Zompist Bboard Again

so X is weird right? we have k and we have s and they're both perfectly workable letters each of which means their respective consonants. but for what are very likely fascinating historical reasons we have this letter x in a lot of languages that means the concatenation of those two particular sounds.

does this kind of thing have a name? european languages can't be the only ones that do this... syllabaries are all multi-phoneme glyphs, alphabets are supposedly phonemic writing, though it's common for glyphs to represent less than a phoneme: i.e. in the english word 'the' each letter represents, on average, 66% of a phoneme.

but surely there exist other such midpoints between a syllabary and an alphabet. abugidas are kind of that, but I'm meaning more like some other alphabetic writing, on... say a random phonology

a e i o u
p t tS k b d g
f s h w j r

and, sure, most of their phonemes are represented by one unique glyph, except not w, w is represented by this set : wa we wi wo oh, and wu? no, we don't write wu, we write it ubu. no, there is no /ubu/ in this language, don't worry. and this one? this one means ka, this other one means tw, this other one is kw, and did I mention we have one for koi ? we really like those fishes. what do you mean I don't have a way to write twat? sure I do, i write it <tuat>. wait, why are you laughing at me. also sometimes /pr/ is written <pr> but in formal writing it has to have its own glyph. this is the ancestral tongue of my people what's funny nooooo stap

It is a little weird. Sampson says that Greek had letters χ ψ because those were the only clusters that could end a syllable.

Greek also had ζ = [zd], according to Allen. There doesn't seem to be any good reason for this one.

<ts> is more common: Polish c, Russian ц, Hebrew צ, pinyin c and z.

Linear B maybe doesn't count because it's syllabary, but besides a large set of CV glyphs it had an inconsistent handful of CCV glyphs, including /twe two dwe dwo pte rjo rja tja nwa>.

Torco wrote: ↑Tue Feb 09, 2021 10:30 pm but surely there exist other such midpoints between a syllabary and an alphabet. abugidas are kind of that, but I'm meaning more like some other alphabetic writing, on... say a random phonology

a e i o u
p t tS k b d g
f s h w j r

and, sure, most of their phonemes are represented by one unique glyph, except not w, w is represented by this set : wa we wi wo oh, and wu? no, we don't write wu, we write it ubu. no, there is no /ubu/ in this language, don't worry. and this one? this one means ka, this other one means tw, this other one is kw, and did I mention we have one for koi ? we really like those fishes. what do you mean I don't have a way to write twat? sure I do, i write it <tuat>. wait, why are you laughing at me. also sometimes /pr/ is written <pr> but in formal writing it has to have its own glyph. this is the ancestral tongue of my people what's funny nooooo stap

You’re talking about the Iberian scripts here, right? Except it’s the other way round: ‘Continuants … were written with distinct letters … but the non-continuants … were written with syllabic glyphs that represented both consonant and vowel together’.

Yeah, totally. It's not like an unattested thing, but I thought it always was kinda marginal. until I saw

bradrn wrote: ↑Tue Feb 09, 2021 11:07 pmYou’re talking about the Iberian scripts here, right? Except it’s the other way round: ‘Continuants … were written with distinct letters … but the non-continuants … were written with syllabic glyphs that represented both consonant and vowel together’.

truly you had what my heart ached for, my man. look at that thing, I don't enjoy featurality that much aesthetically but, functionally? a thing of funky, funky beauty.


does thid have like a name? digraph already means two squiggles one sound... so these would be... hemigrams?

zompist wrote: ↑Tue Feb 09, 2021 10:54 pm <ts> is more common: Polish c, Russian ц, Hebrew צ, pinyin c and z.

But these are or were single phonemes. There's an English rune for /st/, but that often patterns like a single phoneme even today.

As Greek phi and khi replaced digraphs, perhaps it also felt natural for psi and xi to replace digraphs.

A lot of Indian Indic scripts really have a letter for /kṣ/ (or its reflex); often descriptions treat it as just an opaque conjunct.

Torco wrote: ↑Tue Feb 09, 2021 11:43 pm Yeah, totally. It's not like an unattested thing, but I thought it always was kinda marginal …

I always thought this was restricted to Iberian, but from what you say here it sounds like there’s some other scripts which do this — could you give some examples please?

bradrn wrote: ↑Tue Feb 09, 2021 11:07 pmYou’re talking about the Iberian scripts here, right? Except it’s the other way round: ‘Continuants … were written with distinct letters … but the non-continuants … were written with syllabic glyphs that represented both consonant and vowel together’.

truly you had what my heart ached for, my man. look at that thing, I don't enjoy featurality that much aesthetically but, functionally? a thing of funky, funky beauty.

Huh, look at that — it is featural too! Somehow I never even noticed.

does thid have like a name? digraph already means two squiggles one sound... so these would be... hemigrams?

The stops are a syllabary and the continuants are an alphabet; Wikipedia calls the combination a ‘semi-syllabary’.

Richard W wrote: ↑Wed Feb 10, 2021 12:05 am

zompist wrote: ↑Tue Feb 09, 2021 10:54 pm <ts> is more common: Polish c, Russian ц, Hebrew צ, pinyin c and z.

But these are or were single phonemes. There's an English rune for /st/, but that often patterns like a single phoneme even today.

Writing systems can influence what we consider phonemes and what we don't.

I don't have anything to add, except that don't forget the ya, yu, yo stuff in Cyrillic.

Torco wrote: ↑Tue Feb 09, 2021 11:43 pm Yeah, totally. It's not like an unattested thing, but I thought it always was kinda marginal. until I saw

bradrn wrote: ↑Tue Feb 09, 2021 11:07 pmYou’re talking about the Iberian scripts here, right? Except it’s the other way round: ‘Continuants … were written with distinct letters … but the non-continuants … were written with syllabic glyphs that represented both consonant and vowel together’.

truly you had what my heart ached for, my man. look at that thing, I don't enjoy featurality that much aesthetically but, functionally? a thing of funky, funky beauty.


does thid have like a name? digraph already means two squiggles one sound... so these would be... hemigrams?

Ah, Paleohispanic scripts! I don't know about a name for this type of writing system, other than "mixed syllabic and alphabetic" (and I can't think of any examples outside the Paleohispanic family), but the logic is simple: if it can be sustained, it has a letter of its own. Stops can't really, so they don't get their own letters. Compare the Lautiermethode in German primary schools (probably in other countries as well), where fricatives, nasals and liquids (and of course vowels) are just pronounced on their own, while stops get a supporting [ə], e.g. [kə]. This seems to be the logic behind the Paleohispanic scripts, too.

bradrn wrote: ↑Wed Feb 10, 2021 12:10 am

Torco wrote: ↑Tue Feb 09, 2021 11:43 pm Yeah, totally. It's not like an unattested thing, but I thought it always was kinda marginal …

I always thought this was restricted to Iberian, but from what you say here it sounds like there’s some other scripts which do this — could you give some examples please?

oh, nothing off the top of my head besides what's been mentioned already, I'm afraid.

zompist wrote: ↑Tue Feb 09, 2021 10:54 pm It is a little weird. Sampson says that Greek had letters χ ψ because those were the only clusters that could end a syllable.

Greek also had ζ = [zd], according to Allen. There doesn't seem to be any good reason for this one.

ζ can be the reflex of a palatized d. Possibly it occured in more places than other clusters as well?
Or maybe they just assigned it to the closest equivalent of Phoenician [z] they had, which happened to be a fairly common cluster.

As I recall, the [zd] pronunciation is possible but not certain. There's a pretty good chance it was [dz], at least in some positions, historically or in some dialects.
(PIE *dy > zd would be kind of weird.)

Coptic which I've been studying lately has the letter ϯ which stands for ti, a consonant and vowel. It's the only such letter in the alphabet and I've never seen this in any other alphabetic script.

Ares Land wrote: ↑Wed Feb 10, 2021 10:21 am

zompist wrote: ↑Tue Feb 09, 2021 10:54 pm It is a little weird. Sampson says that Greek had letters χ ψ because those were the only clusters that could end a syllable.

Greek also had ζ = [zd], according to Allen. There doesn't seem to be any good reason for this one.

ζ can be the reflex of a palatized d. Possibly it occured in more places than other clusters as well?
Or maybe they just assigned it to the closest equivalent of Phoenician [z] they had, which happened to be a fairly common cluster.

As I recall, the [zd] pronunciation is possible but not certain. There's a pretty good chance it was [dz], at least in some positions, historically or in some dialects.
(PIE *dy > zd would be kind of weird.)

There's already a Z series in Linear B. My guess would be that ζ originally represented a coronal affricate of some sort and that, assuming the [zd] pronunciation is correct, the affricate evolved into that. IIRC there is a similar sound change in Bulgarian.

zompist wrote: ↑Wed Feb 10, 2021 12:21 am

Richard W wrote: ↑Wed Feb 10, 2021 12:05 am

zompist wrote: ↑Tue Feb 09, 2021 10:54 pm <ts> is more common: Polish c, Russian ц, Hebrew צ, pinyin c and z.

But these are or were single phonemes. There's an English rune for /st/, but that often patterns like a single phoneme even today.

Writing systems can influence what we consider phonemes and what we don't.

Yes, but that doesn't apply to the first three examples.

Doesn't the Cherokee "syllabary" follow the same pattern? Most of it is a true syllabary, but there are some symbols used at the ends of syllables that are stand alone consonants.

linguistcat wrote: ↑Wed Feb 10, 2021 2:10 pm Doesn't the Cherokee "syllabary" follow the same pattern? Most of it is a true syllabary, but there are some symbols used at the ends of syllables that are stand alone consonants.

Canadian Aboriginal Syllabics does the same thing. For that matter, so does Japanese.

it's either that or echo vowels, and those aren't everyone's cup of tea. that's what's the weirdest about x (and other such glyphs that represent clusters): it's just a "let's add a touch of funk for funk's sake" on the part of the divine conlanger.

bradrn wrote: ↑Wed Feb 10, 2021 6:51 pm

linguistcat wrote: ↑Wed Feb 10, 2021 2:10 pm Doesn't the Cherokee "syllabary" follow the same pattern? Most of it is a true syllabary, but there are some symbols used at the ends of syllables that are stand alone consonants.

Canadian Aboriginal Syllabics does the same thing. For that matter, so does Japanese.

Though Japanese didn't originally, at least with Hiragana — ん and む descend from Man'yōgana with the "mu" reading, apparently 武 and 无 respectively — though Katakana ン seems to descend from 尓 ("ni", Middle Chinese */ɲie/) — but of course, the development of the final -n characters is in keeping with the evolution of the language; Katakana seem to have been developed more deliberately, though, and possibly after the phonemicisation of terminal -n (I can't get a good timeline on this, unfortunately; perhaps somebody might fill in the gap here?), and Man'yōgana are certainly older than either. On this note, Japanese is arguably developing terminal -k, -t, -s, -sh, -ch, so I should very much like to see how they end up spelled within a few generations (a pity a human lifespan probably won't allow this).

Rounin Ryuuji wrote: ↑Wed Feb 10, 2021 7:28 pm Though Japanese didn't originally, at least with Hiragana — ん and む descend from Man'yōgana with the "mu" reading, apparently 武 and 无 respectively — though Katakana ン seems to descend from 尓 ("ni", Middle Chinese */ɲie/) — but of course, the development of the final -n characters is in keeping with the evolution of the language; Katakana seem to have been developed more deliberately, though, and possibly after the phonemicisation of terminal -n (I can't get a good timeline on this, unfortunately; perhaps somebody might fill in the gap here?), and Man'yōgana are certainly older than either. On this note, Japanese is arguably developing terminal -k, -t, -s, -sh, -ch, so I should very much like to see how they end up spelled within a few generations (a pity a human lifespan probably won't allow this).

Well, how was final -t spelled when Japanese had it?

Wait, what?