I am thinking about making an introductory book to some different "languages", for self learning. But I realize I'm blending the writing system with the pronunciation system, and am starting to get confused. To help ease the confusion, I am wondering if technically it's possible to write any language down using any script (so for example, write English using Hebrew Script, Write Sanskrit with Arabic Script, I guess you can't write English with Chinese Script so there's a counter example I just realized, but still I would like to ask to learn more).

Basically I started with the idea for a "Simplified Hebrew Grammar". I was going to start by taking the letters (symbols/orthography) and writing the pronunciations. But then after a few hours of trying that I realized the Hebrew letters sometimes have multiple pronunciations depending on context. Then I think about English, which uses the Latin script, and you pretty much can't say what the pronunciation of a single letter is without resorting to it's surrounding context in a word or something. The letter "a" isn't "ah", it's "cat", "father", etc.

So then I thought about well what if you had books on (1) Scripts and second books on (2) Pronunciation, or writing using a particular script.

But say you had a book on the Latin orthography. Other than how to actually write the letters (imagine kindergarten templates/guides), it doesn't seem there is much to say about them. They represent sounds all over the place, depending on the natural language being spoken, the dialect, etc. What else can be said of a writing system other than just how to literally do the calligraphy?

Anyways, so then it seems like "we're back to a book combining both orthography and pronunciation" again... Like an "English" book, or a "Spanish" book, not a "Latin Orthography" book.

But then I think of writing systems like Devanagari, which seems much more robust and refined. Each letter/shape has a specific sound, which is modified only according to specific rules (for the most part?). In this case, you could write a book just on "Devanagari" and mapping each letter to pronunciations (a short kindergarten book). So this is why I started to ask this question here. In Devanagari (or Sinhala, or other "Southeast Asian" languages), you can write most other languages it seems to me.

My question is, which writing systems can be used to write other languages? Can these writing systems write all other languages or only some? And which writing systems can't write other languages (like Chinese)?

As a tangent, then I'm imagining if there is ever a language where a letter such as b is pronounced /b/ in one context, and /h/ in another, and /t/ in another, just to make things even more complicated. But just a tangent lol.

  • 1
    Why wouldn't you be able to? You'll generally need to adjust the writing system a bit, because if you pick two languages at random it's very unlikely that one's phoneme inventory will be a subset of the other's, but even in a degenerate case you could say "this Chinese character corresponds to the English letter A, and this one corresponds to B, and this one corresponds to C…" and make a mapping that way. (Not a very good mapping, but it would be usable.) – Draconis May 28 '20 at 4:22
  • That's not what I'm asking, I'm not saying just take a symbol set and map it to phonemes arbitrarily. I'm saying if you could "spell it out" sort of how we can romanize almost every language using the Latin letters. Devanagari seems like it could possibly work too, but I don't know, and don't know about others. – Lance Pollard May 28 '20 at 4:28
  • I guess it's kind of a hard question to answer, but thought maybe you guys have thought about this before. – Lance Pollard May 28 '20 at 4:29
  • 3
    What makes one mapping from graphemes to phonemes "good" and another one "arbitrary", though? Yi Pinyin uses the letter Y for /ʑ/ and W for /ɣ/, for example, but it's a system that's in active use. Zulu orthography uses the Latin letters C, Q, and X for click consonants that would have been incomprehensible to Cicero. Why is Q more similar to /ǃ/ than a random Chinese character is to /a/? – Draconis May 28 '20 at 4:32
  • Good points, that helps me understand better what's out there haha, increasing the complexity! :) – Lance Pollard May 28 '20 at 4:34

Sure, because every script has at least two letters and you can devise a spelling system that mimics binary unicode. It would be a terribly inconvenient system, however.

If you want something less binary, it depends in part on what you think is part of "a script". The Latin alphabet doesn't really include the letter ŋ, although it is used in a number of writing systems (e.g. North Saami). It also depends on what you mean to "write down" a language. A number of Slavic languages have a consonant written "š", which is roughly IPA ʃ. But transcribing in IPA is not the same as writing in the language. For your question to make sense, I think you have to mean "is there a conventional transcriptional resource that is the functional equivalent of the IPA, whereby you could transcribe any language to that degree of precision". If you write Czech [ʃ] with "ʃ", you're not writing Czech, you transcribing it.

Putting the two variables together, it seems that really you are asking if there is transcription system that is not Latin-based, and which covers phonetic transcription approximately as well as the IPA. Apart from the IPA, there is/was a Cyrillic transcription system, the Russian Linguistic Alphabet, that covered a lot of sounds. Apart from that, there are major lacunae in Devanagari and Arabic script for transcription, and most other scripts are tailored to the requirements of that specific language, and not general linguistic transcription.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.