Zompist Bboard Again

doctor shark wrote: ↑Mon Feb 26, 2024 4:14 am

More: show

It's Vianden Castle in Luxembourg.

Of course it is! Do you already know when you will return there (to Mettwurstland, not the castle)?

hwhatting wrote: ↑Tue Feb 27, 2024 4:50 am

doctor shark wrote: ↑Mon Feb 26, 2024 4:14 am

More: show

It's Vianden Castle in Luxembourg.

Of course it is! Do you already know when you will return there (to Mettwurstland, not the castle)?

I'm hoping for end of June, but I need to get the autorisation de séjour temporaire first. We're hopeful for that date, but given that I'll be getting EU funding rather than internal/Mettwurstland funding, I have no idea how long it'll take to get to the stage of submitting the AST application to the ministry.

Good luck!

I found a note today that I made at the end of last year, about making the interrogative particle má in Chusole, which was originally conceived as sentence-final only, a mobile one, so I came up with a set of examples:

[Kele] Tioluni bigyke má?
kele Tiolu-ni bi-gy-ke má
2S.FRML Toilu-ALL go-PST-2S.FRML INT
Did you go to Tiolu?

Kele má Tioluni bigyke?
kele má Tiolu-ni bi-gy-ke
2S.FRML INT Tiolu-ALL go-PST-2S.FRML
Was it you who went to Tiolu?

[Kele] Tioluni má bigyke?
kele Tiolu-ni má bi-gy-ke
2S.FRML Tiolu-ALL INT go-PST-2S.FRML
Was it Tiolu you went to?

[Kele] Tioluni holonnyn má bigyke?
kele Tiolu-ni holon-nyn má bi-gy-ke
2S.FRML Tiolu-ALL horse-INST INT go-PST-2S.FRML
Did you go to Tiolu on horseback (lit., by horse)?

(Tiolu is the capital city of Kiarlon, where Chusole is spoken.)

Note that the pronoun is optional (Chusole is usually pro-drop for sentence pronouns), unless the interrogative particle follows the subject, in which case it is mandatory.

It’s not much, but given how little conlanging I actually do, and how glacial my progress on Chusole has been, it still feels like a step in the right direction.

I have determined the basic/ground stem conjugations of Vrkhazhian's verbs as thus: The terminology is a bit hard to decide. The names could alternatively be named as (in order): realis nonfuture, epistemic irrealis future, deontic irrealis future, epistemic irrealis nonfuture, deontic irrealis nonfuture, and jussive.

The epistemics carry a notion of being purely about probability or likelihood without any reference to the first person, this is why it's considered "indefinite". By contrast, the deontics carry the notion that the first person intends (or intended on) on carrying about the proposition, and thus is why it's called "definite". Alternatively the deontic moods could be called "commissives" (indicating promises or threats): "I might take out the trash tommorrow" vs "I shall take out the trash tommorrow"; "You play with fire, you might get burned" vs. "You touch me, I shall break your nose".

The counterfactuals imply a sort of "If I was there, I would have done this, but alas, I was not."

I must say I think it's a pretty unique way of going about things.

nāyīnanni
[nɑː.jɪː.ˈnɑ́n.nɪ]

na-yya-ynad-ni

EPIST-CTRFCT-hunt_fish\IRR-1sg

"I might have caught fish"

tāyīnanni
[tɑː.jɪː.ˈnɑ́n.nɪ]

ta-yya-ynad-ni

DEONT-CTRFCT-hunt_fish\IRR-1sg

"I would have caught fish"

---

I also have some prepositions that also function as clause-level conjunctions:
ku = at/on/in (locative); about, as for (topic)
sa = with, possessing (ornative); in order to (purpose)
ma = without, lacking (privative); becuase of (reason)

[...] m=irrabbattāsan tīya tālan

CONJ=REL-guard\APPL-3PL<NEG> 1PL.OBL city-IN.INSTR.SG

[mɪr.ˌrɑb.bɑt.ˈtɑ̂ː.sɑn tîː.jɑ tɑ̂ː.lɑn]
"[The men under my authority are very unhappy] Becuase of the ones who did not assign us [to protect] the city"

Perhaps an unsurprising thing that I finished up this week.

They must have a shortage of spaces in the country that money comes from!

doctor shark wrote: ↑Fri Apr 05, 2024 7:09 pm Perhaps an unsurprising thing that I finished up this week.

Tis unsurprising that you have produced more great work. Kudos!

Behold, a series of conjunctive proclitics in Vrkhazhian:

śa= "thus, therefore, so, then" ma= "becuase, for" ī=/āy= "and" uru= "but, yet" nin̮= "still, even so" āy=āni ittāwassaḫni iddāza tāyāni śa tālas
"And I, who shall instill loyalty in them, shall go to the city."

The first two mentioned conjunctive clitics also function as a versatile and contrasting pair of prepositions that perform case-like functions depending on its object's case. They are thus:
śa = allative case (with ACC/ABS), benefactive case (with INS/ERG), ornative case ["with, possessing"] (with GEN)
ma = ablative case (with ACC/ABS), malefactive case (with INS/ERG), privative case ["without, lacking"] (with GEN)

śa tālas "toward the city"
śa tālan "for the benefit of the city"
śa tālaḫ "possessing the city"

ma tālas "away from the city"
ma tālan "to the detriment of the city"
ma tālaḫ "lacking the city"

āyūza śa tālas lumbu śa simbaḫ iddum

go\REAL.NFUT-3sg PREP city-INAN.ACC.SG woman-F.SG.CNS PREP child-N.GEN.SG DEM.DIST-F.NOM.SG

"That pregnant woman went to the city"

---

I have not made much progress in lexicon, but I have made strides in the development of the grammar and syntax.

First attempt at translating something (The King and the God) into the on-off reworking / project of the last few months:

Native orthography

Piru ngikur, kuodeûh nguod. Tzin ritzmi thuthuga na kuodeûh. Tzetzuk na ngeata kuoduar thedmek. Dit 'uq mab thi hiunge naûg: "Tzeb sin thopalh leada!'' Duotze qi naûg kuq mab kuoduar: "Perunos theûk thimos nguodes naûg.'' Perunos thi meûg kuodeûh, guodes naûg 'uethulh. "Dob thopas, Perunos huekeb!'' Zagua lad reri soak Perunos theûk. "Butz hiob nguod tedmetzer?'' "Tzetzuk nab ngeata.'' "Duotze nguomo," daûg 'i Perunos, zeze theûk. Zeze naûg 'i kuoduar meg tzetzutur.

IPA

piru ŋikuɾ, kʷoⁿdew̥ ŋʷon. t͡ʃiⁿd ɾit͡ʃmi θuθuⁿga na kʷoⁿdew̥. t͡ʃet͡ʃuk na ŋe͡ata kʷoⁿdwaɾ θedmek. dit ux maᵐb θi juŋe naŋʷ: "t͡ʃeᵐb ʃin θopaɬ le͡aⁿda!" dwot͡ʃe ɤi naŋʷ kux mam kʷoⁿdwaɾ: "peɾunoʃ θekʷ θimoʃ ŋʷoⁿdeʃ naŋʷ." peɾunoʃ θi meŋʷ kʷodew̥, gʷoⁿdeʃ naⁿgʷ weθuɬ. "doᵐb θopaʃ, peɾunoʃ wekem!" d͡ʒagʷa laⁿd ɾeɾi ʃo͡ak peɾunoʃ θekʷ. "But͡ʃ job ŋʷon tedmet͡ʃer?" "t͡ʃet͡ʃuk nab ŋe͡ata." "dwot͡ʃe ŋʷomo," daⁿgʷ i peɾunoʃ, d͡ʒed͡ʒe θekʷ. d͡ʒed͡ʒe naⁿgʷ i kʷodwaɾ meŋ t͡ʃet͡ʃutur.

Gloss

Old day-OBL, king SBJ-IMPFV. Child lack bear.HAB IMPFV-3 king. Son-NPOSS IMPFV-3 see-STAT king-OBL heart-NPOSS. POSS-3 priest CL:PERSON DAT-3 plea give: "son CL:CHILD PURP-1.OBJ-APPL arrive!'' Thus PFV give priest CL:PERSON king-OBL: "Perunos god DAT-FUT-2 prayer give.'' Perunos DAT go king, prayer give SEQ-3.obj-APPL. "Hear PURP-1.OBJ-2, Perunos father-1.POSS}!'' Sky CL:SPACE ABL-VEN leave Perunos god. ``What CL:WORD SBJ-IMPFV heart-2.POSS-OBL?'' "son-NPOSS IMPFV-1 see-STAT.'' "Thus SBJ-FUT,'' say PFV Perunos, light god. Light give PFV king-OBL wife son-3.POSS-OBL

In the old days, there was a king. The king lacked child(ren). His heart saw son(s). He plead to his priest: "May a child arrive to me!''. Thus spoke the priest to the king: "You will pray to the god Perunos.'' To Perunos went the king, and he gave prayer to him. "Hear me, my father Perunos!'' Perunos came from the skies. "What is in your heart?'' "I want son(s).'' "Thus it will be,'' said Perunos, the light god. The king's wife gave birth to their son.

Notes

The issue is that it's a bit repetitive. Part of this is the original text (or at least the version I used), which repeats the same words a lot. Part of it is that I'm experimenting yet again with small vowel systems and lots of light/coverb constructions, and currently give is one of the most common supporting verbs in this translation. See:

daûg = say, give, ...
zunge naûg = plead (plea give)
guodes naûg = pray (prayer give)
zeze naûg = give birth to (light give)

Not sure if I should draw the line somewhere else (maybe split give and say?) to diversify a bit.

The alternation between daûg and naûg is due to lenition of non-initial elements in compounds and certain phrase types. Similarly, I converted the original text's Werunos to Perunos, because p is the strong form of w (and also of kʷ in some circumstances).

Expressions of thought and emotion often involve θedme- "heart":

heart see = want
have in heart = want

IIRC there's at least one Mayan language that has a lot of emotion terms build in a similar manner.

There's a focus system inspired by Marind which marks the role of the pre-verbal argument. focus + verb + auxiliary/TAM marker form a tight unit - when there is a preverbal focus, the verb shows lenition and the aux moves from post-verbal to post-focus.

The word order and case marking elsewhere differs a bit from Marind though... role marking is mostly minimal (core/oblique, occasionally postpositions derived from nouns, but less common than a fairly semantically wide oblique case). Contrastive topics come before the tight focus/object + verb unit, everything else goes afterwards. I may have accidentally created an OVS language.

Also, the translation of both "bright god" and "give birth" with zeze "light" created an accidental parallel not present in the original version.

I made up a family tree of languages related to Vrkhazhian. Majority of these will just be naming languages.

But I want to start working from a proto-language, so that I can word-gen proto roots, filter them through sound and grammaticalization changes. It will be difficult though because there is some constraints that a word gen may not accommodate.

I have determined the proto-language's phonology to be thus:

Nasal:
/m n ŋ/ <m n n̮>
Aspirate / Fortis:
/pʰ tʰ tɬʰ tʃʰ kʰ/ <p t ć č k>
Ejective / Emphatic:
/pʼ tʼ tɬʼ tʃʼ kʼ/ <ṗ ṭ ć̣ č̣ ḳ>
Voiced / Lenis:
/b d dɮ dʒ g/ <b d ǵ ǧ g>
Fricative:
/f s ɬ ʃ x/ <f s ś š ḫ>
Liquid:
/w r l j/ <w r l y>
Laryngeal:
/ʡ ʔ ħ h/ <ʡ ʔ ḥ h>

Good! I've always loved conlang families.

WeepingElf wrote: ↑Tue Apr 16, 2024 8:10 am Good! I've always loved conlang families.

Ye, I want to generate some words and run them through sound changes so I don't have to think too hard about how something turns out and whether it's consistent or not. Additionally, I overutilize certain consonants and underutilize others, in a way I find too artificial.

Problem is, I haven't found a word generator I've been fully satisfied with using. Mostly because I haven't been able to get any to properly fit these constraints:

• "first and second radicals must not be identical in POA"
• "second and third radicals must not be non-identical in voicing if they are identical in POA"
• "second and third radicals must not both be laryngeals"

If I just make a wordgen generate a series of consonantal root radicals e.g. p-r-s then I can more easily filter out rulebreakers, but it doesn't help with generating actual words.

And I wish there was some crosslinguistic zipf's law distrubution percentages for phonemes of the world

Ahzoh wrote: ↑Tue Apr 16, 2024 3:33 pm Problem is, I haven't found a word generator I've been fully satisfied with using. Mostly because I haven't been able to get any to properly fit these constraints:

• "first and second radicals must not be identical in POA"
• "second and third radicals must not be non-identical in voicing if they are identical in POA"
• "second and third radicals must not both be laryngeals"

If I just make a wordgen generate a series of consonantal root radicals e.g. p-r-s then I can more easily filter out rulebreakers, but it doesn't help with generating actual words.

And I wish there was some crosslinguistic zipf's law distrubution percentages for phonemes of the world

If you don't want to solve the general problem yourself, have you considered just hacking together some code for the specific case? Either generate the words with your own code, or use a dumb generator, then write a Python script or something to check those constraints and discard any that are illegal. It's probably doable in not that many lines unless the constraints are much more complex than they sound here.

chris_notts wrote: ↑Tue Apr 16, 2024 3:39 pm

Ahzoh wrote: ↑Tue Apr 16, 2024 3:33 pm Problem is, I haven't found a word generator I've been fully satisfied with using. Mostly because I haven't been able to get any to properly fit these constraints:

• "first and second radicals must not be identical in POA"
• "second and third radicals must not be non-identical in voicing if they are identical in POA"
• "second and third radicals must not both be laryngeals"

If I just make a wordgen generate a series of consonantal root radicals e.g. p-r-s then I can more easily filter out rulebreakers, but it doesn't help with generating actual words.

And I wish there was some crosslinguistic zipf's law distrubution percentages for phonemes of the world

If you don't want to solve the general problem yourself, have you considered just hacking together some code for the specific case? Either generate the words with your own code, or use a dumb generator, then write a Python script or something to check those constraints and discard any that are illegal. It's probably doable in not that many lines unless the constraints are much more complex than they sound here.

Seconded; this sounds like the job for a simple custom script in, say, Python rather than trying to force some prewritten word generator to do what you want it to.

chris_notts wrote: ↑Tue Apr 16, 2024 3:39 pm

Ahzoh wrote: ↑Tue Apr 16, 2024 3:33 pm Problem is, I haven't found a word generator I've been fully satisfied with using. Mostly because I haven't been able to get any to properly fit these constraints:

• "first and second radicals must not be identical in POA"
• "second and third radicals must not be non-identical in voicing if they are identical in POA"
• "second and third radicals must not both be laryngeals"

If I just make a wordgen generate a series of consonantal root radicals e.g. p-r-s then I can more easily filter out rulebreakers, but it doesn't help with generating actual words.

And I wish there was some crosslinguistic zipf's law distrubution percentages for phonemes of the world

If you don't want to solve the general problem yourself, have you considered just hacking together some code for the specific case? Either generate the words with your own code, or use a dumb generator, then write a Python script or something to check those constraints and discard any that are illegal. It's probably doable in not that many lines unless the constraints are much more complex than they sound here.

I'm not a coder

In anycase, someone showed me this and it seems to have done the job:
https://lingweenie.org/conlang/lexifer-app.html

With such rules as: A bit wordy but it gets the job done

It can even factor in percentages of distribution, such as:
CṾCCS:23
CṾCS:13
CḶCS:13
CṾCVCS:13
CVCḶCS:13
CḶCVCS:7
CLCḶCS:7
CṾCCVCS:4
CVCCḶCS:4
CVCṾCCS:1
CLCṾCCS:1
CVCCṾCCS:1
(The underdots are for the stressed vowel category and S is for suffixes, like case markers)

I think these percentages are naturalistic, about 75% of words generated will be the first five types, while the remaining 25% are the final seven.

Though I'm not sure if a language would have more bisyllables than trisyllables, especially if it has morphology like Semitic languages. A lot of what I put is impressionistic based on the list of Proto-Semitic reconstructed words on wiktionary.

Today I've accomplished... nothing. I haven't had much time recently to do any conlanging. I don't know about you, but I, for one, cannot just do a little bit for half an hour here and there. I need a morning or an afternoon where I can concentrate on it. Anyway, I had a spare hour-or-so and thought I'd look at where I left things, what needs attention, what has fallen by the wayside, and oooooh boy have I taken my eye off the ball! I have two days to myself (if all goes well) in 10 days' time and I hope to spend most of one of them to get things in order, make a to-do list, and see where I am with it all.

At the moment I have lots of notes spread out across emails, text notes on my phone, on my laptop, on paper... some of which are contradictory to other notes... it's a bit of a mess. However, if I can use one of those two days to tidy it all up, I have a week off about a week on from that, so I may be able to put a few shifts in and make some progress.

I didn't need to share this, but putting it out there somehow makes me feel committed to it.

I know that feeling very well - my own worldbuilding and conlanging notes are in a similar state of disarray. It may be comfort for you to know that the greatest conlanger and worldbuilder of the 20th century, J. R. R. Tolkien, wasn't any different (except of course that he had no digital devices, and it was all on paper), which is the main reason why we do not have the book every Tolkienian linguist wishes to find, the Historical Grammar of the Quendian Languages.

Same here.

The trouble is, conlanging and conworlding are last on the list, right after everything else I have to do , and that list is incredibly long these days.

I hope things calm down somehow at some point in the future.

Some sound changes: