A Reclassification of the Occitan Language

Updated April 12, 2014. Long, runs to 65 pages.

Map of Occitania showing the major dialect divisions including Vivaro-Alpine.

Map of Occitania showing the major dialect divisions including Vivaro-Alpine.

According to Ethnologue, Occitan is currently 1 language. This reanalysis will expand Occitan from 1 language to 23 languages.

Occitan, or Langue d’Oc, is spoken in general in a swath across the south of France. It goes a bit into Spain in the Pyrenees and into far northwestern Italy. There is Occitan an outlier in Italy.

There are various classification methods for Occitan. One is to differentiate between Langue d’Oil (French) and Langue d’Oc (Occitan). I do not agree that Occitan is particularly close to French. Occitan is about as far from French as Spanish and Italian are.

I would put the Oil languages (including French) in a Northern Gallo-Romance and Rhaetian and Italian Gallo-Romance into a Southern Gallo-Romance with Arpitan as transitional between the two.

Another view of Occitan.

Ibero-Romance is a different split altogether. Occitan is better placed into a separate Romance category that I would call Catalan-Occitan. This analysis sees Occitan and Catalan as a singular branch of Romance. Catalan-Occitan is then properly put into Ibero-Romance.

It also recognizes that Occitan and Catalan were once a single language stretching across the south of France and into northwestern Spain. This language was very widely spoken, and at one time in the Middle Ages it was very widely used. It is “the language of the troubadours,” the wandering minstrels who plied their trade across Southern Europe in the Middle Ages, though in truth, the troubadours mostly came from Limousin and wrote their songs in a sort of Poitou-Limousin dialect that no longer exists.

Another map of Occitania. This one is a bit harder to read as it is largely in dialect, but if you study it a bit, it makes more sense.

Another map of Occitania. This one is a bit harder to read as it is largely in dialect, but if you study it a bit, it makes more sense.

From 500-1200, there was really only one language – Catalan-Occitan. Occitan only started distinguishing itself after 1200.

At the moment, Southern Languedocien has the closest relationship of all to Catalan – in fact, they are intelligible. Gascon is then the next closest to Catalan, but intelligibility data is lacking. The rest of Occitan is more distant from Catalan. Catalan speakers have a hard time understanding Auvergnat, Limousin, standard Languedocien and Provencal.

It is not the case, as often stated, that Catalan and Occitan are intelligible, but they are close. Catalan and Occitan probably have about 50% inherent intelligibility when spoken, much more in writing. Speakers of only Catalan and Spanish report a hard time understanding Languedocien Occitan TV broadcasts. Cultivated speakers could pick up the other language in a few weeks with coaching. The differences between Catalan and Occitan are said to be greater than among the Scandinavian languages.

A Third Map of Occitania. This is the best-laid out of all, and it is the easiest to read and make sense of.

A Third Map of Occitania. This is the best-laid out of all, and it is the easiest to read and make sense of.

Occitan has been on decline for a long time, as the Langue d’Oil has been been supplanting it for centuries. The decline began in 1539 when a French king ordered that langue d’oil be the official language of all of France. Despite a brief revival in the 1800’s, it’s been downhill ever since. Occitan speakers did not start speaking French in large numbers until 1885. Before that, there was only minor French influence on spoken Occitan. Since 1885, French influence on spoken Occitan has increased, in some cases dramatically.

The codification of the Parisien Langue d’Oil language as Standard French with the victory of the French Revolution and the corresponding fascist Jacobin war on all other languages caused Occitan to recede further into the background.

The French government is reactionary/fascist on the subject of language. The Jacobin Constitution baldly states that “French is the language of the state” and allows for no other languages. Hence, Occitan receives no state support in any way.

Occitan still has about 8 million people who can understand it and 3 million who can speak it to one degree or another. Estimates of the true number of speakers range from 1-3.7 million. Occitan is surely a modern language and does not lack for vocabulary – it has between 250,000 and 1 million words, though many say that this is an exaggeration.

Occitanists like to say that Occitan is all one language, but this is a political statement. They say this in order to unite the dying language and prevent it from splintering.

The Occitanist position is increasingly popular. For instance, Wikipedia is calling all of the Occitan languages “dialects.”

There are two centralized ways of writing all Occitan dialects, one based curiously enough on a Medieval standard. Around 1850, an Occitanist poet named Frederic Mistral invented a standard based on his own Provencal language, but this solution has not caught on well. The second is neo-Occitan, a new koine language created more recently.

Occitan is spoken most often by those over 50, except in Italy and Spain. Occitan is only protected in Spain and Italy, where the respective forms of Aranese and Transalpine Provencal are spoken.

If you hear Occitan, it sounds like some curious cross between Spanish and French, sort of the way that Catalan sounds.

It’s not true that Occitan is one language as the Occitanists centered in the south of the region insist.

The intelligibility among Occitan lects seems to be as I suspected. People with exposure to the other lects can pick them up pretty quickly, but someone who has never heard the other varieties has a hard time understanding them. This is called learned bilingualism. If learned bilingualism is the rationale for saying that Occitan is a single language, it stands on precarious scientific grounds.

Nevertheless, intelligibility in Occitan remains a very controversial subject. On the one hand, speakers say they can’t understand speakers of the same lect a few miles away; on the other hand, Occitan speakers can understand totally different varieties from far away very well.

At this point, it is time for some scientific intelligibility testing to sort all of this contradictory information out. Intelligibility testing has already been done with Occitan. It did find high, but by no means complete, intelligibility between major Occitan lects (Bec 1982). This suggests that intelligibility among Occitan lects is marginal, possibly on the order of 80% among major forms. I would appreciate it if anyone privy to this study could contact me.

Others have put the figure about where I did – at 70-85% intelligibility between major Occitan languages.

It is often said that French and Occitan speakers can communicate well enough. This is not the case. There are many French speakers living in Occitania who say that they cannot understand a word of Occitan.

A good overview of Occitan is here.

Method: Literature and reports were examined to determine the intelligibility of the various dialects of Occitan. >90% intelligibility was considered to be a dialect of a major Occitan language. <90% intelligibility was considered to be a separate language split off from Macro-Occitan. The emphasis was on intelligibility more than than structural factors, but structural factors were included.

Results: This treatment expands Ethnologue’s 1 Occitan language to 23 Occitan languages.

A great map of all of the languages and dialects of SW Europe. It's in Spanish, but you should be able to understand it anyway. The Occitan languages are the light green stretching across all of southern France.

A great map of all of the languages and dialects of SW Europe. It’s in Basque, but you should be able to understand it anyway. Occitan dialects are listed in light green in the area around Southern France.

Gascon is a Southern Occitan macrolanguage spoken in southwestern France and barely over the border into Spain. It has 256,000 speakers, 250,000 in France, but other figures put the number at 500,000. Gascon has some affinities to Basque – it is said to have a Basque substrate – but it is not close to Basque at all. Gascon is probably closer to Catalan than anything else (even closer than it is to Aragonese), however, there is no continuum between Gascon and Aragonese at the border. Instead, there is an abrupt transition.

Gascon is best seen as its own variety of Occitan, outside of both Northern and Southern Occitan. It has heavy influence of Basque and Aragonese that makes it very different from the rest of Occitan.

Along with the Basque substrate, Gascon is a transition between Ibero-Romance (Occitan proper – Aragonese – Catalan) and Gallo-Romance (langues d’oil).

In contradiction to the Occitan centralizers, Gascon speakers say they speak a separate language and resent both being referred to as speakers of a dialect of Occitan and what they see as the cultural imperialism of Occitan politics centered in Toulouse.

In France, Gascon is spoken in the departments of Landes, Gers, Hautes-Pyrénées, the eastern parts of Pyrénées-Atlantiques and the western parts of Haute-Garonne and Ariège, and it is still used actively by many people.

50 years ago, near La Réole, France, there were still monolingual Gascon speakers among the older people. People were still being brought up speaking Gascon as recently as the early 1980’s. However, in France, it is not being taught much to children.

Gascon speakers have a hard time understanding Limousin, and Languedocien speakers say it’s hard to understand Gascon and vice versa. However, it is easier for Gascon speakers to understand Languedocien (though intelligibility is still difficult) than vice versa due to the French-like regularity of Languedocien. For example, “How are you?” is “Quin hes?” and “Cossi fas?” respectively in Gascony and Languedoc. It goes on like that through the Gascon-Languedocien lexicon. It’s clear we have two completely separate languages here.

Around Agen, there is a transition between Guyennais, Gascon and Languedocien. One village can understand the next, but once you get 10-15 miles away, things get difficult. Provencal speakers say Gascon is a foreign language. Gascon speakers can’t understand a word of Auvergnat.

It makes sense to split Gascon into a West and East Gascon. The border would run from Artix-Pau in the south to Marmande-Agen in the north. East Gascon would then start at around Pau and Agen and West Gascon at Artix and Marmande. These distinctions represent the variant influences of Bordeaux in the west and Toulouse in the east.

West Gascon is spoken from Bordeaux in the north to the Basque country in the south. In the east, it runs to Artix in the south and to Marmande in the north. Its differences with East Gascon revolve around the influence of the large city of Bordeaux on the language.

Dialects include Bazadais, Marmandais, Bordalés and Médoc.

Bazadais is spoken around the town of Bazas, famous for its cows. Marmandais is spoken around the town of Marmande.

Bordalés is spoken around Bordeaux. It is probably in very bad shape. It was declining badly even 50 years ago. There is actually a transitional Occitan-langue d’oil (Saintongeais) region around Bordeaux.

The region around Bordeaux is notorious for its sharp linguistic breaks. One early chronicler estimated that the distance between West Gascon Limonde and Saintongeais-speaking Pays Gabay north of Limonde to the Saintonge border was 50% (further than English and German), while the distance between Limonde and Perigord Limousin speaking Montpon-Menesterol just to the east was 25%.

There is evidence of a very sharp ancient linguistic border between Gascon and Old Saintongeais-Limousin on the southern border of Saintonge from St. Cliers to Coutras.

Médoc is a dialect spoken on the Médoc Peninsula south of the Gironde.

Landes is a West Gascon language spoken in southwestern maritime France in the Aquitaine region. As it is not even intelligible within itself (it differs so much that it is hardly intelligible even from village to village), it must be a separate language. Some say that Landes is nearly a dead language, but others say that it is still spoken in the villages.

The coast near Biscarosse gave up Landes long ago, but now even in the inland villages like Rion de Landes and Parentis en Born it is hard to find a speaker. The real Landes died around 1950. The current dialect is very Frenchified.

East Gascon is spoken from Pau to the Ariege River in the south and from Agen to Toulouse in the north. It represents the influence of the large city of Toulouse. Even between the cities and Pau (East Gascon) and Artix (West Gascon), which are very close together, communication is nearly impossible.

This language is probably in very bad shape. It is probably extinct in the Rivière-Basse region around the towns of Marciac, Plaisance and Maubourguet and in the Vic-Bihl region just to the west around Riscle. In Tarbes, Lannemezan and Lourdes, speakers are almost impossible to find. The eastern border with Languedocien is in the Ariege.

Neraqués and Lomagne Gascon are two East Gascon dialects. Neraques is spoken in Nerac, just southwest of Agen in the Lot et Garonne. Lomagne Gascon is spoken to the far northeast of the Gascon language, southeast of Agen down towards Toulouse.

Pyrenean Gascon is a macrolanguage that is unintelligible with the Gascon of the plains. This language is the most divergent member of Occitan, probably due to very strong Basque influence. Some would put it outside of Occitan proper altogether.

The borders of Pyrenean Gascon run from the Ariege in the east to Bearn in the west and to the Spanish border (except in the Aran Valley).

Pyrenean Gascon is nearly a dead language in France, only spoken by 1% of the population. In the region of Tarbes, Lourdes and Bagneres, there are almost no speakers left.

In Bearn, Pyrenean Gascon is still heavily used. In 1994, fully 26% of the population spoke Béarnais. Since then, the number has probably collapsed precipitously, possibly down to 5-6%.

It makes sense to split Pyrenean Gascon into three separate languages. Gascon speakers in the east of Bearnais have a hard time understanding the speakers in the west of Bearnais. They also have a hard time understanding the Couserans spoken in the Upper Ariege near the Foix and Andorra.

Although it makes no linguistic sense, Bearnese is often split off a separate dialect of Pyrenean Gascon. Dialects of Bearnese include Aspés, Ossau Bearnese, and Palois. Bearnese is spoken in Bearn.

West Pyrenean Gascon covers the western part of Bearn. It is here that there is the heaviest Basque influence of all. Speakers in the east of Bearn can understand speakers just to the east in Bigorre and Lourdes than they can the speakers of western Bearn.

Oloronais (Aspois) is a dialect of Béarnais spoken in Oloron that borders on Souletin Basque. The actual linguistic border between Béarnais and Basque is in between Aramits and Tardets.

Central Pyrenean Gascon covers most the Pyrenean Gascon region from eastern Bearn all the way to Ariege. Intelligibility is poor with both western Bearn and the Couserans in the Ariege.

Bigourdan is a dialect of Central Pyrenean Gascon spoken in Bagneres de Bigorre region. Subdialects are Argelès, Aure, Bagnères, and Tarbais. Bagnères is spoken around the city of Bigorre itself and Tarbais is spoken around the town of Tarbes.

Eastern Pyrenean Gascon is spoken in the far east of the Pyrenean Gascon region by the border with Languedocien and Catalan and over the border into the Aran Valley. Central Pyrenean Gascon speakers have a hard time understanding those in the Couserans in the Upper Ariege by Foix, Rousillon and Andorra.

Dialects include Aranese, Ariegois, Commingese, Couseranais, Sauratois, and Contadels.

Aranese is an Eastern Pyrenean Gascon dialect spoken by most of the 6,000 people living in the Aran Valley in the Spanish Pyrenees, where it has official status. It has Spanish, Aragonese and old Catalan influences, but at the moment it is under very heavy Catalan influence such that many Occitanists regard it as an outrageously degenerated dialect.

Aranese is intelligible with Commingese across the border in France. Aranese is not intelligible with Spanish, French, Catalan or the rest of Occitan.

Pujolo and Canejan are Aranese dialects.

Ariegois is a Pyrenean Gascon dialect spoken in the Upper Ariege.

Sauratois is an Ariegois dialect spoken in the Saurat region northeast of Tarascon on the Ariege River. Couseranais is an Ariegois dialect spoken in the Couserans northwest of Andorra. It still has a few speakers. Contadels is an Ariegois dialect spoken in Vicdessos north of Andorra. There is a very heavy Languedocien and Catalan influence on this dialect. This is actually a Gascon-Catalan transitional dialect.

Southern Occitan is a branch of Occitan that stretches across Southern France near the ocean. It includes Languedocien, Maritime Provencal, Nissart and Rhodanian Provencal. This branch has more Iberian influence in the west and more Ligurian Southern Gallo-Romance influence in the east.

Languedocien is a Southern Occitan macrolanguage that has 1 million speakers in an area in a line going from north of Andorra – Aude – Fenoullens – Leucate in the south (border with Catalan), from Toulouse to Oust in the west (border with Gascon), in a line running from Toulouse – Albi – Agde in the north (border with Guyennais) and at Bassin de Thau in the east (border with Provencal).

Languedocien sounds like a mixture of Spanish and French in the north or Spanish and Catalan in the south.

Languedocien speakers have a hard time understanding Limousin, Auvergnat and Gascon. Languedocien speakers have a hard time being understood by the Provencal speakers in Toulouse.

Along with Provencal, this language is more conservative and closer to the Medieval Occitan.

If you try to learn Occitan now as a second language, you will learn Languedocien. Attempts to standardize writing of Languedocien have not been successful. An Occitan koine is being promoted out of the University of Montpellier that some Occitan speakers have referred to as an Occitan Esperanto.

All across Languedoc, most of the older people and many young people still speak Languedocien. In Carcassone, all street signs are bilingual in Occitan, Occitan is an obligatory subject for primary school students, and there are 22,000 speakers in the city. Nevertheless, it is not being learned much by children in general in the region as a whole.

It makes sense to split Languedocien into a Ibero-Languedocien and a North Languedocien (or Franco-Languedocien), the first more like Catalan, Spanish, Gascon and Aragonese and the second more like French and the rest of Occitan.

North Languedocien is a Languedocien language with borders running from Toulouse – Albi – Bassin de Thau in the north and east and around the Bages-Sigues Lagoon in the south. This language lacks the strong Catalan influence of Ibero-Languedocien. Instead, it has more French influence.

There are various dialects within North Languedocien that are quite divergent. Dialects include Besierenc, Narbonés, Carcassés, and Pezenas.

These are spoken around the cities that they are named after and are said to be unrecognizable from one region to the next, but until we get specific intelligibility data, we can’t split them.

Ibero-Languedocien is spoken in the south from Toulouse and Albi down through the Ariege, the Foix, the Aude, the Fenouillines and over to the coast at Leucate, possibly extending north to Carcassonne and Narbonne. This language is rooted in Iberian phonetics.

Ibero-Languedocien speakers feel that they have excellent communication only with speakers of this language and with Catalans. With the rest of Occitan, they feel that they are speaking another language, and there are communication problems.

Ibero-Languedocien is intelligible with Catalan. This dialect is the closest of all Occitan lects to literary Catalan, and is spoken in the part of southwestern France right next to Catalonia. Ibero-Languedocien speakers can understand Catalan easier than they can understand Gascon.

The border between Ibero-Languedocien and Catalan proper begins in the Languedocien-speaking Fenouillèdes along the Agly River. To the south, Catalan is spoken – to the north, it is Languedocien. But that boundary is fairly sharp. On the coast, the transition zone occurs from Leucate to Le Barcares and Salces. The true transition zone occurs in the area north of Andorra. The Catalan of Formigueres is basically the same language as the Languedocien of Usson just to the north.

Tolosenc is a dialect of this language spoken around the city of Toulouse. It has Gascon influences. In the rural areas around Toulouse, almost everyone over 25 understands Tolosenc. In this area, many people over 40 were raised speaking Tolosenc as a first language, but most have forgotten it by now. However, in Toulouse proper, Occitan speakers have gone from 50% in 1950 to 10% today. Foissenc is spoken in the Foix.

Agathois is a divergent Languedocien lect spoken on the coast town of Agde. It is very different from the Besierenc dialect spoken in Beziers and Vias, which were wine-growing regions. Beziers and Vias received many Spanish immigrants to pick grapes in the vineyards, and received many more during the Spanish Civil War. As a result, Besierenc now has heavy Spanish admixture. But Agde, on the coast, received no Spanish influx, and now communication is sometimes difficult between Agathois and Besierenc speakers.

Provencal is a very famous Southern Occitan macrolanguage that is spoken further east than Languedocien all the way to the Italian border. It has 200,000 speakers. Provencal is spoken in the departments of Alpes-Maritimes (except the eastern corner), Bouches-du-Rhône, Var, Vaucluse, in the southern parts of Alpes de Haute-Provence, and the eastern parts of Gard.

Provencal is said to be close to the Gallo-Italic Piedmontese language.

Auvergnat speakers say they cannot understand the Mompelhierenc spoken in Montpellier, and there is marginal intelligibility with Nimes and Sète. People with one parent who spoke South Auvergnat and another who spoke Provencal were not taught Occitan because the lects were too different. This implies that even South Auvergnat has poor intelligibility with Provencal.

Limousin speakers who move to the Provencal region say that the two feel very much like separate languages . Provencal speakers say that Gascon is a foreign language. Provencal speakers cannot understand Vivaroalpine. Provencal speakers even have a hard time with Languedocien.

Provencal speakers, especially those around Marseilles, deny that they speak a dialect of Occitan and instead say that they speak a separate language called Provencal. They see Occitan as a code word for Languedocien and attempts to make that the de facto Occitan language.

Provencal, along with Languedocien, is closer to the Medieval Occitan language and is more conservative.

Dialects include Cévenol, Maritime Provencal, Marsillargues, Mompelhierenc, Bas-vivarois, Lunellois, Aptois, Bagnoulen, Barjoulen-Draguignanen, Canenc, Coumtadin, Foursquare-Manousquin, Grassenc, Marsihés, Maures, Castellane Provençal, and Sestian.

Cévenol is spoken in the Cevennes Mountains north and northwest of Nîmes and is doing well. Maritime Provencal is spoken around the Cote d’Azur, is doing well and is widely spoken, especially as Marsillargues in Marseilles. Mompelhierenc, spoken in Montpellier, has heavy Languedocien influence. Bas Vivarois is spoken in the lower half of the Ardeche region. Lunellois is spoken in Lunel between Montpellier and Nimes and still has speakers.

Aptois is spoken around the town of Apt north of Marseilles. Barjoulen-Draguignanen is spoken around the towns of Barjemon and Draguignan in the hills north of the French Riviera. Canenc is spoken around the Cannes. Grassenc is spoken on the French Riviera.

Rhodanian is spoken around Arles, Avignon and Nîmes, is apparently not intelligible with the rest of Provencal and may be more than one language. Rhodanian speakers from around Nîmes say that they cannot understand other speakers from villages only 12 miles away. This is actually a Languedocien language that underwent Provencal phonetic changes in the late 1700’s, resulting in a Provencal tongue. This probably accounts for its diversity. Rhodanian is more than one language.

Dialects include Arlaten, Bagnoulen, Camarguen and Nimoues. Arlaten is spoken around Arles. Bagnoulen is spoken around the town of Bagnols sur Centre. Camarguen is spoken around Camargue Bay. Nimoués is spoken in Nimes.

Nissart is a separate Southern Occitan language spoken in Nice. It has poor intelligibility with Rhodanian and Maritime Provencal. It has very heavy influence from the local Ligurian Gallo-Italian dialects. It is best seen as a transitional language between Gavot Vivaro-Alpine and Ligurian. Based on its history, a more proper analysis would be of a Ligurian language that became Provencalized by Alpine Provencal speaking immigrants from the mountains coming to work on the coast after 1861.

The Nizzardos, residents of Nice, spoke a Ligurian dialect before Nice was taken over by France in 1860; since then, much French has gone in. It is similar to the Mentonasque and Monegasque spoken in Menton and Monaco (the first Occitan and the other Ligurian). Intelligibility between Nissart and Royasque Ligurian is very limited .

Nissart is in very bad shape; it is a dying language mostly spoken by older people, when it is spoken at all.

Dialects include Esteron, High Vésubie, and Northern Nissart.

Mentonasq is a curious Gavot Alpine Provencal dialect related to Nissart spoken near Monaco in and near the town of Menton. It has a lot of Ligurian influences like Nissart. This is intelligible with Nissart and is apparently a Nissart dialect.

Best seen as transitional between Nissart and Intermelio to the east, a Ligurian dialect with strong Occitan influence. Studies have shown that Mentonasq is between Gavot Alpine Provencal (Nissart) and Royasque (Brigasc)-Pignasque (Ventimiglian) Ligurian (spoken in the Roya Valley in France and Pigna in Italy on the border), with an emphasis on the Occitan. About 2/3 of the words are Provencal.

There are still those who insist that this language is basically Ligurian with strong over layer of Provencal. Intelligibility between Mentonasq and Ligurian Royasque is better than between Nissart and Royasque, but is still somewhat marginal.

Although it is close to Nissart, it is also quite different from it. Monegasque Ligurian is quite different from Mentonasq. Mostly spoken by older people, fisherman and rural types. There is bilingual signage. But the language is in bad shape as the young do not speak it and there are many tourists.

Roquebrunasq is a dialect of Mentonasque, spoken on the Roquebrun-Cap Martin just to the west of Menton. It is somewhat different from Mentonasque. It is dying out. Similar dialects Gorbarin and Castellarois are spoken in Gorbio and Castellar. Gorbarin is particularly close to Mentonasc. Like Nissart, these are Gavot dialects transitional to Ligurian.

Northern Occitan is a branch of Occitan that is spoken to the north of the Occitan region and also over by the Italian border. There are great differences between Northern and Southern Occitan. For instance, 30% of the vocabulary of Auvergnat is not found in Southern Occitan.

One way to look at this is to say that the languages in this region – Limousin, Auvergnat and Vivaro-Alpin, are part of something called Medio Gallo-Roman, which is really in between the langue d’oc proper of the south – Gascon, Languedoc and Provencal – and the langues d’oil to the north and Arpitan to the east. Another way to look at it is to say that Northern Occitan is closer to Arpitan than to the rest of Iberian-dominated Southern Occitan.

Limousin is a Northern Occitan macrolanguage spoken in France, has over 100,000 speakers. It is spoken in Limousin Province and over the western border into the far eastern part of Saintonge and the Perigord in North Acquitaine. North Perigord in Acquitaine Saintongeais influences.  South Perigord speaks Guyennais.

Limousin is still widely spoken in the Limousin region and in northern Dordogne in Acquitaine.

Limousin may have once been many separate languages, at least in the Dordogne department. Older residents in the Périgord Vert near Nontron report that from 1930-1970, it was not unusual for different villages to have Limousin dialects so different that one village could not understand the next, and they had to resort to the use of a koine.

Gascon, Provencal, Languedocien and Auvergnat speakers say they cannot understand speakers of Limousin.

Charente Limousin is a Limousin dialect that is very hard to classify. It extends from Confolens south to Aubeterre. This is an Occitan-Oil transition zone with an emphasis on the Occitan. So these are Limousin dialects transitioning to Charentais langue d’oil.

Between Confolens and Ruffec around Chatain, there is a transitional dialect between langue d’oc and langue d’oil that is nevertheless intelligible with the Charentais spoken in Ruffec. This is probably a Charentais dialect transitional to Limousin.

This province is generally langue d’oil speaking and has been so since the original Limousin speakers were eliminated by the Black Plague in the 1300’s and replaced by langue d’oil speakers, but the area around the Charente River in the far east of the province has long spoke Occitan and never underwent replacement.

Saint-Eutrope and Montberonés are Charente Limousin dialects . Montberones is spoken in Montbron, and Saint-Eutrope is spoken in the town of the same name.

South Limousin is a separate language spoken south of Haute Vienne in Limousin south to the Limousin border. It is closer to Auvergnat and Languedocien.

Haute Vienne North Limousin speakers understand no more than 60% of the South Limousin of Ussel. Between Upper Limousin in Limoges and Lower Limousin in Brive, there are many confusing phonetic changes that make it hard for North Limousin speakers to understand Brive speakers.

Corrèzese is a dialect of South Limousin spoken around the city of Correze. Correzese speakers can understand Auvergnat and vice versa. Corrèzese is best seen as a Limousin dialect transitional to Auvergnat.

Sarladais is a South Limousin dialect spoken in Sarlat in Aquitaine just southeast of Limousin. It has strong Guyennais influences.

Monédières Limousin, a variety of South Limousin spoken in the Monédières Hills near Correze, is a separate language. For one thing, it does not even appear to be intelligible within itself. Some varieties of Bas Limousin in the Monédières Hills near Correze have a hard time understanding each other. For another, Limousin speakers say they have a harder time understanding Monédières Limousin than they do Auvergnat as a whole. This is more than one language.

Guyennais is a highly divergent lect, possibly a separate language, spoken in a swath across central Acquitaine, northern Languedocien and southwest Auvergnat. It is transitional between Gascon, Languedocien, Limousin and Auvergnat. In the South Perigord, the influences are Saintongeais, Gascon and Languedocien. To the east, the influences are Languedocien, Dauphinois Provencal and Auvergnat.

In the north, the boundary with Limousin and Auvergnat is a line from Bordeaux – Bergerac – Carlux – S. Cerre – Latronquiere – southern border of Auvergne to the Ardeche border. To the south, Guyennais borders Languedoc along a line running from Castelsarrasin – Montalban – Cordes – Albi to the border of Languedoc at Millau and the Cevannes.

Guyennais is still widely spoken. In Saint Cirq in Dordogne Department, all of the elderly speak Guyennais as a first language and continue to use it amongst themselves at all times.

Although Guyennais is typically lumped under the rubric of Languedocien, others lump Guyennais in with Limousin, saying that there is no way that Guyennais-Limousin is the same language as Languedocien-Gascon. The best view is that Guyennais was close to Limousin and Auvergnat, but it underwent extensive Languedocienization caused by the expansion of Toulouse to the north from the 800’s to the the 1200’s. At the moment, it is probably closest to Limousin and possibly secondarily with Auvergnat.

There is difficult intelligibility on the border of Guyennais and Gascon. Quercynois, Rouergat and Carladezien are not intelligible with Languedocien.

Guyennais is very similar to the South Limousin spoken in Brive and South Auvergnat. Specific intelligibility data between Guyennais and Limousin and Auvergnat in general is not available.

There is a strong tendency to want to split this off as a separate high level language within Occitan, but there’s no legitimacy to do so yet based on the available intelligibility information.

Haut Quercinois, Bas Quercinois, Rouergat, Carladézien, Bergeracois, Agenais, Gevaudan, and Aurillacois are dialects of Guyennais.

Quercynois (Carcinòl) is spoken in the Quercy in Midi-Pyrenees. Rouergat is spoken around the city of Rouerge. Carladézien is spoken in Auvergne and is still doing very well. It is transitional to Auvergnat. Bergeracois is spoken around Bergerac. Agenais is spoken in Agen and has Gascon influences. Gévaudan is spoken in the southern part of Lozère, and Aurillacois is spoken in the Aurillac. Both have Auvergnat influences.

North Limousin, spoken north of Correze in Haut Vienne to the Marche and over to Nontron in the west, is a separate language. North Limousin speakers only have 60% intelligibility of Ussel South Limousin. Confolentais, a dialect of North Limousin, is a very peculiar Limousin dialect spoken in Confolens in Saintongeais.

Millevaches is spoken on the Millevaches Plateau south of Limoges. Lemojaud is spoken in Limoges.

Monts de Blond Limousin is a North Limousin lect said to be so different from all other Limousin types that it must be a separate language. It is spoken in the Haut Vienne in the Monts de Blond region around Blond between Nantiat and Confolens near the Charente border. Heavy influence from Charentais langue d’oil and Creusois.

Nontronnais is a North Limousin dialect that is so unusual that it must be a separate language. It is spoken in the North Perigord region around the town of Nontron near the Saintonge border. It has heavy Saintongeais langue d’oil influences.

Creusois (Marchois) is a language spoken in La Marche or the Croissant in north Limousin and over into Auvergne. It extends roughly from La Rochefoucauld in Charente to Saint-Priest-Laprugne just over the Auvergne border in Loire in the south and from Bellac in Limousin over to Montlucon and Moulins in Auvergne to the north. The eastern portion in Auvergnat underwent much more extreme changes than the western portion.

Borders on and is influenced by the oil languages Berrichon and Bourbonnais in the north and east and Poitou and Charentais in the west but is intelligible with none of them. In the northeast, there is a 50 mile wide Creusois zone between Limousin and Berrichon. Some say it is a langue d’oil with heavy Occitan influence, but a better analysis is of a langue d’oc with heavy oil influence. To the southeast around Vichy, there is some Arpitan influence.

This language is still widely spoken in places. 15 years ago, the dialect of Saint-Priest-la-Feuille in northern Limousin was still spoken by everyone over 40. A bit to the west, 15 years ago, Gartempaud, spoken in the village Gartempe, was still spoken by most residents over the age of 50.

Dialects include Western Creusois, Eastern Creusois, Central Creusois and Montluçonnais.

Montuluconnais is spoken around the town of Montlucon in Auvergnat.

It is often thought to be a part of Limousin, but Creusois speakers have a hard time understanding Limousin. Auvergnat speakers cannot understand Creusois. Poor intelligibility with Berrichon, a langue d’oil.

This is basically an Occitan-Oil transitional dialect with an emphasis on Occitan.

Map of the Bourbonnais region in north Auvergne and southeast Berry showing Bourbonnais oil, Auvergnat and Forez Arpitan.

Auvergnat is a North Occitan macrolanguage that has 1.35-1.5 million speakers. Auvergnat is spoken in an area covering the departments of Cantal (except the Aurillac region), Haute-Loire, and Puy-de-Dôme, and extending to the Gannat region in Allier, the Saint-Bonnet-le-Château region in Loire, and the western border areas in Ardèche in Rhones-Alps Province.

In Auvergne, reports indicate that nearly everyone over age 35 can speak Occitan, and perhaps 50% of those age 15-35 can at least understand the language. 49% of the population supports bilingual signage.

A neo-language called Aleppo (Literary and Pedagogical Auvergnat) has been created. It is used to teach students who come from a variety of educational backgrounds and by writers who wish to enrich their prose by using loans from other dialects.

Every village has its own dialect, and there is often problematic intelligibility even from one village to the next.

People who learn standardized Occitan fairly well are completely lost listening to Auvergnat. Auvergnat in general cannot understand Limousin, with the exception of the dialect spoken in Corrèze. The reason is that the phonetics, inflections and vocabulary of Limousin are completely different than in Auvergnat.

Auvergnat speakers are completely lost with the Languedocien speech of Toulouse and Carcassone. Auvergnat speakers cannot understand Creusois. Auvergnat is utterly unintelligible to Gascon speakers.

Auvergnat speakers cannot understand the Provencal spoken in Montpellier, and there is marginal intelligibility with Nimes and Sète. The Languedocien influence on these Provencal dialects is what makes them hard to understand for Auvergnat speakers. People with one parent who spoke South Auvergnat and another who spoke Provencal were not taught Occitan because the lects were too different, implying that even South Auvergnat has poor intelligibility with Provencal.

Auvergnat is closer to French than the rest of Occitan, and it has the strongest Arpitan influences of any Occitan language.

There area two major splits – South Auvergnat or Upper Auvergnat south of the region and North Auvergnat or Lower Auvergnat in the north of the region, which are separate languages. The names upper and lower do not correspond with north and south here, which is curious.

South Auvergnat is spoken from Mauriac in the west through Brioude in the center to Crappone sur Arzon south to the border of Auvergne. It has difficult intelligibility with the North Auvergnat spoken in Allier and Puy de Dôme. South Auvergnat is still in good shape, with 67% of the residents of Cantal speaking the language. South Auvergnat is quite similar to Sarlat Guyennais and the South Limousin spoken around Brive.

Dialects include Brivadois, Mauriacois, Yssingelais, and Sanfloran.

Brivadois is spoken around Brioude and Sanfloran around Saint Flour. Brivadois cannot understand the North Auvergnat spoken in Allier and Puy de Dome. It is in between North and South Auvergnat, but is best characterized as South Auvergnat.

Mauriacois is spoken in the southwest in Mauriac, but it is very different from Aurillacois. It has some old influences from San Floran and Gevaudan. Yssingelais is spoken in Yssingeaux in far southeast Auvergne. It has strong Arpitan and Alpine Provencal influences. Some have classed this as an Alpine Provencal dialect, but this seems uncertain. Intelligibility data is lacking. San Floran is spoken in St. Flour. This is a very influential dialect, having influenced many nearby dialects.

North Auvergnat is a macrolanguage spoken in Allier and Puy de Dome. It is close to the langues d’oil, especially Bourbonnais, but is probably not intelligible with them. North Auvergnat is not doing well.

Speakers of Brivadois, a South Auvergnat dialect transitional to North Auvergnat, have a hard time understanding the North Auvergnat of Allier and Puy de Dome, so it is separate from South Auvergnat. North Auvergnat, especially in the east, is possibly the most divergent lect in Occitan after of Gascon due to very heavy Bourbonnais and Arpitan influence. Some even think it is outside of Occitan proper altogether.

North Auvergnat can be divided into two separate languages – Northwest Auvergnat and Northeast Auvergnat. The differences are so dramatic that they must be separate languages.

Northeast Auvergnat is spoken in the eastern part of the North Auvergnat from Jumeaux and Arlanc north to the west bank of the Allier River near Vichy and Cusset. From Vichy-Cusset to the Loire border, Forez Arpitan was spoken. North of Vichy-Cusset to the Champagne-Ardennes border, langue d’oil Bourbonnais was spoken. Northeast Auvergnat has very heavy Arpitan influences that make it so different from Northwest Auvergnat that it must be a separate language. In fact, Livradois speakers cannot understand Besse-en-Chandesse speakers.

Livradois is a Northeast Auvergnat dialect spoken on the broad Lemange plain in the east-central part of Auvergne bordering on Loire. In the southern part of Livradois around St. Antheme, there are strong Forez Arpitan influences.

Northwest Auvergnat is spoken from about Champes sur Tarentaine to Lempdes north to Pionsat and Gannat. The heavy Arpitan influence on Northeast Auvergnat makes it so different that it must be separate from Northwest Auvergnat. And it is true that Besse-en-Chandesse Northwest Auvergnat speakers cannot understand Livradois Northeast Auvergnat speakers.

Alpine Provencal (Vivaro-Alpine) is a macrolanguage, part of the Provencal macrolanguage, and is often considered to be a separate branch of Northern Occitan.

Overview of Alpine Provencal here. In France, Alpine Provençal is spoken by perhaps over 100,000 speakers, but most of them are middle-aged or elderly.

Maritime and Rhodanian Provencal speakers cannot understand Vivaroalpine, so it is a separate entity.

Dauphine Provencal (Vivaro-Dauphine) is a separate language within Alpine Provencal. It is spoken in the departments of Ardèche (except the north and the western border areas), Drôme (except the north) and the southernmost parts of Isère.

Dialects include Ardechois (Mid Vivarois), spoken in the center of the Ardeche and Dauphinois or Drômois, spoken in the Drôme River area.

Gerbier de Jonc is an Ardechois dialect spoken in the Ardeche region of that name. It differs greatly from the north to the south, with words changing from village to village.

Other dialects are Albenassien, Albonnais, Annonéen, Southeast Ardèchois, Boutierot, Northeast Drômois, Southeast Drômois, Montilien, Privadois, Valentinois, and Vernoux-Doux.

Privadois is spoken in Privas in the Ardeche. Montilien is spoken in Montelimier in the Drome. Albonnais, spoken in the village of Albon in the commune of St. Pierreville in the central Ardeche, was still widely used in everyday life in the town as of 15 years ago.

In areas east of Haute-Loire in the Southern Auvergnat region, Dauphine Provencal resembles South Auvergnat. Apparently not intelligible with Rhodanian or Maritime Provencal.

Gavot Provencal is a divergent Northern Occitan language within Alpine Provencal in France. There are intelligibility problems between this and the Dauphine Provencal spoken in the Drome and the Ardeche such that the others say that Gavot is a separate language.

This language is spoken in an area bounded by Gap – Embrun – St. Paul on the north, Sistern – Digne – Anot – Nice on the west, Nice to Menton on the south and  Menton – Roya Valley – Italian border to St. Paul on the east. Gavot is an eastern group of Vivaro-Alpine spoken in the French Occitan Alps. Speakers of Maritime and Rhodanian Provencal say they cannot understand Gavot.

Apparently all Gavot dialects, while differing from village to village in vocabulary, morphology, verbal conjugation and phonetics, are mutually intelligible.

Dialects include Molliérois, Embrunais, and Seynois. Molliérois is a dialect of Gavot spoken north of St. Martin Vesubie and Beaui near the Italian border. It differs significantly from the dialects of St. Martin Vesubie and Isola very close by. Embrunais is spoken in Embrun. Embrunais has problematic intelligibility with the Transalpin Provencal spoken in Briançon. Seynois is spoken in the town of Seyne and in the surrounding towns of Auzet Barle, Montclar, Selonnet, and Le Vernet.

Transalpin Provencal is a Northern Occitan language, the Italian group of the eastern section of Alpine Provencal, spoken in the Piedmontese Valleys in the Alps along the northwestern Italian border with France and just over the border with France in the Briançon region. There are about 100,000 speakers in Italy, about 50% of the population. The language is in much worse shape in France, where it is near extinction. It is best seen as a Gavot-Piedmontese transitional language.

It is spoken in 14 Piedmontese valleys in the Alps (in the provinces of Cuneo and Torino) and in one community (Olivetta San Michele) and a few hamlets in the Liguria region (in the province of Imperia).

A lot of parents in this region still pass Transalpin Provencal on to their children, but the language is declining, being replaced with Piedmontese or Italian. It is spoken in the highest valleys only, having been replaced in the lowest valleys first and then the middle valleys. The highest valleys often lack schools, courts, post offices, etc. The people live in homes that often lack heating and bathrooms and sometimes lack electricity.

Of the young people under age 20, 40-50% of them speak the language. There is an increase in the number of cases where two Occitan speaking parents speak Italian only to their children. Of the population of 180,000, about 50,000 are Occitan 1st language speakers.

In Italy, it is spoken in the upper valleys of Piedmont (Val Maira, Val Varacho, Val d’Esturo, Entraigas, Limoun, Vinai, Pignerol, and Sestriero) by speakers of all ages, but younger people are reportedly shifting to Italian. Nevertheless, there are reports that the number of speakers of this language has actually risen in recent years, and it is now recognized as an official language by the state of Italy.

In the Estura Valley, Piedmontese (with heavy Transalpin Provencal influence) is spoken in the lower valley from Demonde up the valley to Aisone, and Transalpin Provencal is spoken from Aisone to the top of the valley. In this  area, 100% of the population speaks either Transalpin Provencal or Piedmontese or both. It is only down by Cuneo that you start running into a lot of Italian speakers.

Transalpin Provencal is not intelligible outside of the region.

Escarton is a dialect of Transalpin Provencal that is spoken near the town of Briançon on the border of France and Italy where the Gavot Provencal, Piedmontese and Savoyard Arpitan languages all come together. All three languages influence this dialect, especially Savoyard, but at base it remains an Occitan dialect. Spoken in the Cottian Alps.

There are many different dialects included under the Escarton rubric. Briançon dialects include Viaran and Montegenevre. Escarton also includes the Queyras spoken around Abries and Aigilles in France to the southeast. In Italy, it includes Oulx in Oulx, Bardonecchia in Val Susa and Val Chisone in the town of Sestriere in Val Chisone.

Escarton has difficult intelligibility with the rest of Occitan. It has better intelligibility with the Transalpin Provencal across the border in Italy than with the Embrunais Gavot of the lower valley in France.

Gardiol is a diaspora Alpine Provencal language spoken in Guardia Piedmontese, an Occitan-speaking town in southern Italy. The town, located in the Cantabria region in Cosenza Province, was established in the 1300’s by people from the Waldensian or Vaudois Protestant movements who were fleeing Catholic religious persecution. They were thought to be heretics and were massacred in the 1300’s.

The language is a Vivaroalpenc dialect formerly spoken in Briançon and in the Varaita and Pellice valleys of France. It is still taught from K-12, and has 340 speakers. Gardiol is under strong southern Italian influence. Gardiol is said to be incomprehensible to French Occitan speakers due to the fact that it has been diverging for over 700 years in isolation in Italy.

The outlandish costumes of the women of Guardia Piedmontese, Italy, based on clothing from the 1300's in southeastern France.

The outlandish costumes of the women of Guardia Piedmontese, Italy, based on clothing from the 1300’s in southeastern France.

There are more speakers in Germany’s Württemberg, in the US (especially in North Carolina in the town of Valdese), in the Argentinian town of Pigüé, and in Canada’s province of Quebec. Intelligibility of these diaspora lects with the language in Italy is not known.

References

Bec, Pierre. 1982. “Occitan”, Chapter in Posner, Rebecca and Green, John N. Trends in Romance Linguistics and Philology, Volume 3. La Hague, Paris, New York: Mouton.

This research takes a lot of time, and I do not get paid anything for it. If you think this website is valuable to you, please consider a a contribution to support more of this valuable research.

33 Comments

Filed under Dialectology, Europe, France, Indo-European, Italy, Language Classification, Linguistics, Regional, Romance, Spain

33 responses to “A Reclassification of the Occitan Language

  1. Sleep

    You seem to be on the extreme end of what linguistics people call “splitters”. I would say that there is one German language, one Occitan language, one English language, and so on. Chinese is a more difficult question because of its unique writing system, but I hear that when they teach Mandarin to country bumpkins who speak only their local Chinese “dialect”, they pretty much have to start at square one, so I could agree that there are probably a lot of Chinese “languages”. Curiously this problem doesn’t seem to exist in Japan even though Japan has been evolving for 2000 years.

    • Actually that is not the case about Japan, but you are right that there is not a lot of linguistic diversity. For one things, there are about 6 Okinawan languages that are not intelligible with Japanese. I am not sure if you are counting those. And I think there are some Japanese dialects that could be split off, but it’s such a radical thing to do that I don’t want to get into it right now.

      It’s a good intellectual discussion why mainland Japan ended up with only one language after 2000 yrs.

      It’s true that I am a splitter, however, Ethnologue, supposedly radical splitters, has been given the task of handing out ISO codes now. ISO codes is how the scientific community determines what’s a language and what is not. So science has given the language definition task to a bunch of splitters.

      In my classifications, I am following Ethnologue’s methods of splitting, so I’m not really going beyond them. Their method is mutual intelligibility and they seem to set it at around 90%. Below 90% comprehension, I split dialect to language.

      There is not much opposition in the linguistic community to the splitters, as most linguists support the splitters or really don’t care one way or the other.

      The whole opposition to splitting comes from either amateur linguists not at a university (cranks) and nationalists, usually speakers of the languages that are getting split up. The nationalists say that the splits are just dialects of their great language. I’m not sure what motivates these people, but it seems to be some sort of tribalism or nationalism. In part they are afraid of sucessionism but that is not all of it.

      It might be interesting to investigate why German and Italian speakers are so furious about saying their precious dialects are really languages.

    • mumbaki

      Actually,it is very easy to say why there are only few divisions,idioms or languages in japanese,it is because the japanese were concentrated to one area in the first millenium of their development as a people,the northern parts were inhabited by ainus,Japanese expanded to north few centuries before the dialect that became the ancestor of modern day japanese dialects existed,in short if the ainus were easy to assimilate,japanese could be now composed of more unintelligible idioms.

    • Interesting, thx for that mumbaki.

    • Mumbaki

      I think more study should be done in philippine languages because some philippine languages are different to each other in the same way as Kansai Ben and Tokyo Ben are different from each other.

    • Actually, Mumbaki, I am a strong believer that more than one Japanese language is spoken in Japan proper. Some of the Japanese dialects are very strange and have poor intelligibility with Standard Japanese.

    • Mumbaki

      But there are variants between Kansai Ben and Tokyo Ben.

      I really believe that Everyday Language should be also used in education and media,the linguafranca should also be taught as well….

    • Mumbaki

      Kansai Ben is flourishing actually…

  2. Curiously this problem doesn’t seem to exist in Japan even though Japan has been evolving for 2000 years. – Sleep

    Interesting observation. Off the cuff I’d say it’s because Japan is practically all coast. Japan has had excellent sea access to itself, probably for longer than almost anyone has had roads. Thus, well-spoken upper-class types could sail up to any coastal town and the peasants could emulate their grammar, etc. The ability of the latter to do this so well, too well in the eyes of the aristocracy, may be why so many exchanges in Japanese are coded by status. I.e., it’s something like, “Yes, dirty peasant, you are required to use good grammar like we do in the palace, but you’d better not think that makes you belong in the palace. Thus you must use this laundry-list of words and inflections indicating utmost respect for me.”

    China, with its huge hinterland, had no such advantages. On the other hand, I’ve never heard of any such thing as a “coastal dialect” or language which bound all the Chinese folks with good sea access together … so my theory may be bunk. Perhaps there is a coastal lingua franca that allowed, say, a Mandarin-speaking junk captain to tell his Cantonese-speaking crew to talk to Hakka-Wu dockworkers.

    • There was no coastal lingua franca in China. I don’t think they were even sailing up and down the coast all that much. The written language was de facto lingua franca. When my Dad was in China in 1946, the rickshaw drivers, if they could not understand each other, would get out paper and write down the words in Chinese and communicate that way.

      Even during the Chinese Revolution there were horrible problems with units unable to communicate with other units, leaders unable to communicate with rank and file, etc. That’s why one of the first things the Commies did in 1949 was select Putonghua and shove it down everyone’s throats. The Nationalists had been trying to promote Mandarin but had done a crappy job of it, as they did a crappy job of everything.

      Interestingly, there was a big debate in 1913 or so in the Chinese government about which language to pick as a lingua franca. They chose Mandarin, but Cantonese only lost by one vote!

      To this day, the South is more resistant to Mandarin than anywhere else. They’re almost actively hostile to it. Down around Hong Kong, everything is in Cantonese, even the TV. They do speak Mandarin, but they speak it so differently that they may as well be speaking Cantonese. Cantonese Mandarin is not really intelligible outside the region.

      • WFS

        “To this day, the South is more resistant to Mandarin than anywhere else. They’re almost actively hostile to it. ”

        1/Only the South might (i said might) properly “resist” to Mandarin. How would actually “the North” (as opposed to “the South”) which already speaks dialects of Mandarin (roughly standard Mandarin with regional features and colloquialisms) resist to Mandarin itself.

        2/Only Cantonese (Yue) almost has active hostility, because it has a standardized form AND a legal status in HK. so that it can thrive under PRC Mandarin policy.

        Otherwise a 90% mutual intelligibility among Chinese languages is utopia. Even cities close like Shanghai and Suzhou (50km) have different speeches, and even though they belong to one only SuHuJia subgroup. (I’m native of shanghainese)

  3. Jean-Christophe

    Then, if occitan is as many, how many French languages would you consider ?

  4. Hi Jean-Christophe, I am working on French right now and will be done with it soon. French is quite a few languages, even in France. Standard French is just Parisien, one of the langues d’oil that was chosen to be the standard. I have 32 French languages so far. I think Ethnologue has only 4. I have not yet done the creoles – this does not count the creoles.

  5. So what you’re telling me, Lindsay, is that if I am to learn a dialect of Chinese I should flip a coin??

    Naw, just kidding, but I wouldn’t really know which one. Once I decided I’d figure out which language HK, RoC, and Singapore Sinophones had in common, figuring two out of three would be okay. No can do – they are all different! Probably Mandarin classes are easiest to find.

    • No, you need to learn Putonghua. That Standard Mandarin. There is no point to learning any of the other languages unless you are going to work over there. Putonghua is the one that all the foreigners are learning, and most Chinese speak it anyway. Even in Taiwan and Singapore they speak it. They speak it in Hong Kong too, but they prefer Cantonese. Putonghua is hard enough!

  6. Hello Robert,

    I’m quite puzzled by your classification. What are the criteria for intelligibility? My mother from the Gascon area understands my father’s family (from Languedocian area). As a postman, my father worked some time in Limousin area and was very pleased to spend some time on the fairs and discuss about the local cattle virtues with limousin-speaking locals. I even remember him discussing at Puigcerdà with a catalan speaker, each using his own language.
    Personally, I’ve had opportunities to discuss with Provençals and even with people from the occitan-speaking Piedmont without any trouble.

    Last but least, let me quote a mail from the Ethnologue staff I got last January:
    “Dear Mr. Blanc,

    Thank you for pointing out the information on The Ethnologue website concerning Occitan. Earlier this year, we updated our database to unify all the Occitan dialects, which were previously coded as separate languages.

    The current website still reflects the 15th Edition of our book, which came out in 2005. We are now preparing to publish the 16th edition in February or March, at which time the website will be updated with the new information. In the 16th edition, Auvergnat, Gascon, Langedocien, Limousin, and Provençal will appear as dialects of Occitan.

    We appreciate your use of and interest in The Ethnologue. Thank you.

    Chuck Fennig
    Managing Editor
    http://www.ethnologue.com

    So my question again: what are your sources for mutual intelligibility?

    Best regards,

    JF Blanc

    • The sources are Ethnologue.

      However, it is interesting that Ethnologue notes that there are intelligibility problems with some of the dialects.

      My sources indicate that Nissart is not intelligible with the rest of Provencal, and Northern Gavot is said to be hard to understand for the rest of Provencal. Transalpin Provencal is not intelligible outside the region. Since Mentonasque is even further from Provencal than Nissart, it must surely be a separate language.

      I am going to wait until the new Ethnologue publishes to update my classification. According to you, we need to make Catalan and Occitan one language! Catalan and Occitan have about 50% intelligibility, but after a few weeks of close contact, they can start to figure each other out.

      Scientific studies have recently investigated the 6 major Occitan languages that Ethnologue has listed. Those studies show high, but by no means complete, intelligibility among those 6 Occitan languages. I’d prefer to go on science rather than someone saying that “I can understand everyone.” Especially when that person claims that Catalan and Occitan are one language.

      If you could get me a link to any of those studies I would appreciate it.

      There is also a question of “structurally separate languages”. A number of the Occitan languages have been described as “structurally separate languages.” That generally trumps intelligibility.

      Intelligibility in Occitan is very controversial. On the one hand, you have folks saying that they can’t understand the folks in the next village, then you have folks like you saying everyone understands everyone. It’s time for some scientific intelligibility studies to sort all of this out.

      • Foix gras

        Interesting stuff. It’s an old post, but if you don’t mind, what kind of sample sizes are we looking at here? In determining intelligibility values, are the ages of respondents taken into account? Could the younger generation possibly have had earlier exposure to French, and thus a leg up in easily understanding other Romance speakers as compared to the older generation, for instance (even without prior exposure to that particular lect)?

        • Hi, I got intelligibility figures from various websites and from quite a few informants who told me that, say, their parents said that they had a hard time understanding this or that lect or could only understand 50% of it. Actually the intelligibility figures would be much greater for the older generation than for the younger generation.

  7. Hello Robert,
    Staying away of scientific studies for a long time, I can only provide information from my personal experience. I’ve also a position in NGOs trying to keep alive the Occitan language, so you may consider my POV as biaised… Nonetheless, note the information about “structurally separate languages” comes from a very biased source, especifically P. Blanchet – also a strong contender in the not really scientifical controversies on occitan as one vs as many separate languages. The other ethnologue source, Agard, came to other conclusions on subsequent publications on the Romance languages.
    (All of this written with no bookshelf at my hands, sorry for the lack of references)
    Another point if your “It’s time for some scientific intelligibility studies to sort all of this out.” I clearly agree, but it is maybe too late as there is few or none natural monolingual speakers left…

  8. Another point about Catalan and Occitan as a unique language… the “declaration of independence” of Catalan is a manifesto published in 1934.

    A manifesto. Not a scientific study.

  9. Ok, I am having a hard time reconciling this. On the one hand, we have many reports that speakers of Occitan dialects can’t understand each other, even people from one village to the next claim that they cannot understand each other. Then OTOH there are folks like you who say they can understand all the way from Aranese to Transalpin Provencal. There is something really wrong with this picture. I think that going back in time a while, it’s clear that many Occitan speakers could not understand each other. See this reference from 1978 that states:

    “…the fact that ‘Occitan’ comprises speakers of different varieties who cannot understand each other, is an instance of lack of a modern standard variety…”

    Book Review. F. Abel, Le Mouvement Occitaniste Contemporain. Tubingen: Gunter Narr, 1973. Pp. 105.. Language in Society (1978), 7 : 289-290 Cambridge University Press.

    We don’t need monolinguals to test intelligiblity. We just record samples of Limousin and play them back to Provencal, etc. The studies have already been done and found high, but by no means full, intelligibility, but I can’t find the studies. In the real world, that often means something like 80-90% intelligibility. We have reports that people were not taught Occitan by their parents because one spoke Provencal and the other South Auvergnat.

    We have reports of Provencal speakers saying that they can’t understand the folks over the mountains only 12 miles away. We have lots of reports saying North Gavot is hard to understand. How to reconcile all of this with reports that everyone understands everyone? There is a real disconnect here.

    For the moment, I am just following Ethnologue. Keep in mind: I could have split Bearnais, Aranese, all sorts of little dialects, but I could find no basis to do so.

  10. Jean-Christophe

    Thank you for your work.

    However, I wonder how far intelligibility is liable to delimit a language.
    Which part plays “variation” in your study ?

    As for occitan, being myself a speaker in the limousin dialect, I have noticed many paradoxes.
    The same native speaker can first say the “patois” in the closest village is different from his, though fully understanding it, and next say Catalan or Provencsal speakers were easy to understand during WW2″.

    Many times, I have come across limousin speakers who easily understand faraway occitan varieties and hardly understand some other limousin varieties (for instance provençal can be more accessible to a limousin speaker than a variety of bas-limousin) spoken in the area of the Monédières hills.
    Moreover, you can be sure Occitan in Italy (at the least in the North Valleys) is largely accessible for a limousin speaker.
    I agree most often a while is needed between two native speakers to get used to each other’s way of speaking.
    How would they have communicate before the coming of standard forms (as French in France…) ?
    Even during 19th, most occitan speakers did not speak French however understanding it.

    In my opinion, the question will never definitely be sorted out if occitan is one language or several. The same problems arise for many other languages around the world.
    Therefore, I shall not really take part in it as the question appears to me as irrelevant the sake of the language, though interesting linguistically but unsolvable.
    As far as I am concerned, I acknowledge occitan has some unity and believe one shall insist rather on similitudes than differences, though admitting them.

    Your criteria is intelligibility. Mine is likely to be morphology.
    Your study is nonetheless useful.

    To conclude, I may just ask a few more questions:
    How many tests have you performed yet to delimit occitan languages ?
    Are your results really reliable ?
    Finally, how would you define “a language” ?

    • Thank you. I have not conducted any studies to determine intelligibility.

      Your comment that speakers of one variety say they can’t understand speakers of another variety but then then they can understand another Occitan language is typical. The Occitan dialects or languages are quite strange indeed.

      Reportedly, intelligibility studies were done at one point for the major Occitan varieties and it was concluded that there was “high but by no means full” intelligibility them. In practice, this often means around 80% intelligibility.

      For a language, I want <90% intelligibility. For a dialect, I want 90%+ intelligibility.

  11. Pingback: A Reworking of German Language Classification, Part 1: Low German « Robert Lindsay

  12. Pingback: A Reworking of German Language Classification « Robert Lindsay

  13. Don Sayenga

    Robert, I could use some coaching with Occitan.
    I’m transcribing a travel journal kept by a young American engineer who went looking for a canal reservoir at Lampy near the Black Mountain just outside of Toulouse in March 1831. He became desperately lost in the fog. Describing this, he wrote in the journal:
    “imagine my joy at beholding at a short distance before me a girl pulling a species of herb which grows upon the tops of these mountains; and also my disappointment when I found she could not speak a word of French, or any written language. Nevertheless she understood the word Lampy and pointed towards it – and accompanied the indications which she gave me with a volley of Patois that made me run from the spot as rapidly as I had approached it. I found next a shepherd, who likewise spoke nothing but the same tongue, yet who asked me if I was in want of food, fire or of the grande route. I understood the mantiga to signify manger the French verb to eat; focs or fume I forget which, at present, to come from the Latin fumus or Italian fuseo; and the granda camina to be the same as the Italian grande camino, and I was not mistaken. I could not but admire the good intentions of the innocent or ignorant shepherd , though I was out of all patience.”

    His handwriting is difficult to decipher.
    Do I seem to have transcribed the Occitan words correctly “mantiga” for example?

    Thanks, Don

  14. Christine

    That map~ which I am glad to have~ is not in Spanish. I am pretty sure “Espainia” is Euskara.

  15. Pingback: Language Death Can Occur Very Rapidly | Robert Lindsay

  16. Tim Weir

    Political, cultural and geographical influences on Aranese/Aranais
    I was in the Val D’Aran earlier this year. When you go there you realize why the valley has retained its own language for so long; although politically part of Spain, the valley is on the French side of the Pyrenees, easily accessible from the Occitan-speaking French side (the river Garonne flows down the valley into Gascony) but much harder to access from the Spanish side. Until a tunnel was built in 1948 from Alta Ribagorca to Vielha the valley was inaccessible from the south during the winter. Similarly French radio and TV is easier to pick up than Spanish or Catalan. Hence the valley has continued to speak a Gascon dialect of Occitan. Franco’s regime attempted to impose Castilian here, as everywhere in Spain; this has helped to drive the push back towards ‘local’ languages since his death in 1975.
    Most people we met spoke 4 languages: Spanish (Castilian), Catalan, Aranese and French. We spoke to them in French (my wife is half-French and I speak the language fluently; the kids and I also speak some Spanish). Even in the tourist office they don’t speak English (which would be their 5th language). Signs in tourist areas are normally in those 4 languages. The locals to whom we spoke – probably because they spoke so many languages – had a very sophisticated understanding of how languages developed. We had a random spontaneous seminar with a lady in a bakery on how Latin forms had changed in various parts of the Mediterranean, of a kind which would be hard to imagine in England!
    As a bizarre aside, the lack of English spoken in the valley meant that, while a written guide to the valley exists in English, it has apparently been translated using Google-translate, probably via Spanish. This method apparently didn’t recognize that ‘Aranese’ is a proper name, so it was translated throughout by ‘spider’ (arraignée in French, araňa in Spanish) eg ‘the distribution of spiders in Vielha is slightly different to in the rest of the valley’.
    We concluded is that it is correct to regard Aranese as a separate language, rather than as a dialect of either Occitan or Catalan. Unique local factors (protection from France’s homogenizing project by being politically Spanish, protection from Spain by the Pyrenees) contributed to an unusually pure preservation. I disagree with the view that it is an ‘outrageously degenerated dialect’. This is certainly a political statement, made my non-native Aranese speakers, for whom ‘Occitan’ is almost a religious project. Of course there are Catalan influences, which are probably growing; much modern vocabulary comes from Spanish or Catalan, but this is normal. The Aranese to whom I spoke were quite clear that their language was not only separate but vigorous and growing, notably because it was now taught in schools (unlike in Franco’s era). I would want to see broad survey evidence that the locals regarded their language as degenerating into Catalan before accepting the truth of that.
    IMO in areas where the special factors which have kept Aranese alive don’t apply it’s much more dubious to talk about separate languages, rather than dialects. As a total neutral on the political issues it may be easier for me to take an objective view than some ‘committed’ writers do; IMO there is very substantial common intelligibility across all the Occitan dialects of South West France, which is why I would call them dialects. There is at least as much intelligibility between these dialects as between for example RP English and Geordie (spoken in and around Newcastle), a classic example of an English dialect. One has to be very careful deriving a view on the dialect versus language issues from individual writers who almost invariably have some political point to make; even deciding whether to call the major Spanish language Spanish or Castilian is in effect taking sides in the Spanish Civil War, which for many people in this area ended only in 1975 with Franco’s death.

  17. daurade

    Tolosenc is a dialect of this language spoken around the city of Toulouse. It has Gascon influences. In the rural areas around Toulouse, almost everyone over 25 understands Tolosenc, and most speak it regularly.

    I’ve travelled all over the rural areas around Toulouse and am a bit dubious about this claim. If you mean by speaking it regularly people use a phrase or two now and again, okay, but the only people I’ve met who can actually construct something other than rote phrases are over 75.

    But everyone knows a few standard phrases, as do I. You can’t live here long without hearing things like “va pla” or “cossi vas?” or “Ques aquo?”

    Just a quibble, your article is very thorough and interesting. Thank you.

  18. Pingback: Everywhere I run there’s another Gariotte - Lot Cycling Holidays Blog

  19. Pingback: Giving it a go | Operation Freedom

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s