I know that the idea that everyone thought the Earth was flat until Columbus got lost trying to sail around it is a fallacy. What I’m asking is, did the maps, navigational techniques, devices and methods they used to determine their position, etc. – did they all rely on the presumption that the Earth is round?
Probably not. A lot of early navigation was based on “sailing directions”, which were sometimes depicted in chart form, but the charts look very different from modern maps. Such sailing directions were probably based on earlier experiences, and weren’t always the best route to get where you were going, but they were known routes, that you could trust would get you there.
Any method of determining your latitude from celestial objects is based on the assumption that the Earth is round. It’s easiest with the North Star, if you’re fortunate enough to live at a time when there is a North Star, but it can be done with anything.
But where did the charts come from? One fundamental technique (among others that various navigators used) is to estimate your latitude by measuring the elevation of a star (or the sun):
(a card + knotted string is known to have been used by Arab navigators in the 9th century, not exactly a recent innovation)
I don’t know if this philosophically equivalent to asuming the earth is round, but one needs to understand the concept of latitude somehow.
Erastosthenes calculated the diameter of the world in 240 BC by measuring shadows at different latitudes. This required a good grasp of the true geometry of our planet, in a very literal sense.
Yes, you could do that, and they probably did fairly early on. But sailing directions are even older, and could be based on almost anything. Landmarks, general headings, things like that. “I sailed due east for two days, and saw an island with two mountains aligned north to south. I then turned due south, using the mountains to maintain my course until they dropped out of sight, and then sailed a half day more to arrive at the island where I bought this fancy food. It featured a wide harbour facing north, surrounded by groves of trees.”
There were literally whole books of such directions.
Such books were called rutters. Rutters and the secrecy surrounding them (to the level of state secrets) was a plot point in Clavell’s book Shogun. The earliest maritime charts were the combined information from many rutters.
Maps didn’t. They didn’t develop charts for big enough distances that it mattered. Even the “global” maps were anything but.
Like Chronos said, some navigation techniques do depend on a non-flat Earth, though. But that wasn’t how they drew their charts or sailed their ships.
The group that would have cared the most would have been Muslims in distant lands. That’s one reason why the pre-Modern advances in this area were Islamic
(note that Ancient, as used in the OP, has specific time implications for the actual field of history, that I don’t think was the actual intention given the Columbus cutoff)
Apart from the [to us] logical extension of a rolling horizon and sails that appear over it implying a great curvature without end, Western Mediterranean and Islamic Indian Ocean sailing mainly relied on bearing off from a fixed point and following a path, guided by a star [including the sun] or following a latitude.
AIUI Polynesians used stars as “landmarks” in the same way Mediterranean sailors used mountains. I.e. “sail towards that star until it sets, then aim for this star” and so forth (among other techniques). The fact that the stars obviously would rotate overhead (and some disappear entirely if you sail far enough north or south) coupled with islands appearing/disappearing over the horizon, strongly suggests they would have known the earth was round, but a quick Google plus what I can recall of Christine Thomson’s “Sea People” hasn’t turned up anything definitive.
Following the star until it sets does not make sense, 'cause it moves. However, observing them at rising or setting gives you a definite bearing and Arab, Polynesian, et al. navigators knew all the bright stars and were able to do just that (at night, in good weather).
ISTM if you start to work this out to any non-trivial degree it becomes pretty clear the surface of the planet is spherical, but, on the other hand, an apprentice ancient navigator would have been busy memorizing dozens of stars and landmarks and specific directions to follow, and it does not immediately follow that every single one was giving it abstract or quantitative thought.
Fair, that’s why they don’t ask me to navigate.
A lot of pre-modern navigation was based on landmarks and guesswork, but they still would have had accurate latitudes. For any place out of sight of other land, you’d get there by first getting to the right latitude, and then sailing either due East or due West until you ran into the place. You might not know exactly how far due East or West you traveled, but you could still get there.
If you look at any of those old distorted world maps, the distortion is all in the east-west direction. Find any distinctive point on them, like a horn of land or a river mouth or whatever, and it’ll be at the right latitude.
The problem with the North Star is that you usually want to measure the angle of some celestial object above the horizon, and you can’t see the horizon on a dark night.
That’s not too hard. Above this line, there are stars, and below it, there aren’t. On a truly dark night, there will be enough stars that the edge of them will be fairly well-defined.
Or you can use a measurement based on a plumb line, rather than the visible horizon. Down is down, regardless of what you can see. You can make a simple astrolabe with elementary-school students (admittedly the fact that the protractor is already marked makes it a lot easier).
Indeed, as I understand, the traditional Mercator projection we all remember fondly from our grade school classroom maps, where Greenland is as big as Africa, and northern Canada is “yuge!”, has a more sailorly purpose - any line from A to B drawn on that map is a straight line of constant compass heading. (except, of course, magnetic north is not true north and moves around over the centuries.)
Which gives a good hint - navigation without landmarks involved learning (hence, rutters) what direction it was from A to B or to intermediate waypoints. Stars near the horizon would give a hint as to true directions. Half the fun was probably estimating to what extent you were drifting off course, and recognizing new landmarks that gave you a hint how close you came to the next waypoint.
That helps, certainly, but how do you determine what direction you’re traveling in? It’s not as simple as just what direction the prow of your ship is pointed in, even in simple cases like a wind that happens to be directly astern of you, because the water is moving, too.
There are many different sorts of maps used for many different purposes in navigation. On a Mercator projection, a straight line is a curve of constant heading. On a polar projection (which could be centered on any point, not necessarily a geographic pole per se), a straight line is a great circle. Both of those will distort distances; there are other maps you can use that preserve distances, at least from some points. Or maps that keep areas accurate, or maps that try to compromise and do an adequate but not completely correct job on all of these.
None of those things are as accurate as actually seeing the horizon, especially on the moving deck of a ship at sea.
I’ve actually used a sextent on a ship. To be really accurate, I had to estimate the height of my eye above the water level. My view to the horizon was at a slight downward angle. It was only a fraction of a degree, but enough to make a difference.
After doing all the math, I was off by about 20 miles. The second mate was off by half a mile.
That’s why, I presume, rutters - someone else has made the voyage, perhaps several times, and noted “when you think your prow is headed X° it could be off by such-and-such due to current - so use this heading instead.”
it was only in the Pacific, and perhaps if aiming for the Azores, where the problem is if you miss, you don’t hit anything for a while. By the time sailors were worried about the Pacific islands, most of those navigation problems had been solved. (one clue used by both Polynesians and Columbus on his first voyage, was seeing land birds, one of the clues that land was nearby - and possibly helped by the direction the birds tended to fly)
Off by 20 miles might be plenty good enough. Let’s say that you’re trying to reach Tristan del Cunha, the most remote inhabited island on Earth. The highest point on the island is over 2 km above sea level, meaning that you’ll be able to see it from 160 km away (more, if your ship has a high crow’s nest). And as @md-2000 mentions, you might even be able to determine that there’s land nearby from further than you can actually see it, from clues like shorebirds or deflection of waves or whatever. Almost really is good enough, here.