economic significance, or some other original story. Examples of this are Southport, a—you guessed it—port on the southernmost tip of Uunili, or Cale’s Hope, a settlement named after a businessman’s mining venture. It’s all details, and chances are that most readers won’t even pay attention, but I personally found that this added a lot of plausibility and immersion.

I could cover a lot more, but this post is already way too long, so I’ll leave it at that—if there’s enough interest, I’d be happy to make a part two. If not, well, maybe at least a couple of you got something useful out of this. If you’re looking for inspiration/references to show to your illustrator/cartographer, the David Rumsey archive is a treasure trove. Finally, for anyone who doesn’t know and might be interested, my novel is called Dreams of the Dying, and is a blends fantasy, mystery, and psychological horror set in the universe of Enderal, an indie RPG for which I wrote the story. It’s set in a Polynesian-inspired medieval world and has been described as Inception in a fantasy setting by reviewers.

Credit for the map belongs to Dominik Derow, who did the ornamentation, and my friend Fabian Müller, who created the map in QGIS and answered all my questions with divine patience. The linguist’s name is David Müller (no, they’re not related, and, yes, we Germans all have the same last names.)

I don’t think “island continent worlds” are inherently unrealistic. An “island continent world” is just what a world might look like in an era when seafloor spreading is at a high point; it’s just the opposite side of the coin of the Pangaea-type continent configuration where most of the continental crust is in a single landmass. In the Cretaceous and Eocene Earth looked a lot more “island continent world” than it does now. An “island continent world” is also more likely if your world has less continental crust than Earth, which is pretty plausible.

Aside on that: a world with highly fragmented continental crust and lots of seafloor spreading will likely have high sea levels, like Cretaceous Earth did. Lots of seafloor spreading implies lots of midocean ridges, which displace water, which means higher sea levels. Sea levels during the Cretaceous got a couple of hundred meters higher than they are now (that’s why you can dig up fish fossils from that era in what’s now the middle of the United States; there was a giant shallow sea there back then), and it wasn’t just because there were no polar ice caps back then. This would tend to accentuate the “island continent world” effect and create a world with small continents and lots of shallow seas and islands and complex fractal coastlines (which means less land, but most of it would have a relatively nice maritime climate, and sea-oriented societies are likely to be prominent). Again, an “island continent world” is the opposite side of the coin of a Pangea world, which would probably have low sea levels and a lot of arid land.

But yes, if your world has an amount of land similar to Earth, some clustering of the continents is likely, and it’s likely that at least some of them will touch or almost touch at some point. And islands will tend to cluster around continents, because islands are often part of a continental platform or part of a subduction zone complex. Which is actually good if you want a low-tech setting that’s big but can be explored and travelled around by your characters! Deep ocean voyaging is very challenging, in a low-tech setting a continent surrounded by thousands of kilometers of ocean in every direction is likely to be very isolated from the rest of its world! In a low-tech setting, continents that almost touch at some point are much more likely to be part of the same cultural, political, and economic “world,” and are much more likely to be places your characters would be familiar with and could feasible travel to. Ironically, a world with less isolation also actually makes worldbuilding easier; less isolation means more cultural diffusion, so you don’t have to work with the assumption that the cultures, religions, etc. of each continent will be totally different.

For anyone interested in these subjects: I wrote a long post of worldbuilding-relevant information a while back, and I also recommend Brett Devereaux’s essays on cities (here and here) and Chris Wayan’s “Carpentry Tips For Worldmakers” page. I’d also suggest James C. Scott’s The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia and Against the Grain: A Deep History of the Earliest States for a look at how the limitations of pre-modern transport shaped pre-modern politics.