Zompist Bboard Again

bradrn wrote: ↑Fri Apr 24, 2020 7:07 pmThanks for the excellent explanation Ser! I was aware that it was devised by a native speaker, but I didn’t know the details. The orthography does make a lot more sense now that you’ve explained it.

(Of course, that doesn’t take away from the fact that SENĆOŦEN is still very, very weird by the standards of Latin-script romanizations…)

Oh, and I forgot to mention, the use of that one lowercase <s> for the Saanich possessive suffix -s, meaning '-3' (i.e. 'his/her/its/their') is simply a visual reference to English possessive -'s. I'm kinda surprised he didn't use an apostrophe too.

1SG prefix: /nə-/
1PL suffix: /-ɬtə/
2 prefix: /ʔənˀ-/ [ʔənʔ], sometimes reduced to /n(ˀ)-/
3 suffix: /-s/

Also, he perceived Saanich /a/ as Canadian English /ɑ/, which explains the choice of <O> for /a a:/, as in <SENĆOŦEN> /sənˈtʃaθən/ 'Saanich', <SPOL> /spa:lˀ/ 'raven'.

The Kiowa alphabet was also devised by a native speaker, Parker McKenzie, an amateur linguist (eventually he received an honorary doctorate) who did a bunch of language documentation work, some of it in concert with white linguists, for many decades, and was a really interesting guy in general. (Also: more on McKenzie and the development of the alphabet.)

bradrn wrote: ↑Fri Apr 24, 2020 7:52 am Northern Qiang, which (in)famously uses ⟨v⟩ for /χ/.

The Northern Qiang romanization has many more problems than that. IIRC there's something in the grammar about how speaker uptake is low because it's too confusing. The contrast between <ae> and <ea> probably doesn't help, nor the fact that voiced consonants are written by doubling their unvoiced counterparts except <zh> which 'doubles' to <dh>, the rhotic being <rr> because <r> is used to mark vowel rhoticity, the extremely long words, etc.

The Saanich alphabet is at least a reasonable response to the given constraints. The Northern Qiang one is just poorly designed by any metric. Except similarity to Pinyin.

The Pumi alphabet is I think pretty similar but with tone letters and <zh> doubles to <zzh>.

Nortaneous wrote: ↑Fri Apr 24, 2020 11:45 pm

bradrn wrote: ↑Fri Apr 24, 2020 7:52 am Northern Qiang, which (in)famously uses ⟨v⟩ for /χ/.

The Northern Qiang romanization has many more problems than that. IIRC there's something in the grammar about how speaker uptake is low because it's too confusing. The contrast between <ae> and <ea> probably doesn't help, nor the fact that voiced consonants are written by doubling their unvoiced counterparts except <zh> which 'doubles' to <dh>, the rhotic being <rr> because <r> is used to mark vowel rhoticity, the extremely long words, etc.

The Saanich alphabet is at least a reasonable response to the given constraints. The Northern Qiang one is just poorly designed by any metric. Except similarity to Pinyin.

After going through the Northern Qiang romanization carefully, I’m not convinced it’s quite as confusing as you say. (Although that doesn’t mean it’s anywhere near good enough for real-world usage.) In particular, the doubled letters for voiced consonants make more sense when you realise that there’s a contrast between tenuis, aspirated and voiced stops, which can be tricky to represent orthographically. (But doing it for fricatives as well is definitely weird, and as you say, it’s stupid to have ⟨zh⟩ ‘double’ to ⟨dh⟩.) And, given that it’s spoken in China, I don’t think it’s a bad thing that it’s similar to Pinyin.

Linguoboy wrote: ↑Fri Apr 24, 2020 5:39 pm

Birdlang wrote: ↑Fri Apr 24, 2020 4:45 pmAlso, there are some inconsistencies in the Manx, Irish, and Scottish Gaelic alphabets which I’m not a fan of, mostly because of the multiple spellings for one sound.

That's not something you can fully get away from with an interdialectal orthography. Of course, the logical alternative is an orthography strictly based on a single dialect (such as Ó Cuív's proposal for Irish), but I don't think that's desirable for endangered varieties.

What was Ó Cuív's proposal like?

Id be interested to read more too just out of pure curiosity even though i dont know much about Irish. I found this:

http://www.dinglename.com/articles/article.asp?a=56

where the son of the spelling reformer Brian Ó Cuív, who has some position in the government of Ireland, defends the spelling of his surname against people such as Denis Healy who suggested that he change it back to O Caoimh because V is not an Irish letter. Funnily enough Brian Ó Cuív's wife also had a V in her surname.

anteallach wrote: ↑Sat Apr 25, 2020 2:58 pmWhat was Ó Cuív's proposal like?

You can get some idea by browsing his Irish Made Easy on Google Books. Here's a typical sentence with its CO equivalent:

Ó Cuív: Ní dá wint san às do viàl é. "Not interrupting you."
de Bhaldraithe: Ní á bhaint [/i](sin) as do bhéal é. "Pardon my interrupting you."[*]

Notice how slavishly he follows West Muskerry pronunciation. His dialect has breaking of /eː/ before unpalatalised consonants, thus viàl for /w′eːl/ where we might have expected *véal. San represents /sən/[**], which is the default form of the distal demonstrative in Munster speech. (I actually would've expected /s′ən′/ here because of the preceding slender consonant.) Elsewhere in the corpus you find very West Muskerry-specific verb forms like vèach (CO bheadh) and raying (CO rachainn[***]).

So, essentially (since he advocated adoption of this orthography throughout the Irish-speaking world and not just for West Cork speakers), he's trying to teach everyone to speak the dialect of his Gaeltacht.

[*] Literally the translation is something like "It's not that this/it is being taken from your mouth". So Ó Cuív's translation is more literal whereas de Bhaldraithe's better reflects the pragmatics. (It's a phrase used when you are, in fact, interrupting someone.)
[**] <a> in CO can represent both /a/ and /ə/. In general, though, stress placement is enough to disambiguate because [ə] never occurs in stressed position (primary or otherwise). Ó Cuív introduces use of the grave accent to mark short stressed vowels and the circumflex for long nasalised vowels (e.g. lâv for CO lámh), another Munster-specific feature.
[***] West Cork has voicing of the medial fricative followed by diphthongisation, i.e. /raxaN′/ > /raɣaN′/ > /raiN′/, phonetically [ˈrˠəiŋʲ]. (The realisation of the slender fortis dental nasal as [ŋʲ] is another Cork-specific feature.)

Linguoboy wrote: ↑Fri Apr 24, 2020 10:08 amFrench orthography is more consistent than English, and Irish orthography is far more consistent than either (even when you take into account considerable dialect variation).

This made me have a good laugh, and not because I think you're wrong, but because I think you're right.

But the last of these only makes sense if you understand the phonological structure (which, honestly, is how it should be; orthographies serve speakers, not foreign learners).

It is a funny thing when you read the RAE's present and past official books on orthography though, in which they state that the orthography is partly designed for native speakers and partly for foreign learners.

I do wish they simplified their rules for the acute accent though... Hardly anyone understands the RAE's accent rules, so hardly anyone applies them. Various amusing warts:


Why does [pɾo.ˈi.βo] <prohíbo> 'I forbid' have an acute, when [ek.ˈsi.βo] <exhibo> 'I exhibit' and [i.ˈni.βo] <inhibo> 'I inhibit' don't? Oh, it's because <h> is ignored when it comes to applying the rules for adding the acute accent, so <prohíbo> is really "proíbo" underlyingly (what?).

Why does [ra.ˈiθ] <raíz> 'root' have an acute, when [kons.tɾu.ˈiɾ] <construir> 'to build' doesn't? Oh, it's because /u/+/i/ always counts as a diphthong in spelling, regardless of how a given word may be usually pronounced (what?). Oh, but if the word ends up as monosyllabic underlyingly, you can reflect real pronunciations optionally if you want (what?). So, as [u.ˈi] 'I escaped' ends up underlyingly as monosyllabic "ui", this means both <hui> and <huí> are allowed (even though no one pronounces it as "hui" [wi] or [gwi]). 'I was' is only [fwi], so naturally it is only <fui>, but theoretically *[fu.ˈi] would be <fuí>.

Why does [ˈfaθil ˈmente] <fácilmente> 'easily' have an acute but [aˈmaβle ˈmente] <amablemente> 'politely' doesn't, when both are written as single words? Shouldn't the stress of amable be reflected, the way that of fácil is? Oh, -mente is special in that it is completely ignored underlyingly for the acute accent rules (what?). While we're at it, why are these adverbials written as one word, when even you, RAE people, say they do sound like two? Oh, French influence (what?).

How come [komo] can be <cómo> or <como> in Mira cómo/como corre? Oh, it's <cómo> if it means 'Look how she runs' (how fast, with what technique) but <como> if it means 'Look at her running', because in the latter spelling (<como>) it stands for a simple subordinator conjunction (commonly used with verbs of perception like mirar and oír) and not a subordinator proadverb of manner (what is that again?). Oh, and there's also No hay cómo llegar 'There's no way to get there' vs. No hay como llegar 'There is nothing like getting there ~ Nothing compares to getting there' (getting there is very good).

The adverb and conjunction [a.ˈun] (also [awn] when said faster and at least a bit unstressed) 'even [something], even, still, yet' is written <aún> with an acute "when it means todavía" or "(ni) siquiera, hasta, incluso", and <aun> otherwise. What specific senses of todavía/siquiera/hasta/incluso are we talking about here? Oh, "still" and "yet" and positive "even", so, the temporal sense "it started in the past and surprisingly still goes on" (still), the temporal sense "it hasn't happened but it is expected" (yet), and in a positive manner the intensive sense "to a greater degree" (positive even). Which creates distinctions such as aún mejor 'even better (than before, than you'd think, than otherwise even though it was already good...)' vs. aun mejor '(when talking about a bad thing) even if it was better'. Or aún así 'still like that' vs. aun así 'even so'.


As you can imagine, >99% of native speakers can't bother with all this little crap. So it's not like any learner should care anyway.

Linguoboy wrote: ↑Sat Apr 25, 2020 4:45 pm

anteallach wrote: ↑Sat Apr 25, 2020 2:58 pmWhat was Ó Cuív's proposal like?

You can get some idea by browsing his Irish Made Easy on Google Books. Here's a typical sentence with its CO equivalent:

Ó Cuív: Ní dá wint san às do viàl é. "Not interrupting you."
de Bhaldraithe: Ní á bhaint [/i](sin) as do bhéal é. "Pardon my interrupting you."[*]

Notice how slavishly he follows West Muskerry pronunciation. His dialect has breaking of /eː/ before unpalatalised consonants, thus viàl for /w′eːl/ where we might have expected *véal. San represents /sən/[**], which is the default form of the distal demonstrative in Munster speech. (I actually would've expected /s′ən′/ here because of the preceding slender consonant.) Elsewhere in the corpus you find very West Muskerry-specific verb forms like vèach (CO bheadh) and raying (CO rachainn[***]).

So, essentially (since he advocated adoption of this orthography throughout the Irish-speaking world and not just for West Cork speakers), he's trying to teach everyone to speak the dialect of his Gaeltacht.

[*] Literally the translation is something like "It's not that this/it is being taken from your mouth". So Ó Cuív's translation is more literal whereas de Bhaldraithe's better reflects the pragmatics. (It's a phrase used when you are, in fact, interrupting someone.)
[**] <a> in CO can represent both /a/ and /ə/. In general, though, stress placement is enough to disambiguate because [ə] never occurs in stressed position (primary or otherwise). Ó Cuív introduces use of the grave accent to mark short stressed vowels and the circumflex for long nasalised vowels (e.g. lâv for CO lámh), another Munster-specific feature.
[***] West Cork has voicing of the medial fricative followed by diphthongisation, i.e. /raxaN′/ > /raɣaN′/ > /raiN′/, phonetically [ˈrˠəiŋʲ]. (The realisation of the slender fortis dental nasal as [ŋʲ] is another Cork-specific feature.)

Thanks. I suppose I can see what he was trying to do, but as well as being too dialect specific it just looks too radical and that like a lot of English spelling reform proposals there's a fair amount of fixing what isn't broken.

In the version of Irish Made Easy I found Ó Cuív's first name is spelt Seán, but he's often referred to as Shán. So did he at one point propose an even more radical reform which replaced slender <s> with <sh>, Manx style?

anteallach wrote: ↑Sun Apr 26, 2020 11:28 amIn the version of Irish Made Easy I found Ó Cuív's first name is spelt Seán, but he's often referred to as Shán. So did he at one point propose an even more radical reform which replaced slender <s> with <sh>, Manx style?

The use of <sh> for /s′/ is already found in IME, so I don't really have an explanation for why he changed the spelling of his surname before he changed the spelling of his given name.

I was thinking about the states of the US earlier today, and it occurred to me that the poor conworlder who planned them out must have had some pretty interesting conversations about it…

‘So, this is the country you’re working on, is it?’
‘Yes, it is. I’m calling it the United States of America — USA for short.’
‘Very… ah, imaginative name. I see that you’ve planned out lots of states already.’
‘Yes — I have fifty of them so far. Do you think that’s enough?’
‘It’s probably enough. Gosh, some of them are pretty strangely shaped, aren’t they? Like, why is that one at the top split in half?’
‘I don’t know. I’ll figure something out eventually.’
‘And that bottom-right one has a really funny bit sticking out, doesn’t it?’
‘You like it? I’m calling it a ‘panhandle’.’
‘Good name. Very appropriate.’
‘I like the idea a lot, so I’ve added some more as well.’
‘Are you sure that’s realistic?’
‘Don’t worry — I’ve worked most of them out using the geography as a guide.’
‘That is realistic. But I can’t help but notice there’s two big states right in the middle which are just squares. How’s that geographically motivated?’
‘Um… er… well, I sort of kinda didn’t remember to fill that part in…’
‘Did you seriously forget to work out the geography for that part?’
‘Ah, well… not forget as such, I just was a bit preoccupied… I’ll tell you what, I’ll come back to them later and work the borders out properly, once I have the geography…’
‘I bet you won’t remember to redo the borders. But how did you manage to forget to fill in the geography for a whole piece of a continent?’
‘I was busy making all the names.’
‘Oh, right — I do see the names, now that I look.’
‘I made names for all fifty states — that’s why it took so long.’
‘You seem awfully proud of it.’
‘I am. I even managed to memorise all of them! Listen — [points] Alabama, Alaska, Arizona, Arkansaw—’
‘But that says ‘Arkansas’!’
‘It was named by the French. Spelled ‘Arkansas’, but you say ‘Arkansaw’.’
‘Oh, right. I see you’ve been making it very realistic. So, that state next to it — the one spelled ‘Kansas’ — I say that one ‘Kansaw’, right?’
‘No, that one’s ‘Kansas’.’
‘‽‽‽’

Mind you, now that I think about it, my own country is even worse:

‘What’s your country called?’
‘I’m calling it Australia.’
‘That’s a pretty name.’
‘It is, isn’t it? It just means ‘Southern’.’
‘Very imaginative.’
‘I’ve been naming the states the same way… Western Australia, Northern Territory, South Australia, Australian Capital Territory…’
‘I see. Now, I can’t help but notice that most of your borders here are straight lines.’
‘Well, it’s just desert.’
‘How does that even make sense?’
‘I’ll figure something out.’
‘Sure you will. You do know that the whole ‘desert island’ thing is a cliche?’
‘It’s a bit bigger than an island.’
‘Well, ‘desert planet’ is also a cliche.’
‘Oh, fine. There’s some forests as well.’
‘So this desert… that’s why all the borders are lines?’
‘Yep!’
‘Fine, fine. I see you’ve added three squiggly borders though.’
‘Well, I didn’t want to make all the borders lines.’
‘They seem awfully random, those squiggles…’
‘Well, the one follows a river.’
‘And the other two…?’
‘I just drew something that looks good.’
‘Spare me.’
‘Anyway, it doesn’t matter, because no-one lives near the borders anyway.’
‘…why?’
‘It’s all desert!’
‘But you said there’s forests as well…’
‘People don’t live there because… let me think… there’s bushfires. Yes! Lots of bushfires! And then there’s floods! And then the rivers go dry! That’s such a cool idea — I’ll write that down right now.’
‘That isn’t realistic.’
‘I’ll make it work, trust me… For the bushfires, I’m working on this cool ‘global warming’ project with a couple of other people here.’
‘Forget I asked. Anyway, wouldn’t there be indigenous people living all over?’
‘Actually, I was worrying a bit about that. I don’t want to make hundreds of languages… look at that Papuan project, it’s dragging on for ages, there’s just so many languages to make…’
‘You’re just lazy.’
‘Actually, no, wait — I have an idea! Colonisation! I’ll have a bunch of people from the other side of the world colonise and wipe out everyone else! Then I only have to make a couple of languages, the rest have all gone extinct. And I’ll give them all the same phonology, no need to put too much effort in.’
‘Soooo lazy. But, in this country with, as you say, deserts and bushfires and all the rest, why would anyone want to live there?’
‘Hmmm… I know — prisoners! I’ll find some tiny little country on the other side of the world which has no more space for its prisoners, and they can send them here.’
‘You can’t have an entire country made only out of prisoners.’
‘Why not? It sounds so cool! But if you insist… wait — coal! Coal and lead and whatnot. The rest of them can be miners.’
‘Fine — I’m not arguing any more.’
‘Actually, this is perfect! I can keep everyone in a few big cities because that’s where the prisoners go and that’s where the coal is, and no-one wants to move out. Now I can put pretty much whatever I want in the rest of the continent.’
‘Oh, god…’
‘I know — giant hopping rats! That sounds so cool^†! A lizard with two heads! Worms scribbling on trees! I’m going to have so much fun with this!’
‘…I’ll just go back to talk to the US guy, why don’t I. At least his country was a bit realistic.’

_______
^† A common misconception. Kangaroos aren’t nearly as interesting as people think — they mostly just stand around all day and eat grass. So they’re really just like furry cows which can hop.

(Credit goes to zompist for first doing this.)

lol. Thats awesome.

It does remind me though .... I would like to see a conlang with a deliberately irregular orthography. I suspect people dont tend to do it because they would have trouble reading it themselves, but maybe its just the sort of thing that never occurs to people.

Pabappa wrote: ↑Wed May 06, 2020 11:16 amIt does remind me though .... I would like to see a conlang with a deliberately irregular orthography. I suspect people dont tend to do it because they would have trouble reading it themselves, but maybe its just the sort of thing that never occurs to people.

It has been done various times. Zompist's Modern Hanying has an etymological orthography (he doesn't use it in the article, but does include a paragraph about the main etymologized features), and you can ask Nortaneous about Arve [ˈhɛu.wɪj] sometime. I suspect deep orthographies might be more common on CONLANG-L though: I recall that when I tried to read it regularly ca. 2013, I came across the word "maggelity" for the quality of irregular orthographic depth (as in, to talk about a conlang having a lot of maggelity), apparently a joke about one "Maggel" guy who loved irregular deep orthographies.

Ser wrote: ↑Wed May 06, 2020 3:22 pm

Pabappa wrote: ↑Wed May 06, 2020 11:16 amIt does remind me though .... I would like to see a conlang with a deliberately irregular orthography. I suspect people dont tend to do it because they would have trouble reading it themselves, but maybe its just the sort of thing that never occurs to people.

It has been done various times. Zompist's Modern Hanying has an etymological orthography (he doesn't use it in the article, but does include a paragraph about the main etymologized features), and you can ask Nortaneous about Arve [ˈhɛu.wɪj] sometime. I suspect deep orthographies might be more common on CONLANG-L though: I recall that when I tried to read it regularly ca. 2013, I came across the word "maggelity" for the quality of irregular orthographic depth (as in, to talk about a conlang having a lot of maggelity), apparently a joke about one "Maggel" guy who loved irregular deep orthographies.

Not the guy was named Maggel, his conlang was. Otherwise correct.

Pabappa wrote: ↑Wed May 06, 2020 11:16 am lol. Thats awesome.

Thanks! (I was a bit worried about what people would think about it, but found it too funny not to post.)

It does remind me though .... I would like to see a conlang with a deliberately irregular orthography. I suspect people dont tend to do it because they would have trouble reading it themselves, but maybe its just the sort of thing that never occurs to people.

I’d like to see this as well. As Ser said, it’s been done a few times, but it would be nice to have more of it.

Oh, and that reminds me… I’ve complained about Irish and French and English orthographies a couple of times on here, but Tibetan quite possibly has the worst orthography on the planet. See, the Tibetan abugida was made for Old Tibetan, which was spoken in the 11th century. Since then, Old Tibetan has evolved into about 40 different Tibetic languages — which are all now written using exactly the same orthography. So to read text in any Tibetic language, you essentially have to redo 9 centuries worth of sound changes in your head. And the worst bit of it is that this means that the written orthography often seems to have no relationship to the spoken words — at least French has a little correspondence. So you get words like Tashi spelt ⟨bKra-shis⟩, Kagyu ⟨bKa'brgyud⟩, Drebung ⟨'Bras-spung⟩. It’s not even too hard to find such horrible words. And you have to memorise a different set of sound changes for every single Tibetic language!

Tibetan is always my go-to example whenever anyone complains about English having "the worst" orthography. I trotted it out just last week during a work call. I'm like, "Imagine if we stopped modernising our spelling like four centuries before we actually did. Oh, and if we had added special characters just to write words from Latin even though we didn't pronounce them any different." And that's without even touching on the tonogenesis!

bradrn wrote: ↑Wed May 06, 2020 6:32 pmOh, and that reminds me… I’ve complained about Irish and French and English orthographies a couple of times on here, but Tibetan quite possibly has the worst orthography on the planet. See, the Tibetan abugida was made for Old Tibetan, which was spoken in the 11th century. Since then, Old Tibetan has evolved into about 40 different Tibetic languages — which are all now written using exactly the same orthography. So to read text in any Tibetic language, you essentially have to redo 9 centuries worth of sound changes in your head. And the worst bit of it is that this means that the written orthography often seems to have no relationship to the spoken words — at least French has a little correspondence. So you get words like Tashi spelt ⟨bKra-shis⟩, Kagyu ⟨bKa'brgyud⟩, Drebung ⟨'Bras-spung⟩. It’s not even too hard to find such horrible words. And you have to memorise a different set of sound changes for every single Tibetic language!

I'm a bit mystified by your comment that "at least French has a little correspondence". French has a lot of correspondence from writing to sound, more so than English in fact. And there are lots of correspondences in Tibetan by all means, you've probably just read an inadequate intro. In general, you're supposed to think of words in terms of minor syllables + major syllables, and then apply the sound changes to the minor and major together, while separating from other minor-and-major pairs (or just contiguous major syllables).

For example, in compounds like bka'-brgyud, you treat bka' and brgyud separately. I'm not really all that familiar with Tibetan, but as far as Lhasa pronunciation goes, pre-onset b- is generally ignored, and so is a coda glottal stop, so bka' ends up as simply ka-. Same goes for bs- and br-, and IIRC coda -d (Classical Tibetan [t]) has some effect in the tone or something? Either way it's not [t], at least in word-final position. So bka'-brgyud ends up as ka-gyu. A pre-onset glottal stop is also ignored, br- generally ends up as the retroflex [ɖ] which you write dr- (there are a number of Cr- onsets that undergo straight retroflexion), and the rhyme -as as [ɛ] (I think; probably through [ai]), so 'bras- becomes "dre-" [ɖɛ].

Yes, Tibetan is worse than English and French (and Burmese and pre-WWII Japanese...), but there is some sense to a lot of parts of the orthography, at least in synchronic terms going from etymologizing writing to modern sound.

Ser wrote: ↑Wed May 06, 2020 7:19 pm

bradrn wrote: ↑Wed May 06, 2020 6:32 pmOh, and that reminds me… I’ve complained about Irish and French and English orthographies a couple of times on here, but Tibetan quite possibly has the worst orthography on the planet. See, the Tibetan abugida was made for Old Tibetan, which was spoken in the 11th century. Since then, Old Tibetan has evolved into about 40 different Tibetic languages — which are all now written using exactly the same orthography. So to read text in any Tibetic language, you essentially have to redo 9 centuries worth of sound changes in your head. And the worst bit of it is that this means that the written orthography often seems to have no relationship to the spoken words — at least French has a little correspondence. So you get words like Tashi spelt ⟨bKra-shis⟩, Kagyu ⟨bKa'brgyud⟩, Drebung ⟨'Bras-spung⟩. It’s not even too hard to find such horrible words. And you have to memorise a different set of sound changes for every single Tibetic language!

I'm a bit mystified by your comment that "at least French has a little correspondence". French has a lot of correspondence from writing to sound, more so than English in fact.

True, ‘a little’ was inaccurate. (Although admittedly I do find words like ⟨langue⟩ /lɑ̃ɡ/ and ⟨emprunt⟩ /ɑ̃.pʁœ̃/ to be decidedly odd.)

And there are lots of correspondences in Tibetan by all means, you've probably just read an inadequate intro.

In fact, I haven’t read any intro at all — I just wrote about what it looks like to an outsider. As you say in your post, there’s plenty of logic, but it’s very, very hard to see if you aren’t familiar with it.

bradrn wrote: ↑Wed May 06, 2020 7:34 pm(Although admittedly I do find words like ⟨langue⟩ /lɑ̃ɡ/ and ⟨emprunt⟩ /ɑ̃.pʁœ̃/ to be decidedly odd.)

Those two are completely regular and predictable actually, if you're familiar with French orthography.

For weird cases to mention for fun, I'd suggest using fils (/fis/ 'son(s)' in singular and plural, /fil/ 'threads' plural of fil), distiller /distile/ (cf. regular briller /bʁije/), second /səgõ/, or cinq /sæ̃k/ (cf. regular il convainc /kõvæ̃/, blanc /blɒ̃/, but note cinq did have the standard reading /sæ̃/ before a consonant until a century ago or so).

Ser wrote: ↑Wed May 06, 2020 8:00 pm

bradrn wrote: ↑Wed May 06, 2020 7:34 pm(Although admittedly I do find words like ⟨langue⟩ /lɑ̃ɡ/ and ⟨emprunt⟩ /ɑ̃.pʁœ̃/ to be decidedly odd.)

Those two are completely regular and predictable actually, if you're familiar with French orthography.

But I’m not familiar with French orthography, so these look weird to me.

(I think that the vast majority of the posts in this thread are much like this, actually: completely unexplainable from the outside, but they make sense if you’re familiar with the language in question. The fun comes from looking at these languages from the outside and pondering how on Earth that system could ever occur in a natlang… in which case French orthography certainly qualifies.)

For weird cases to mention for fun, I'd suggest using fils (/fis/ 'son(s)' in singular and plural, /fil/ 'threads' plural of fil), distiller /distile/ (cf. regular briller /bʁije/), second /səgõ/, or cinq /sæ̃k/ (cf. regular il convainc /kõvæ̃/, blanc /blɒ̃/, but note cinq did have the standard reading /sæ̃/ before a consonant until a century ago or so).

Oddly enough, most of these look fine to me. I think that’s because my mental model of French orthography is ‘chop off a bunch of letters at the end and switch around all the vowels’.