Great read, felt quite informative. I can't shake the sense that everything on wikipedia is naive historiography in the truest sense of the term, but this story about Occitania seems fun.

Expand full comment

> The first written account of the future language of Gaul is from 842 when the Oaths of Strasbourg were written down. This oath was supposed to be taken by the Frankish soldiers and it therefore needed to be notated in the language these soldiers were actually speaking and could understand.

I don't think this is right; weren't the Oaths spoken by the Frankish knights in Vulgar Latin for the sake of the Gaulish soldiers' understanding?

(Yes, "Vulgar Latin." If the term was good enough for H.P. Lovecraft, it should be good enough for any Anglophone linguist. What French linguists have to say about it is their own affair)

Also Anders I don't know if you like retro games, but you might like Locomalito - he's a Spanyard who uses a lot of historical themes. He has a free game called l'Abbaye des Morts based on the Albigensian Crusade: https://locomalito.com/abbaye_des_morts.php The younger Pie children play it a lot.

Expand full comment

H.P. Lovecraft aside, I think one of the problems with the Vulgar Latin terminology is the difficulty of determining where it starts and where it ends. The Romance language in the Oaths of Strasbourg is of course some sort of descendent of Latin. But what sort?

Ironically, I understand more or less nothing of the Latin version of the Oaths of Strasbourg, despite it being a predecessor of French, which I do speak. However, I can understand at least some of the Germanic version of the same text, because of its similarity to Swedish. I am not sure what point I want to make out of this. But one interpretation is that the proto-French of the 9th century went through more development in the following centuries than the proto-German.

Expand full comment

I know what point I want to make out of this. I want linguistics to be a better science. Properly speaking, something like the rate of linguistic change could be traced and measured. The tendency for words to be lost or neologisms to be coined, for phonemes to be dropped, added, or changed to specific words, or for entire sound shifts to occur, should be something linguists follow and attempt to predict. Whenever I ask linguists about this they hem and haw and fall back on abstruse vocabulary and the IPA. They know the difference between plosives and nasals, SVO and SOV languages, ergative and absolutive languages, languages with and without tone or gender, and so I know those differences as well because they are all extremely simple. But how to predict whether, when, why, or how a language will change is something extremely elusive. The only predictions I've ever seen, I just found this moment:


But I know linguistics enough by now that I'm not holding my breath.

Expand full comment

Fascinating history, and I thank you for writing it.

One question bugs me about the geography, though: there are mountains (The Pyrenees) right through the middle of the Catalan-Occitan region. Why didn't the mountains do more to split up the language?

Expand full comment

My geography isn't great, but looking into the Pyrenees it looks like a lot of the western range is pretty low. And if you look at the Camino se Santiago, there are two pilgrimage routes that cut through the Western Pyrenees:


Imagine further all those Cathari chased into the hills as the persecution just began; that might do something to increase linguistic interchange along both sides.

Expand full comment

Good question. To which I do not have a good answer.

In the book about Montaillou we can read that there was quite some traffic across the Pyrenees by shepherds and others. But none of these can be called culture-bearers and should not have affected the development of the respective languages in any significant way.

Along the coasts traffic was easier, which might explain part of the affinity. Borders were definitely more fluid along the coasts. If you look at the very first map in this article, the one with remnants of Occitan language in France, you see that there is a small area in the far south of France, between the Pyrenees and the Mediterranean, which is marked as not Occitan-speaking. This is not really correct. The area in question is Roussillon, which was Aragonese and then Spanish until the French conquered it in 1659. The reason Occitan is not spoken there is because Catalan is spoken instead (which is more or less the same). This area is sometimes also known as North Catalonia. This should prove that there could be some communication between the two sides of the Pyrenees.

If I should make a guess as to what the main reason that Occitan and Catalan have not diverged is, I would opt for the explanation that Occitan was suppressed in France. I have not heard of any post-Medieval Occitan literature. On the other hand, there is plenty of post-Medieval Catalan literature. I suspect that without any domestic Occitan "high culture" the Occitans looked to Catalonia for cultural inputs which made the Occitan language stay fairly close to Catalan. But I have absolutely no proof for this, it is speculation, pure and simple.

Expand full comment

> there is a small area in the far south of France, between the Pyrenees and the Mediterranean, which is marked as not Occitan-speaking.

On the other hand, there is a little corner of Spain marked as Occitan-speaking: the Aran Valley (<https://en.m.wikipedia.org/wiki/Val_d%27Aran>), the only remaining bit of Spanish land north of the Pyrenees (Llívia is on the Pyrenees themselves, <https://en.m.wikipedia.org/wiki/Ll%C3%ADvia>), and the only place where Occitan (called Aranese) is an official language today.

As far as I know, Occitan and Catalan do have considerable dialectal variation (the loss of a learnèd standard will do this to any language), and, while it was probably arbitrary to label some of those dialects as Occitan and others as Catalan, the distinction is now entrenched; for example, in the Aran Valley, there are separate official standards for Aranese and Catalan—in addition to Spanish, of course.

Modern standard Catalan was developed mostly in the 20th century, and there was—sadly, I’d say—little interest in reïntegrating it with the already dying Occitan. They chose an arguably awkward orthography, close to the Spanish one, with the idiosyncratic digraph _ny_ (corresponding to Spanish _ñ_ and French and Italian _gn_), but never got around to extending this paradigm to _ly_, so Spanish-like _ll_ has stuck, and, in turn, forced the consolidation of _l·l_ for a geminate ell, which was once regarded as a temporary stopgap. They didn’t bother to preserve written final _n_s still pronounced in some Occitan dialects.

Meanwhile, what little Occitan is still written these days, including Aranese, tends to use the classical orthography, which retains those pesky supradialectal letters dropped from Catalan, and uses _nh_ and _lh_ for the Romance palatal consonants. These digraphs are much more widely known today from Portuguese, but it actually borrowed them from Occitan in the Middle Ages, when the latter was an important literary language.

Expand full comment

Today in 'losing battles has consequences', the good lord bless you for this fascinating write up

"King Peter moved north in 2013 to assist Toulouse which was being harassed by the crusaders. At Muret, a small town 25 km south of Toulouse, he made contact with the crusader army. The severely outnumbered crusaders made an immediate attack and in one of the most famous battles of the High Middle Ages not only routed the Aragonese army but also killed king Peter."

Expand full comment