#conlang problems

[ranahan]

i've been pondering how mando'a-as-first-language speakers might tend to parse [θ] and [ð]... we know <x> = <sh> thanks to ryan kaufmann (which i'd say must be canon enough since traviss didn't contradict it) and [f ~ p] and [z ~ s] from Traviss, (and maybe <vh> = [f] at least dialectally a la Vhett -> Fett, but I'd prefer to think that's more a matter of Basic speakers parsing [ɸ] as [f] than [ɸ ~ f] being widespread in mando'a) but I don't think we have any data on how they'd tend to parse [θ] and [ð]. I like the idea of th-fronting where [θ] becomes [ɸ] and [ð] becomes [v ~ ʋ] (Kiwi English is allegedly developing this phenomenon), but since mando'a phonotactics don't seem to allow [ɸ] outside of a syllable-initial position (of course since vhe- words are the only example we have of <vh> the data is a biiit limited, but if we assume it's true) I do wonder how likely mando'ade would be to parse, for example, the [θ] in "cloth" as [ɸ], or if they'd be more likely to parse it as [t] or [s]. Tho I do also like the th-fronting bc I wanna distance <vh> from <f> as much as possible :') Not a cipher! 🗞️

On the other hand I'm a bit charmed by [θ] -> [s] as well, mostly bc I associate it with the Askie pronunciation of ת‎, historically [θ] which has in different regions either remained as such or become [t] (Modern Israeli) or [s] (Askenazi). But then the lack of [z] for [ð] (since [s] and [z] are not readily differentiated among mando'ade) makes it less ideal imo. I mean ofc they could be mismatchy, it's not like [ð] and [θ:] were replaced with other sounds neatly and evenly in Finnish (and there's that variation across dialects, which of course would be a thing among mando'ade as well--) But yeah, [ð] could be, if not [v ~ ʋ], then [d] or even [ɾ] or [l]... for obvious reasons I find the idea of the [ɾ] realisation pretty charming :')

anyway, uh, thoughts?

(yes i am thinking abt this bc loan words mostly)

(also. while generally i prefer [f] -> [p], ~kevhe [or whatever final vowel would be best...] for caf is much nicer than kep, lol...)

I’m afraid your ask is going to be better thought out than my answer!

I did think about the same thing, then thought about the option of ð being loaned as dh in Mando’a, and then got hung up in determining what the heck Mando’a phonology even is, and what do the dh etc. spellings stand for. And then never really got back to loanwords.

But now that I’ve found an answer that satisfies me, I do have a whole bunch of things I’d like to loan to Mando’a actually, because I for one don’t buy that a language of an empire, itself a creole (if you subscribe to that headcanon), does not have many loanwords. You know that joke about English mugging other languages and riffling through their pockets for loose vocabulary? I think it would be more realistic for people to make similar jokes about Mando’a, and how Mandalorians plundered half of the galaxy and half of the dictionaries too. (Yes, this is the hill someone else will die on.)

Anyways, yes I do have thoughts. Sorry this is going to be very long and rambling because these are more of immediate thoughts rather than something I’ve formed a considered opinion on.

Like you, I’ve also noted that vh only occurs in syllable initial position in the Mando’a dictionary. But! There’s a Mandalorian world called Ceravh, which as far as I can tell comes from Traviss. I’d give that even odds of being Mando’a or being (perhaps a Mandalorianised version of) a native name of the colonised world, which could indicate that Mando’a could use syllable-final vh in loanwords at least.

But anyways, another option I have thought about is that a word-final φ could get loaned as v. Even if Mando’a doesn’t allow a syllable final φ, the voiced counterpart v is very common in that position. So e.g. caf > kav. Or heck, how about h? > kah. Idk how sensible h would be, it just popped into my mind as something that could sound similar and is allowed in the coda.

Another option of course is that Mando’a could insert extra vowels into loanwords, like e.g. te reo Māori which basically keeps all of the original sounds and adds vowels until the word conforms to te reo phonotactics. So e.g. caf > kavha or kevhe or whatever the vowels would be. There’s this quote, which has a different context, but be used as a precedent if you wanted? “Pronouncing terminal consonants varies in songs. They often become extra syllables. For examples, tor becomes to-rah and tang becomes tan-gah to maintain rhythm and meter.”

On the other hand, loanwords, especially old ones, could also be subject to Mando’a’s habit of contraction. Which tbh seems as much fanon as canon to me in that canon shows some contraction and elision (especially in compound words), but fanon has observed that pattern and taken it as a licence to go an extra mile. I kind of love that kind of an emergency and it’s what got me interested in Mando’a in the first place. I digress, but I’m trying to say faithfully borrowing every single sound is not necessarily the way I’d go for Mando’a. Maybe borrowing syllables as the closest equivalents would feel better? Idk, maybe I’ll try a couple of different approaches with some test sets of loanwords and see what happens.

I do generally like f > p, but… Idk, the results don’t always compel me? They don’t sound unlike Mando’a, but also not exactly like Mando’a either? I mean kav feels more like Mando’a than kap to me. Same for vhervhek vs perpek (although for obvious reasons, perpek does appeal as a curse). But on the other hand, fan > pan vs van vs vhan? Not much of a difference. And Fi > Pai, Felucia > Pelucya/Pelusha, even Falleen > Palin/Paliin do sound good to me. This is probably because while there are words that end in p (like kotep, taap), they are pretty rare. While words ending in v or beginning with a p are a dime in a dozen.

I agree that Vhett being spelled as Fett is a transliteration from Mando’a to Basic, and not necessarily indicative of how Mando’a L1 speakers would adapt loanwords from Basic to Mando’a.

Re: θ/ð, I’ve previously considered t/dʰ, but I like both s and th-fronting better! Those would be the closest fricatives, wouldn’t they.

Mando’a doesn’t seem to have /z/, which seems to be an allophone of /ts/ rather than /s/, if you go by Traviss’s recordings. Or maybe it’s an allophone of /s/, if you go by some comments. Heck, maybe it depends on the dialect. So perhaps you could go with s/ts, or level them both as s? I guess it would depend on whether z gets loaned as ts or s?

Or go with th-fronting (θ > ɸ, ð > v) in syllable initial positions, and for the coda, either voice the θ > v, or add an extra vowel after it, or allow φ in the coda for loanwords. I think that at least for dialects that realise <v> as [β], th-fronting sounds really close. I like it!

And since some Mando’a dialects seem to have [b ~ v] (or perhaps more likely [v ~ β], or maybe some dialect has /β/ which is heard as /b/ in some others…), after making the rounds in Mando’a dialects, a loan word might even end up with ð > v > b.

Full disclosure, I also like how either would make loanwords a little less immediately recognisable. But on the other hand, loanword phonology is also affected by the speakers’ understanding of the donor language, which might bias e.g. /t/ over /s/. Idk though how much we need to nitpick realism in an artlang though. Enough to ground it, not so much the art goes out.

And then we of course have the loan words Traviss picked from English:

besom /ˈbiː.zəm/ > besom /ˈbɛ.sɔm/

Bergen /ˈbəɹɡn/ > birgaan /bɪɾ.ɡaːn/

TAB /tæb/ > taab /taːb/

hyem /jɛm/> yaim /ʝaɪ̯m/

charva /ˈt͡ʃɑːvə/ > chaav’la /ˈt͡ʃɑːv.la/

chebs /tʃɛbs/ > shebs /ʃɛbs/

scran /skɹan/ > skraan /skɾaːn/

scunner /ˈska.nɐ/ > skanah /ska.nah/

(Geordie) way aye /ˈwaɪ.eɪ/ or (Pompey) weee /wiː/ > wayii /wa.jiː/

dinny /ˈdɪ.nɪ/> dinii /dɪ.niː/

jack /d͡ʒæk/ > jag /d͡ʒaɡ/ ?

(I’m not counting mhi and kando from that linked post, since they—or at least kandosii—are pre-Traviss Mando’a.)

Bit of a guess at the pronunciations because I’m not a Geordie and Traviss doesn’t do IPA, but I’m getting the feeling that the vowel qualities don’t match up that well. I might have loaned Bergen as *bargen or *begen instead. Besom i > ɛ I can excuse as levelling the vowels to the same height, which seems like something Mando’a is prone to. ɛ > aɪ̯ is interesting. But generally, I’m not sure what I’m looking at here? Maybe just artistic licence?

Of course if actual historical loans, we’d have to consider that those words might’ve been loaned from a dialect or at a time when the pronunciation differed from modern/RP pronunciation. Plus I at least like to think that Mando’a itself has many different dialects, so I’d be willing to excuse a degree of artistic licence and inconsistency, especially when it comes to vowels which are more fluid to begin with. Afaik it’s also not weird for natural languages to exhibit even considerable variation in loan word adaptation.

Ch is loaned both as ch and sh. Syllable-final r seems infrequently loaned (birgaan) or not (chaav’la) or loaned as h (skanah)—perhaps this reflects the (non)rhoticity of the dialect from which the loans came from, because Mando’a does have syllable-final r’s. Otherwise the consonants in these words seem unremarkable.

Talking about r though, I think there are some words in Mando’a where n > r in certain environments (or at least that could explain some etymologies). I guess it wouldn’t be too much a stretch to extend that to d and maybe ð. You can hear [kote], [kode] and [koɾe] or the Republic Commando OST. Imo this is simply the effect of singing on pronunciation (I was a trained singer once upon a time, I could go on about this), but Traviss seems to have taken it as t and d being interchangeable, or /kode/ being the archaic pronunciation and /kote/ modern.

Also as an aside, I kinda want to work out the phonetics of Huttese, because clearly Huttese (or trade/pidgin/creole Huttese, whatever they speak as a lingua franca in the Outer Rim) has lots of Basic loanwords, and I think that many Basic loanwords would probably arrive to Mando’a via Huttese rather than directly from Basic.

tldr: You have seem to have thought more about this than me. Do you have more thoughts? I’m not sure how much sense my reply made; it was bit of a brain vomit. I’ll probably go read some things and come back to this later with hopefully more sense, but I wanted to reply now because I’m not sure how busy I’ll be in the near future.

P.s. you wouldn’t happen to remember where you read about <x> = <sh>, [f ~ p] and [s ~ z]? I mean I remember reading some of those same things, but it was at least a year ago and I don’t seem to have saved the source (whoops). I can probably look up the source myself if you remember where it was. Might it have been this forum thread? “So, Xaga might have been "Shaga" (incorporating that "sh" sound the Mando like) and Zuka could've been Suka.”

[ranahan]

Correction:

Every conlanger ever: Always start your conlang with the phoneme inventory…

…and then fuck it up with sound changes.

Hope this helps!

every youtuber ever: Always start your conlang with the phonetic inventory! always start here! you can't do anything without it! let's make one together: slap some vowels here, some consonants there, and you're done!

me: But what if I want my conlang's phonology and phonotactics to be complex, both for phonaesthetic purposes and to accurately reflect the natlangs I'm basing it on? What if this task becomes an insurmountable roadblock that prevents me from making any further progress?

youtuber: sucks for you then ig

[adragonsfriend]

“Sand people isn’t a good name” “tuskens is a worse name”

Lmao they’re all exonyms just invent a conlang real quick so you can make a name for them to call themselves in their own language. duh

[godlymiseryclub]

[kelasian]

okay i might get back on with sentian because i just had an idea to not mark person on verbs at all and only mark mood, aspect, tense and voice (i guess). which isn't anything crazy but i have actually never done that + it should make it way easier to derive morphemes from the protolang now

[ranahan]

Here are a few other options:

5. Give your phonemes some allophones that make pronunciation easier in certain positions/combinations. Maybe your ejectives become tenuis in intervocalic positions: jak’aq-rəkʷ > jakaq-rəkʷ. In other words, you can get rid of the ejectives only in some positions and keep them in others. Maybe your /r/ becomes more of a [ʁ] when following uvular consonants. Etc.

6. Restrict phonotactics and make rules how roots/morphemes combine when they break phonotactic rules. E.g. maybe your language doesn’t allow consonant clusters, so your language inserts a vowel or omits one of the consonants making up the cluster: jak’aq-rəkʷ > jak’a-rəkʷ

7. Give your inflectional morphemes some allomorphs (depending on the phonological features of the base) that simplify pronunciation. A very simple allomorph pair would be -rakʷ after a vowel and -akʷ after a consonant: jak’aq-rəkʷ > jak’aqəkʷ

Or more than one of the above.

thinking aloud: conlang darlings

i'd appreciate any input from any seasoned conlangers

so working on my lang i end up with this word:

jakʼaq-rəkʷ ("forest"-LOC)

and for the life of me, i just cannot pronounce it. there's just to many things i have to do with the back of my tongue in such a short time, and what's more, i'm learning that i hold to long on the medial ejectives so they have the rhythm of a consonant cluster when they shouldn't. the ejective may actually be causing most of the problem (though the coda /q/ is not at all easy for me either).

i'd like to be able to pronounce this lang... i don't need to like, speak it fluently, but i'd like to be able to say the words and it not be too difficult.

it looks like my options are:

leave it the way it is

design roots specifically to be easier to pronounce

come up with complicated ruki rule-style sound changes to make words like these easier

get rid of the ejectives

leaving it the way it is sacrifices my ability to pronounce it, and i do care about that (should i?).

manually changing roots might sacrifice naturalism but maybe not?

i definitely don't want to do the sound changes.

but i really don't want to get rid of the ejective. i even simplified things by only including /kʼ/, but apparently even then i struggle with it in medial position

idk

what do

[beebfreeb]

favorite ghs artist also has a conlang ? my day is made ??

I even have multiple.

[hungwy]

i have a few streaming ideas including playing sudoku, geoguessr, conlanging, and (less creatively) playing apex and cs2. obviously three of these are a better niche than the other two by virtue of fewer people doing them. but i wont lie i am an fps player at heart and obviously would enjoy playing those more in terms of regular streaming content

[pogostick61]

@whatisamildopinion

You ever read an ao3 fic so good you make a co land and then spend 2 weeks working on it until you have enough grammar to write a couple sample sentences? no I could never

Idell u’irteaczire xa’tallokasz due llunu’ikisza wisza, due adżewi’dwole favathilera ar obati’dwole xomana cza’uviszi lir nemiralita, eva’i lanqi’e.

Gableni’e, qo’li ar ule. Opaxaczanbe te orfasigi erera ar szeler. Orfas ope’emira ar żo’ora ar nema’er.

Noczire oge’e ule i qo’li lir ayakasz, lloge veqexwe nemiraliszi kirizesza xa’tallor.

I hope this is even close to what you were imagining, because I know I was always imagining a horrifying mixture of polish and welsh

so I made it

enjoy

[possiblyatransgirl]

Having clear phonotactic rules in your conlang is great and all but they don’t tell you about the cost (having to make phonotactic rules)

[hey-missy-youre-so-fine]

I love how I have to learn an entire dialect of a Tolkien conlang just to properly thirst over Legolas's dad

[nototherwisespecified]

one of the problems with falling in love with dense series about wretched people & their horrible adventures in dark fantasy worlds is that their tumblr fandoms are unbearable. be rabid about the Character with me no not like that

[notnulli]

i make conlangs exclusively so I can yearn in a language only I understand

(also bc its an hyperfixation and I can't focus on anything else bc my ADHD meds don't work)

[lemonthepotato]

Cringe warning: very bad Esperanto.

Mi povas mergi min en Esperanto čar mi ne faris multe da progreso ekde 2019. Mi kredas mi komencis lerni en 2019, sed mi rezignis dum (por?) unua jaro. Mi rezignis la hispana, la franca, la japana kaj la irlandano čar… nu, estas evidente kial. Tro da lingvoj lerni (por lerni? Lerni sentas malgxuste.)

Sed, mi restis kun la Esperanto(n?) čar gxi estas facila. Mi volas lerni lingvon por la sakeo (that’s… is that seriously the word? Sakeo? I joked once that Esperanto is 80% English words with -o and the end and 15% other languages with o- at the end, but I digress) de lerni lingvon. Homoj diras ke tio estas malbono kialo, sed, kial? Estas amuza… ne estas krimo amuziĝi.

Honeste, mi estis (estis for ‘have been?’ doesn’t feel right…) uzi Google Translate por helpi min, sed ne por lambastono. nur por kontroli se mia gramatiko estas bona. Ne estas, evidente, sed… mi estas nesekura pri gramatiko. Mi scias ke gxi ne estas bona, sed gxi estas probable pli bona ol mi sed mi ne uzis gxin.

Cxiuokaze… mi havas punkton kun ĉi tio; estas malfacila mergi en konlang! Jes, mi povus aligxi servilo de Discord, sed… la embaraso. Mi estus kiel, “Bonvolu… mi estas…” kaj havas furzo de cerba! Cerba furzo? (Googling how to stutter in Esperanto. Great.) (also I’m realising I said bonvolu instead of… oh my god? Am I seriously forgetting hello? Oh, Saluton!)

Cxu mi probable lernu la lingvon de miaj lando, la irlandano? Probable, sed honeste? Neniu parolas la irlandano en la nordo. Ili apenaŭ en la sudo. (Ne estas sude, mi ne zorgas se Google translate diras alie… ne sentas gxusta.)

Cxu mi havas punkto kun cxi tio? Ne. Sed, hej, diras al mi kiel CLAPPED mia Esperanto estas. Kaj, jes, mi eĉ ne provis traduki clapped cxar gxi estas pli amuza al ne.

Mi estas tiel malbona pri Esperanto. Mi devas fidi al tradukistoj por helpo. Mi uzas Google Translate por helpi kun tempoj kaj gramatikoj, sed la vortoj estas plejparte el mia cerbo, se tio havas sencon.

Mi ne havas kialon pri ĉi tio. Mi supozas, ke ĉi tio estas testo de miaj kapabloj. La rezultoj? Tre malbona, sed, hej, mi afiŝos ĉi tio, ĉiuokaze.

Edit: after writing this post, I got an easy, actually video about languages recommended… lol

[bakuliwrites]

Entirely unrelated to anything, but I'm working on a conlang and have been for a number of years, now. As a linguist, I know that it would be good to do syntax trees for it but what if I just, like, didn't??? Because I hate them and they hurt my brain and I simply do not want to make syntax trees???

[oviraptoridae]

creating alien species who are colonial organisms with decentralized consciousness/sense of selves is all fun and games until you have to deal with the “english doesn’t have a distinct second person plural pronoun” problem when writing dialogue for them