When you realise, aged ten in a museum or railway station, that other places shockingly have names for themselves, different to the ones you’re familiar with, it’s like developing a cultural Theory of Mind. This contributes substantially to my enjoyment of international departure boards, even without the clak-clak-clak noise (what will Prague be called here?!)

Last week it occurred to me that I have never seen a map of European cities labelled with their endonyms (names for themselves) and exonyms (external names). And so, I thought I would make one. 

I used a list of the 108 highest-population cities in the EU and rendered the names into its five most spoken languages as best I could (English, German, French, Italian, Spanish). This is the result:

My working hypothesis is that the earlier a city became highly connected, the more exonyms it should have. More languages named it before conventions were standardised, and political control itself was sometimes up for grabs. So an ancient foundation like Naples (Neapolis), which remained great and contested throughout the Middle Ages and into the modern era, has four versions (IT: Napoli, ES: Nápoles, DE: Neapel, EN/FR Naples). On the other hand, nobody ever had occasion to create exonyms for the equally estimable Wigan (EN) or Wuppertal (DE), because both experienced their growth in the industrial age.

In fact, my map includes a long tail of German and British cities whose growth lies mostly post-1700, and who mostly don’t have exonyms. Of the 108 cities, 42 have no exonyms across the five languages at all (allowing for some accent mark differences). German and UK cities together are overrepresented in this humdrum sub-group (24 have no exonyms, from a total of 39).

Even so, sufficiently heavy modern contact can create some orthographic waves. Manchester, a small town before 1700, is spelt the same way everywhere except Spain, where they put a jaunty little quiff on top of the a, and honestly, who can begrudge them that?

Counter-examples suggest other processes. Amsterdam was a medieval and early modern trading city and ought to be prime for an exonym or two, but everyone calls it Amsterdam. Is this a case of the phonetics being so uncontroversial that no other renderings sprang to mind? Is Madrid similar?

At the other end, these are the cities with the greatest number of exonyms (note that some use one of the five test languages).

Mostly these are the historically weighty places one would expect, but again there are odd exceptions. The Hague has five exonyms, though it lacks the historicity of an Athens or a Florence. Probably the explanation is that it’s a toponym – the medieval term for “enclosure/hunting ground”. French (La Haye), Spanish (La Haya) and Italian (L’Aia) seem to be substituting their own terms for that thing, “translating” the name rather than varying it.

I’ve also, hesitantly, included cities (Brussels, Antwerp) with multilingual autochthonous populations, so strictly speaking these are not exonyms at all. It may be that a bigger analysis would find regions where exonyms are more likely to arise regardless because borders are closer and political control has shifted around. The areas flipping in and out of the Holy Roman and then Austro-Hungarian empires, or trading coasts, would be a neat study.

English has proportionately the fewest exonyms and Italian has the most. Often these are of the “put an -o on the end” school (Berlino, Dublino, Salonicco – conceivably this is revenge for Turin, which everybody else refuses to call Torino). Elsewhere, they preserve an appealing Romanitas – Italian is the only language to have an exonym for Plovdiv in Bulgaria, so they make it count with the classical Philipoppoli. It was also news to me that Munich is Monaco di Baviera because one Monaco near one border wasn’t enough, so they duplicated it and then had to differentiate again.

Using this dataset made for some haunting omissions of places which have never grown enough to make the cut (Venice, Venedig, Venise, Venezia, Venecia!). Adding Venice, plus half a dozen French cities of similar venerability, would have somewhat undermined my investigation, but made for a much more fun map. This is why we can’t have nice data.