43 great journeys over 2,500 years, mapped out and back, and graded by one plain question: how sure are we that they happened?
Some journeys are as certain as anything in the past. Magellan's last ship sailed into Sanlucar with 18 men; Amundsen left a tent at the South Pole with a letter in it, and Scott found it. Others rest on a single line in a single chronicle, a saga written down two centuries later, or a story a king told an official in Cairo. Maps of "great journeys" usually draw them all in the same confident ink.
I drew 43 of them, from about 600 BC to 1947: famous voyages, envoys and pilgrims, Viking crossings, lost expeditions and a handful of beautiful legends. Each route is a chain of 405 named stops taken from the Wikipedia article about the journey, with every stop's coordinates and source recorded. Then I graded each journey by its evidence, in four inks: Proven (archaeology or several independent records), Probable (one good written source), Disputed (serious scholarly argument) and Legend (myth, legend or oral tradition). Outward legs are solid, return legs dashed, and a cross marks where a party never came back.
Tap a line or pick a journey to read its story card: who went, when, where they turned, whether they came home, how far, and why it gets its grade. Play the timeline to watch the routes appear in order.
Of the 31 journeys that began before 1450, only 7 are proven. Of the 12 that began after, 11 are. Before AD 500 just one of 12 counts as proven: Zhang Qian's mission to Central Asia, told by a court historian within living memory and followed by documented embassies and trade. All 7 legends begin before 1450.
That is not because ancient travellers went less far. It is because the evidence for an ancient journey is usually one text, often written long afterwards by someone who was not there: Herodotus on the Phoenicians who sailed round Africa, the Book of the Later Han on Gan Ying, the Saga of the Greenlanders on Bjarni. Printed books, logbooks, telegrams and photographs changed that. From Columbus onwards, the question stops being "did it happen?" and becomes "exactly where?".
The grade shows up in the map itself. Of the stops on proven journeys, 3% are conjectural (a place we can only guess at); on probable journeys 13%; on disputed ones 48%; and on legends 62%. Measured by distance, 91% of the line on a proven journey runs between places we can identify, against 8% on a legend.
A legend is not only a story nobody can check. It is a line with nowhere solid to stand.
22 of the 43 parties came home whole, at least as their sources tell it. In 9 more, some came back and some did not: of about 170 men with da Gama, roughly 55; of about 270 with Magellan, 18 on the Victoria; of the four men of Burke and Wills who crossed Australia, only John King. Seven parties never came home, and for five nobody knows.
Coming back is usually how we know. Most of these journeys survive because somebody returned to write them down or tell them, which is why the "never returned" column is so odd. It holds the best-documented losses of all, Franklin, La Perouse and Scott, found through notes, wrecks and bodies, and two legends whose whole point is that nobody returned: Xu Fu, who sailed for the isles of the immortals in 210 BC, and the Roman legion said to have ended up in China.
Laid end to end, the drawn routes come to 725,858 km. The longest single journey on the map is La Perouse's lost expedition at 69,705 km, followed by Ibn Battuta's great journey at 67,563 km. These are drawn distances along my lines, so they are minimums for the real tracks, and they mean the least where the stops are guesswork.
Speed needs dates at both ends, which only 21 passages on the map have, all from the better-documented journeys. Ships under sail made a median 128 km a day across 9 dated passages. Da Gama's crossing from Malindi to Calicut took 26 days with the monsoon behind him (164 km a day); the return, against it, took 96 days (45 km a day), and about half the crew died on the crossing. Horsemen riding to the Mongol court covered 45 to 47 km a day, as fast as a balsa raft drifting across the Pacific.
The two polar parties show how little pace alone explains. Amundsen's dog teams went out at 22.8 km a day and came back faster, at 33.7. Scott's man-hauling party managed 18.0 going out and 18.6 coming back, counting to the day they were stopped by the last blizzard, eleven miles from food and fuel. The slowest passage of all was not a journey anyone chose: Endurance drifted in the pack ice of the Weddell Sea at 3.7 km a day for 281 days before she was abandoned.
Some journeys make most sense in pairs: travellers heading in opposite directions along the same routes, sometimes in the same years. Tap one to see both on the map.
For six journeys the article itself gives a distance I can compare with mine. My drawn routes come out between 71% and 116% of the stated figures. Carpine's ride from Batu's camp to the Khan, which he estimated at 3,000 miles (about 4,828 km), measures 4,701 km on my map. Burke and Wills is the outlier, at the low end: a line through a handful of stops cuts every corner of a real overland route.
The grades are judgements. I set them from what each article says about its sources, using four simple rules, and I give the reason on every story card. Reasonable people would move some journeys: Marco Polo (probable here, because it rests on one book, though details are confirmed elsewhere), the Norse (proven for the crossing, which archaeology shows, not for every saga detail) or Polynesian contact with South America (disputed here, though genetics now makes some contact very likely).
Wikipedia is the source. Every route, date and debate comes from the English Wikipedia article on the journey, read in October 2026. That makes the work checkable and the sources easy to follow, but it inherits the articles' choices, gaps and errors. I checked a few facts against other sources and note them where I used them.
The lines are drawn, not logged. Stops come from named places in the articles. Between stops I drew the shortest plausible path, and I added 306 routing points to keep lines on sensible courses. None of these is a stop, and none is shown as one.
The selection is mine. These 43 are not a sample of all journeys. I chose famous ones, some lost ones and some legends, and tried to cover the world rather than only Europe. A different set would give different shares.
Sources. The full English Wikipedia text of 42 articles about these journeys and their travellers. For each journey I read the article's own account of the route, dates, evidence and debates, and noted the article and section behind every stop.
Places. 405 stops: 128 take the coordinates of the Wikipedia article for that place, 258 come from the GeoNames gazetteer (a nearby town where the article names a region or an old site), and 19 are positions given in the article text itself, such as the latitude where Endurance was abandoned. Each stop is rated known (276), approximate (77) or conjectural (52).
Routes. Land legs are great circles between stops. Sea legs are the shortest path on a half-degree grid of open sea (from Natural Earth coastlines, with narrow straits opened by hand), smoothed into straight runs. Distances are measured along the drawn line.
Grades. Proven: archaeology, physical remains or several independent records of the journey. Probable: one good written source, usually first-hand, with nothing serious against it. Disputed: scholars seriously argue over whether, or how far, it happened. Legend: myth, legend, oral tradition or modern speculation without evidence. "Did they come back?" follows the main source.
Speed. Only for passages with a departure and an arrival date in the article: drawn kilometres divided by days.
Every journey with its grade and outcome. Tap a row to show it on the map.