Blog de l'Arpetani

(EN) Diasystemic spellings

Last update: 2023, November 7th

French version (original)

Arpitan version

Definition

A diasystem is a linguistic continuum, with variation within its lects, but having enough common features to group different lects together. It’s also called Abstand language.

For example, Arpitan is a diasystem: this language has plenty of internal variation, but its lects have many common linguistics features so we can group them together. Same for Catalan-Occitan, Hindustani (Hindi-Urdu), Serbo-Croatian, Dutch-Flemish-Afrikaans-Low Saxon, Arabic, Malay-Indonesian or Galician-Portuguese.

With a diasystemic spelling, we can write a diasystem with the same conventions, regardless the lect. Note that some diasystems doesn’t have one unique common spelling, some of them can even use more than one writing system, such as Hindustani language, which can be written with the Devanagari or the Arabic abjad, for cultural and political reasons.

The principle

The principle is: one diaphoneme = one pronunciation for each lect = one grapheme. The spelling is not phonetic, but it is almost totally deductible. It implies to learn some rules to know how the diaphonemes are pronounced in a specific lect, but when you know them, you can read everything. In many Romance languages, the letter ⟨c⟩ is generally pronounced /k/ before ⟨a⟩ ⟨o⟩ or ⟨u⟩, but it is palatalised before a ⟨e⟩ or a ⟨i⟩ (such as /s/ or /θ/), for etymological and morphological reasons.

A diasystemic spelling can have a large version, written the same way for every lect, and tighted versions corresponding to specifics geolects or sociolects. In Arpitan ORB large, the diaphoneme from the Latin -ELLUM is written ⟨él⟩, but there is also another version ⟨iô⟩ in a tighted spelling for lects such as mine, having a pronunciation closer to /jo/. “Mantellum” in Latin can result “mantél” and “mantiô” all the same.

So, a diasystemic spelling is not totally unified and includes local variations. In some Gascon lects, many initial F become H, this modification is written in the spelling. “Flower” is “flor” in Lengadocian, but “hlor” for Gascon lects having this mutation.

But finally, most graphemes are written the same everywhere. In Arpitan ORB, the ⟨ch⟩ can be pronounced /ʃ/ /θ/ /s/ /t͡ s/ /t͡ ʃ/ /t͡ ̪θ/ or /st/ depending the lect, and it’s not a problem. It’s the same diaphoneme, which evolved differently depending the regions.

A diasystemic spelling can also have phonetic graphemes, which are pronounced the same way for every lect, in order to write recent loans. That’s why I can use the ⟨w⟩ in ORB, for example to write “pengwen”, so the ⟨gw⟩ is pronounced /gw/ everywhere. It couldn’t be written another way, and certainly not with ⟨gou⟩, pronounced /g/ or /v~w/ depending the region.

A diasystemic spelling can be used to be close to other languages’ spellings, such as Arpitan ORB close to Occitan’s classic spelling. It increases the inter-comprehension, shows the common linguistic features, and makes easier the writing of transition lects. On the other hand, some diasystemic spellings are only looking for the internal coherence, sometimes without caring about the common features outside the diasystem, which makes the inter-comprehension more complicated, and I think we should avoid it as much as possible

Examples of diasystemic spellings

The classic spelling of the Occitan is diasystemic. Bearnese, Nissard and Vivaro-aupenc can be written with the classic spelling without problem because their functioning is enough close.

The Arpitan ORB is also diasystemic. From Rouana (Forês) until Pont-Sant-Marten (Aosta Valley), even if the oral inter-comprehension is very difficult because Arpitan language is extremely diverse, it’s totally possible to understand other geolects by written because the ORB spelling shows the common diaphonemes.

Comparison with phonetic spellings

Phonetic spellings use a grapheme (often only one letter but not always) for a sound, often using spelling conventions from another language. For example, for languages dominated by French, it’s easy to see a phonetic spelling if it uses ⟨ou⟩ to write /u/, whereas most languages in the world use ⟨u⟩.

It allows to write one particular lect without many linguistic knowledge, but for people who speak another lect, it’s generally very tricky to understand, except if they know the diaphonemes of the other lect.

Phonetic spellings are symptomatic of the minorisation of a language and shows an asymmetry in the languages. For minorised languages, the written tradition was lost or never existed, so the language is written with the conventions of the language which dominates locally.

Some diasystemic spellings are sometimes criticised for being too close to dominant languages, such as ORB which seems too close to French, whereas the Conflans spelling uses ⟨k⟩ and ⟨w⟩ which are infrequent in Romance languages. So many people reproach ORB to don’t free itself from French. But I personally don't agree, I think it's on the contrary phonetic spellings who don't free themselves from French. ORB shows how the language works, the similarities are overall because Arpitan and French are closely related languages, so it's logical to have similar spellings. This is even more true for Oil languages.

However, texts and resources in phonetic spellings can be interesting to read and to study, they give interesting informations about a local pronunciation, and can help for studying and learning a language.

In Arpitan, phonetic spellings (such as Conflans's spelling) are often used by Arpitan speakers, who never really learn to write in Arpitan. As there isn't any written local tradition, it was frequent to write in local French and translate directly when you should recite it, or to write in a phonetic way.

Pros

Cons

The spelling can be difficult to deduce from the pronunciation, unless if you know the pronunciation in more than one lect, or if you have some etymology knowledge.

When the spelling is not built with the close diasystems, it's difficult to write efficiently the transition lects.

Standardised languages (such as French) can be hard to write with a diasystemic spelling because irregularities can be added gratuitously by normative institutions such as the Académie Française, in order to make the language more complicated or return to a supposed "original purity" which never existed. For example, "arcus" in Latin has become "arc" /aʁk/ in standardised French, this is irregular, because the final -c wasn't pronounced during a long time, such as in "porcus" which became "porc". The pronunciation of the -c of "arc" was added arbitrarly.

A diasystemic spelling is able to write the words following diasystemic rules, but it doesn't work for every word. Mainly:

Political issues

Choosing a spelling more than another is very political. By choosing a diasystemic spelling, we show which unifies the lects of a diasystem, and sometimes it shows also the links with the closely related languages, so it places the language in its linguistic continuum.

Nowadays, we communicate a lot from different regions and by written means (allover thanks to Internet), so I think it's not in our best interest to use very specific spellings, and it's better to unify ourselves to fight against the dominations from the states oppressing our languages.

And for languages belonging to a same family, diasystemic spellings can show what unify them, and also by using them, we can understand ourselves more easily. Savoyard written in Conflans spelling is totally unintelligible for Gascon speakers without any preceding linguistic exposition, but the same Savoyard written in ORB won't be very difficult to understand.

You may think: "why should we use a complicated spelling, which even doesn't write languages as they are pronounced?" But it would omit that a diasystemic spelling inserts itself within the continuity of the previous ways to write the language, so if you like the tradition and the status quo (which is not my case), it's coherent to defend them. And as we communicate many between different regions, it's pretty ineffective to write our micro-lects in very different ways. We have a political interest to converge our ways to write.

Revitalisation

Note: this section was written by a person speaking minorised languages of the French state, so it's quite France-centric, even if it can apply to many other languages of many other states.

You may think a phonetic spelling makes the learning easier, because you don't need to learn how the diaphonemes work. But nobody wants the same thing for the languages no dominated by French. Nobody says that Urdu, Japanese or Kalaallisut should be written with a phonetic spelling with the French's rules, saying that without, French people would struggle to learn these languages. So there is an asymetry between the languages, depending if they're dominated or not, and it disadvantages the minority languages, seen as "not-really-languages".

Each language is written differently, and it's totally normal to learn the new rules of a new language, even if the spellings are close. Why shouldn't be the case for dominated languages by French? Why specifically for these languages, we shouldn't use similar graphemes, because in French they are not pronounced the same? When people say that ⟨au⟩ can't write anything else than /o/ for Romance languages close to French (allover for Oil languages, but not only), and to write /aw/ it's better to write ⟨ao⟩ for example, it confirms the rules of French over other languages, and it doesn't fits to the historical usages which wrote ⟨au⟩. Moreover, ⟨au⟩ exploits an articulatory logic, because it's generally an ⟨al⟩ with a /l/ vocalised in /w/, and historically noted ⟨au⟩. And it evolved differently depending the language: in French it's /o/, but it could result by /ɑw/, /ɔ/ or /ɑː/ elsewhere. Even the Mistralian spelling for the Provençal, which is not diasystemic (without being totally phonetic) and which uses ⟨ou⟩ for /u/, uses ⟨au⟩ for /aw/, and not ⟨ao⟩ nor ⟨aou⟩.

I think don't liberate from French's rules, not only it confirms the French domination, but also it may contribute to kill the dominated languages.

But anyway, it's quite false to say a phonetic spelling makes the learning easier. On the short term, it can be useful for one specific lect, because it's easy to know how the language is pronounced (even if French's phonetic conventions can't distinguish some close phonemes, such as /ʃ/ and /ɕ/).

But on the long term, if you learn with a phonetic spelling, you shoot yourself in the foot. It's way more efficient to learn a diasystemic spelling, showing the internal functioning of the language, and making a bridge towards other close languages, with the pronounciation noted in international phonetic alphabet or audio recordings, the same way for almost every language worldwide (once again, there is no reason it should be different for dominated languages).

I feel like diasystemic spellings are a concept quite difficult to accept, but since you know how does it works, you can't stop to use them. It's true for me, and it's also true for many of my friends.

Little interesting anecdote to finish: one of my friend is re-vitalising the Romance Lothringian, and when she started to develop a diasystemic spelling for her language, her first aim was to show the internal functioning of the language. She was creating a diasystemic spelling without knowing it.

See also