Conlang Random Thread

Conworlds and conlangs
User avatar
Man in Space
Posts: 1566
Joined: Sat Jul 21, 2018 1:05 am

Re: Conlang Random Thread

Post by Man in Space »

Raphael wrote: Tue Dec 27, 2022 12:58 amVery interesting etymological development, but I'm not entirely sure I'm following how they got from "mother of" to "a brother" in the case of helium. And while the "mother of" system is very good for elements that were isolated at some point during the development of modern chemistry, I'm less sure about those that were already known in their elemental form in times before the development of modern chemistry, like various metals. Wouldn't these already have have long-established traditional names by the time it became known that they're "elements" in the modern sense?
I'm kind of glad you brought this up because I was actually just writing about this both in the megathread and on my blog, so it's on the brain!

In the case of helium, there's a couple factors at play here. First, the idea of a second or secondary thing being associated with or outright being the "brother" of that thing is a deep-rooted one in the PTO cultural complex (like, even the word höhsë 'four' was originally a compound of 'brother' + isë 'three'). Second, it's shorter than saying eámr 'father', which further would obscure the "brotherly" symbolism. Third, it was kind of a kludgy workaround for using móm in this specific application. Normally the ikłe X construction is used to specify atomic weights, full stop—for instance, móm ĝ kr ikłe 14 'carbon-14'—but because of cultural stubbornness they wanted to still use móm for a three-nucleon nucleus, even though this would be ambiguous between tritium and tralphium (helium-3's older, snobby name that I like to use). was kind of agreed upon as the least-worst option, though it's essentially a -2 on the weight implied by the ikłe X. (I totally did not just pull this out of my ass because you saw an inconsistency I missed.)

(Incidentally, electrons are egáma (sg. keáma) 'grandmothers'.)

Anyway, there are nonetheless other strategies for naming elements (and some noteworthy substances or materials).
- Elements from antiquity have their own names: ûn 'gold', gákłi 'silver', ürḫar 'bismuth', ëtsín 'tin'.
- Iron is a special case of the above, as it has two words. Raw, unworked, or elemental iron is ĝaĝ. Worked iron is tiên, a loan from Mute Caber (the old saw goes that the Caber gave the Tim Ar the secret of writing in exchange for metallurgy, but the historicity of that is questionable). Steel is kahál 'strong', used substantively as a shortening of tiên kahál 'strong (worked) iron'.
- Bronze also gets its own special name, lugna kiĝ 'red metal'.
- The móm n X formula is alive and kicking: móm ĝ kr 'mother of charcoal' > 'carbon', móm ĝ ḫuú 'mother of breath' > 'oxygen', móm n raáḫ 'mother of division' > 'U-235'.
- Some get named by just repurposing an existing word: ëslug 'hardy' > 'tungsten'.
- Descriptive phrases are also a Thing: lugna sígna 'yellow metal' > 'uranium'.

A few honorable mentions:
- The adjective hún 'dark-colored' (here, 'blue') is used to describe nuclear things: agamári hún 'nuclear weapons' (lit. 'blue weapons').
- A rocket's exhaust/thrust/thrust plume is called its saman 'voice'.

Also I just realized how I can kind of show that î is not a preposition but a derivational marker: Given some placenames Ákmrgámr and Kuasagua, you could say something like tiên î Kuasagua 'Kuasaguan iron'; tiên ĝ Kuasagua 'iron from Kuasagua' would just imply that the iron had maybe been shipped there. As well, for purposes of syntax, the î X construction is treated as an adjective and not an oblique phrase.

I also also just spent like two hours typing up mathematical conventions in CT for my blog…my brain's kind of…fried. If I can manage not to fall asleep maybe I'll post about it in the megathread.
User avatar
Raphael
Posts: 4194
Joined: Sun Jul 22, 2018 6:36 am

Re: Conlang Random Thread

Post by Raphael »

Thank you for that long and very informative reply!
Man in Space wrote: Tue Dec 27, 2022 2:30 am
I also also just spent like two hours typing up mathematical conventions in CT for my blog…my brain's kind of…fried. If I can manage not to fall asleep maybe I'll post about it in the megathread.
Don't forget to rest!
User avatar
Man in Space
Posts: 1566
Joined: Sat Jul 21, 2018 1:05 am

Re: Conlang Random Thread

Post by Man in Space »

Raphael wrote: Tue Dec 27, 2022 2:40 amThank you for that long and very informative reply!
You're welcome!
Raphael wrote: Tue Dec 27, 2022 2:40 amDon't forget to rest!
Thank you!

--------------------------------

Fractions take the form [numerator] n [denominator], i.e. it's a regular genitive construction: dún n dúnki 'one of five' = 'one-fifth'. You might not expect that 'ninetieth' and 'one hundred tenth' are irregular: dún n arihkêgê 'one-ninetieth', dún n ténḫo arig 'one-one-hundred-tenth'. Improper fractions are permissible and, when unqualified by a governing whole number, not considered "improper": ki m hága 'five halves'.

The decimal point is késög, an inposition meaning 'after, post-, once s.th. has ended'. It can be used to separate a fractional as well. This is a formal mathematical convention, though; in common speech, people usually use łá instead. Łá is a very important word to know in mathematics: Originally a nominal conjunction—it meant 'and' between nouns, and still does—it basically serves as a 'this number/quantity/part of this equation is over' marker. You can't have two numbers in a mathematical expression next to each other without some sort of intervening word if you don't want them being read as a number (like how one might dictate the number one million, two hundred seventy-two thousand, three hundred sixty-eight as "1-2-7-2-3-6-8" for brevity); when there's no other word to disambiguate, you default to łá. (You can partially credit this rule to CT's VOS sentence structure.)

As you get into more complex mathematics, you'll run into doí (pl. odí) a lot. Literally it means 'quantity'. It often is used to bookend complex expressions to which other operators can be applied. For instance, doí hága łá x lé, doí łá isë ïrëĝ '(x + 2) cubed'. Note that the first doí doesn't take a łá, although the second one does require one.

Instead of saying áge 'is' or using a similar term, the contronym kaá is used. It means both 'build, make' and 'collapse, fall apart'. For operations that increase value, the former meaning is relevant; for those that decrease it, the latter is.

Negative numbers are X ág: höhsë ág 'negative four, minus four'. If you have to use késög (or łá colloquially), it comes before the ág: dún késög ág dún m höhsë 'one and one-quarter'.

--------------------------------

Your simple A-B-C formula is kaá C łá A łá B X, where X is an inposition. Some mathematical terms/operators in CT:

- To add, the inposition is 'on, on top of, atop, alighted on, over'.
- To subtract, it's ê 'off, up, into itself'.
- Multiplication is handled with ĝöl 'because of, due to'.
- Division uses mió 'with, by means of, using, via'.
- Raising to a power is handled with ïrëĝ 'again'.
- Taking a root is expressed with séḫ 'for the cause of, for the sake of, for'.

A couple examples of this in action:

- Kaá hága łá dún łá hága lé '1 + 2 = 3'
- Kaá dún łá isë łá hága ê '3 - 2 = 1'
- Kaá rarê łá höhsë łá hága ĝöl '2 * 4 = 8'
- Kaá höhsë łá rarê łá hága mió '8 / 2 = 4'
- Kaá rarê łá hága łá isë ïrëĝ '2 ^ 3 = 8'
- Kaá hága łá rarê łá isë séḫ '8 ^ 1/3 = 2'

Because CT absolutely loves zero-deriving everything from everything else, these can be used as nouns or postpositions in their own right: X łá dún lé 'X + 1', hága łá lé ĝöl 'two times the sum'.

--------------------------------

Something really stupid:

Kaá arig łá doí hága késög dún n ténḫo łá höhsë késög dún dún ê doí X łá hága ê, doí łá hága ïrëĝ, doí łá isë séḫ
'The quantity (2 1/100 - 4.11 + (X-2)^2), the cube root of the quantity equals ten'
Ahzoh
Posts: 456
Joined: Wed Jul 18, 2018 1:52 pm

Re: Conlang Random Thread

Post by Ahzoh »

I decided that Vrkhazhian will have this system of cases:

Unbound Cases (marked noun cannot be modified by other nouns):
Nominative (the doer of the action), marked by the suffix -s
Accusative (the recipient of the action), marked by the suffix -n
In Neuter Nouns, the Nominative and Accusative are combined, marked by -s or -m (maybe an Ergative case)

Bound Cases (marked noun modifies another noun in the construct state):
Genitive (what the head noun comes from or belongs to)
Equative (what the head noun resembles)
Ornative (what the head noun has)
Privative (what the head noun lacks)
Not sure how I'll mark these, whether they'll be opaque or transparent in their morphology.
User avatar
Jonlang
Posts: 351
Joined: Tue Sep 04, 2018 8:59 am
Location: Gogledd Cymru

Re: Conlang Random Thread

Post by Jonlang »

In one of my three related conlangs, I want it to put numerals after the noun - however the languages are based on IE languages and, according to WALS, there isn't a single language within 4000 miles of the UK that has noun-numeral order (unless you count Arabic which is given by WALS as having "no dominant order"). Given that I can't look to IE languages, what could push a language towards putting the numeral after the noun, rather than before it?

EDIT: Well it looks like Greenlandic is within 2000 miles of me and uses noun-numeral, but due to the way the WALS's maps present data, I didn't initially see Greenlandic.
Twitter won't let me access my @Jonlang_ account, so I've moved to Mastodon: @jonlang@mastodon.social
User avatar
Rounin Ryuuji
Posts: 2994
Joined: Wed Dec 23, 2020 6:47 pm

Re: Conlang Random Thread

Post by Rounin Ryuuji »

Jonlang wrote: Fri Jan 06, 2023 4:04 am In one of my three related conlangs, I want it to put numerals after the noun - however the languages are based on IE languages and, according to WALS, there isn't a single language within 4000 miles of the UK that has noun-numeral order (unless you count Arabic which is given by WALS as having "no dominant order"). Given that I can't look to IE languages, what could push a language towards putting the numeral after the noun, rather than before it?

EDIT: Well it looks like Greenlandic is within 2000 miles of me and uses noun-numeral, but due to the way the WALS's maps present data, I didn't initially see Greenlandic.
Why not just have a very strict noun-adjective order emerge, having first ordinals shifting to follow the noun, and the cardinal numbers following suit?
Kuchigakatai
Posts: 1307
Joined: Mon Jul 09, 2018 4:19 pm

Re: Conlang Random Thread

Post by Kuchigakatai »

Latin had both noun - cardinal and cardinal - noun as allowed word orders; maybe other old IE languages did too. It could simply be a conservative feature of your European IE conlang.
User avatar
WeepingElf
Posts: 1386
Joined: Sun Jul 15, 2018 12:39 pm
Location: Braunschweig, Germany
Contact:

Re: Conlang Random Thread

Post by WeepingElf »

Yes. Numerals after the noun occasionally occur in various IE languages.

Latin:

Gallia omnia est divisa in partes tres.
(Caesar, Bellum Gallicum)

English:

Old King Cole was a merry old soul,
and a merry old soul was he.
He called for his pipe and he called for his bowl,
and he called for his fiddlers three.
... brought to you by the Weeping Elf
My conlang pages
Travis B.
Posts: 6309
Joined: Sun Jul 15, 2018 8:52 pm

Re: Conlang Random Thread

Post by Travis B. »

WeepingElf wrote: Sat Jan 07, 2023 6:28 am Yes. Numerals after the noun occasionally occur in various IE languages.

[snip]

English:

Old King Cole was a merry old soul,
and a merry old soul was he.
He called for his pipe and he called for his bowl,
and he called for his fiddlers three.
I should note that lyrics have a tendency to wax poetic, and poetic language often does not follow quite the same rules as normal spoken or written language, whether formal or colloquial.
Yaaludinuya siima d'at yiseka ha wohadetafa gaare.
Ennadinut'a gaare d'ate ha eetatadi siiman.
T'awraa t'awraa t'awraa t'awraa t'awraa t'awraa t'awraa.
Moose-tache
Posts: 1746
Joined: Fri Aug 24, 2018 2:12 am

Re: Conlang Random Thread

Post by Moose-tache »

More to the point, languages don't just copy their neighbors. English didn't switch to natural gender because all the cool kids were doing it. If you want noun-number order, just come up with an internal morphological or phonological motivation for it.
I did it. I made the world's worst book review blog.
hwhatting
Posts: 1090
Joined: Mon Jul 09, 2018 3:09 am
Location: Bonn
Contact:

Re: Conlang Random Thread

Post by hwhatting »

Besides what Elf said, several IE languages have a construction of genitive (pl.) plus numeral which has a partitive or indentifying use, e.g. in somewhat outdated German der Schwestern drei waren schön "Three of the sisters were beautiful" ich sah der Riesen fünfe "I saw five of the giants / five giants (but there could have been more)".
When Russian puts the numeral after the noun, it implies an approximate number: pyat' domov "five houses", but domov pyat' "about 5 houses". Your language could have similar constructions spreading beyond these applications.
User avatar
Rounin Ryuuji
Posts: 2994
Joined: Wed Dec 23, 2020 6:47 pm

Re: Conlang Random Thread

Post by Rounin Ryuuji »

I think when there's a will to do something with a language in this sort of way, there's generally a plausible way to get to it, even if it isn't something that ended up happening in our reality.
User avatar
foxcatdog
Posts: 1608
Joined: Fri Nov 15, 2019 7:49 pm

Re: Conlang Random Thread

Post by foxcatdog »

What's the best minimal consonant phoneme inventory you can come up with? For me i'm thinking of making a language with the following phoneme inventory. Possibly with 3 vowels or 5.
*p *t *k
*s
*w *j
*m *n
bradrn
Posts: 5747
Joined: Fri Oct 19, 2018 1:25 am

Re: Conlang Random Thread

Post by bradrn »

foxcatdog wrote: Sun Jan 08, 2023 6:00 pm What's the best minimal consonant phoneme inventory you can come up with?
/t k b d a/ + epenthetic (non-phonemic) [ə]

EDIT: oh, didn’t notice you said ‘consonant’ only. In that case, /t k b d/.
Last edited by bradrn on Sun Jan 08, 2023 6:32 pm, edited 1 time in total.
Conlangs: Scratchpad | Texts | antilanguage
Software: See http://bradrn.com/projects.html
Other: Ergativity for Novices

(Why does phpBB not let me add >5 links here?)
User avatar
Rounin Ryuuji
Posts: 2994
Joined: Wed Dec 23, 2020 6:47 pm

Re: Conlang Random Thread

Post by Rounin Ryuuji »

foxcatdog wrote: Sun Jan 08, 2023 6:00 pm What's the best minimal consonant phoneme inventory you can come up with?
/n t k ʔ s j w/
Kuchigakatai
Posts: 1307
Joined: Mon Jul 09, 2018 4:19 pm

Re: Conlang Random Thread

Post by Kuchigakatai »

/t ʔ m n w l/, with /i a o/ as vowels (/i/ is sometimes [j])
Travis B.
Posts: 6309
Joined: Sun Jul 15, 2018 8:52 pm

Re: Conlang Random Thread

Post by Travis B. »

bradrn wrote: Sun Jan 08, 2023 6:09 pm EDIT: oh, didn’t notice you said ‘consonant’ only. In that case, /t k b d/.
I would say something similar, /t k b d/, but with /t/ being realized as [s] or [z] before /i/ (depending on whether its intervocalic), /d/ being realized as [j] before /i/, /b/ being realized as [w] before /u/, /b d/ being realized as [m n] adjacent to nasal vowels, and intervocalic /t k b d/ (except for /t d/ before /i/, /b/ before /u/, or /b d/ adjacent to nasal vowels) being realized as [b d β ð] respectively.
Yaaludinuya siima d'at yiseka ha wohadetafa gaare.
Ennadinut'a gaare d'ate ha eetatadi siiman.
T'awraa t'awraa t'awraa t'awraa t'awraa t'awraa t'awraa.
User avatar
Jonlang
Posts: 351
Joined: Tue Sep 04, 2018 8:59 am
Location: Gogledd Cymru

Re: Conlang Random Thread

Post by Jonlang »

Thanks everyone, I've read your comments but I'm very short of time at the moment so I'll bookmark them for further consideration.
Twitter won't let me access my @Jonlang_ account, so I've moved to Mastodon: @jonlang@mastodon.social
chris_notts
Posts: 682
Joined: Tue Oct 09, 2018 5:35 pm

Re: Conlang Random Thread

Post by chris_notts »

It's a trivial one, but I want to change an allomorph of the 1st person inclusive plural agreement marker, undecided about whether /χ/ is distinctive enough given the other ones. Basically, the verb is made up of a prefix chain and then a root, which are a grammatical (distributional) word but have a certain amount of phonological independence. Currently, the person markers are:

Code: Select all

                       Before vowel     As coda/at end of prefix chain
1st person      ʃ                         j/0 (lost after i)
2nd person    m                        N (nasal assimilating to following consonant)
3rd person     r/0                      0
1st+2nd         k                         kV (V = copy of preceding vowel)
For various reasons including some ambiguity with the -kV- form, I'm thinking about shifting to a single consonant for the inclusive, but that single consonant cannot really be a stop due to phonotactic constraints. The obvious one would be the dorsal fricative x~χ, but for some reason that feels a bit close to the post-alveolar-palatal ʃ~j. This is a bit worrying because the coda forms of the agreement morphemes may be buried 2-3 syllables into the verbal word, as in:

chanayunteha'
[cha-n-a-yu-n]-[te-ha']
[SAP-SUBORD-IMPERF-CISLOCATIVE-2]-[APPL-be]
"(When/as) you coming to me, ..."

I was also considering using /f/, since x->f is a fairly common sound change (e.g. English and other Germanic languages), so I think k~f as a synchronous alternation would feel vaguely plausible.

Compare:

1excl > 2: chanayuyteha'

3 > 1inc:
ernayu_ku_teha'
OR ernayu_χ_teha'
OR ernayu_f_teha'
OR ???

So which set of consonants do you think is most distinctive in coda position mid-word, {j, N, χ, 0} or {j, N, f, 0}?
bradrn
Posts: 5747
Joined: Fri Oct 19, 2018 1:25 am

Re: Conlang Random Thread

Post by bradrn »

chris_notts wrote: Sat Jan 21, 2023 12:58 pm So which set of consonants do you think is most distinctive in coda position mid-word, {j, N, χ, 0} or {j, N, f, 0}?
I think they’re both equally distinctive. [χ] and [ʃ] already feel very far away to me, and [χ] and [j] even further apart. But really, I think it all depends on how your language structures its consonants — e.g. in Hebrew they’d be straightforwardly different, whereas in Spanish (where [j~ʝ] can merge into [ʒ] and [x] can be velar) they might be considered a lot more similar. I recall Nortaneous talking about the fact that which phonemes ‘choose’ to interact with which others can be quite different between languages; something similar applies here, I think.
Conlangs: Scratchpad | Texts | antilanguage
Software: See http://bradrn.com/projects.html
Other: Ergativity for Novices

(Why does phpBB not let me add >5 links here?)
Post Reply