Relationship between language features and writing systems

[StrangerCoug]

I was looking at HoskhMatriarch's conscripts thread and thought this question should be asked in a separate thread: Generally speaking, what kind of phonotactics/grammar/other language rules lend themselves better to what kinds of writing systems? (To give you an idea of the kind of answer I'm looking for, I was thinking of the way tri-consonantal roots work in Hebrew and Arabic and how abjads fit well with them.)

Also, conversely, how do I know I've picked a good writing system for my conlang? As a hypothetical example, suppose I like the idea of abugidas. I want it so that my conlang SHOULD be written as one, but in practice the vowel marks are sometimes stripped in some informal contexts, rendering it an abjad (paralleling the concept of txt-speak). What factors should I consider in deciding if that's a good idea?

[masako]

At the risk of sounding silly; almost any writing system can be adapted/used/developed to write almost any language.

As for types, I prefer this list: http://www.ancientscripts.com/ws_types.html and as you can see there is a bit of bleed over from language to writing system. Korean used to be written with Hanja until an alphabet was developed, Hieroglyphs are essentially a logograph-alphabet mixed system used to write a consonantal root-like language. Tibetan is hardly a strict CV language like its sister languages, but is uses an abugida much like they do.

Try not to get too bogged down in the question "What is the ideal WS for X type of language?" Do what feels right and make it work linguistically later, once you've got the foundation.

[Chagen]

At the risk of sounding silly; almost any writing system can be adapted/used/developed to write almost any language.

That doesn't change the fact that the original writing systems were rather decently suited to the languages in question. The Semitic abjads for instance work well with the triroot grammar and structure of the languages they were meant for. Meanwhile, Chinese's extremely analytical grammar meshes well with Hanzi's distinct logograms for each syllable...and was so horrible for Japanese that the latter invented an entire syllabary just to make the system even work at all. People don't just borrow writing systems, they take them and then hammer them into something more comfortable. Even the Greek and Roman alphabets were vastly different from the Phoenician one they derive from.

I can't imagine a language with English's phonology using a native syllabary, for instance. One they borrowed from a more powerful nation and kludged into something useable? Sure. But there's probably a reason why most Syllabaries were made for languages that have very simple phonotactics. But I'd expect a completely native writing system, barring extremely ancient ones, to reasonably accommodate the language it was meant for.

[masako]

People don't just borrow writing systems, they take them and then hammer them into something more comfortable.

You are - in fact - making my point for me.

The Semitic abjads for instance work well with the triroot grammar and structure of the languages they were meant for.

But those abjads developed almost as a necessity after centuries of those languages either not having a dedicated writing system, or using something like Hieratic, derived from phonographs like Hieroglyphs, which, in turn led to the Byblos syllabary.

I can't imagine a language with English's phonology using a native syllabary, for instance.

Then your imagination is hideously limited.

[sangi39]

So many people forget Linear B. It kind of sucks for writing Greek, and for a long time people thought it wasn't used to write Greek precisely because of that fact. Syllables are written with missing codas, plosive MOA is unmarked, and clusters are either simplified or written out using multiple syllabic signs.

However, this could be in part because of the way it was used. Linear B was never really used beyond inventories, list and tax documents, and they were intended to be temporary. The vocabulary was limited and the texts pretty formulaic, plus semantic determiners were used as well. Had Linear B been more widely used, it may well have developed to better represent Greek, but that's at least one example of a language borrowing a script that was relatively ill-suited for the language without modifying it.

The Japanese kana systems took something like four centuries to develop so the "hammering out" of Chinese characters took some time. The same, IIRC, is more or less true of Akkadian borrowing its script from Sumerian.