(Conlangs) Q&A Thread - Quick questions go here

A forum for all topics related to constructed languages
User avatar
Reyzadren
greek
greek
Posts: 684
Joined: 14 May 2017 10:39
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Reyzadren »

No, I only include things that I can pronounce, though my conlang has a simple phonological inventory anyway.
Image conlang summary | Image griushkoent thread
shimobaatar
korean
korean
Posts: 10373
Joined: 12 Jul 2013 23:09
Location: UTC-04:00

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by shimobaatar »

sangi39 wrote: 26 Feb 2020 20:11
KaiTheHomoSapien wrote: 26 Feb 2020 19:53 […] Do you include sounds in your conlang that you yourself can't pronounce? […]
[…]

However, I don't think it's unreasonable to include those sounds within a conlang, but, for me personally, I try and do some reading about them first, see how they behave in languages that have them, e.g. how do they interact with other sounds (do they cause vowels to shift around), do they tend to occur only in certain environments (or, conversely, is there a restriction on what sounds can occur around them), and stuff like that. That way, when it starts to coming to word-building and describing allophony, those sounds (hopefully) seem to fit better.

But, basically, if it turns up as "pronounceable" somewhere in the real world, then there's nothing stopping you throwing it at your conlang (strictly speaking, there's nothing stopping you throwing anything you want at your conlang, but that's a wider discussion).

[…]
[+1]

Ideally, I intend for my conlangs to make sense as fictional languages spoken by fictional people, including in terms of phonology. I care about whether certain contrasts or distributions or whatever seem plausible for a theoretical human language far more than I care about whether or not I, myself, can pronounce the languages I create in a way that comes close to how their imaginary native speakers would pronounce them.

Although that hasn't always been my goal, I've never intended for any of my conlangs to be used in any way by any real people other than myself, but even then, I've never intended to speak any of them, or to use them for actual communication.
sangi39 wrote: 26 Feb 2020 20:11 EDIT: I'm also attempting (rather slowly), to work on a conworld mainly centred around being a place for languages to go, but with the same sort of variety that you might find here on Earth. If I stuck to just what I could pronounce, every language would basically be pretty Englishy, which would just be boring.
This sounds more or less like what I'd like to do as well.
User avatar
KaiTheHomoSapien
greek
greek
Posts: 641
Joined: 15 Feb 2016 06:10
Location: Northern California

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by KaiTheHomoSapien »

Thank you all for your answers. [:D] I'm honestly a little disappointed to hear that some people don't even try and pronounce their own conlangs. But I guess the next time I see a conlang with a word like bnšxćelk’ih’xkuq, I'll take some solace in knowing that not even the creator knows how to pronounce it. [:P]
shimobaatar wrote: 27 Feb 2020 01:42 Although that hasn't always been my goal, I've never intended for any of my conlangs to be used in any way by any real people other than myself, but even then, I've never intended to speak any of them, or to use them for actual communication.
I don't intend for anyone else to use my conlangs either nor do I use them for actual communication, but I like to read sentences and paragraphs I've written in them out loud nonetheless. I love the sound of Lihmelinyan and have made a few recordings of me speaking it. I may share those here some time. [:)] My phonological inventories also tend to be a bit simpler and IE-like and that is partially because I've always been more interested in morphology than in phonology.
Last edited by KaiTheHomoSapien on 28 Feb 2020 19:16, edited 1 time in total.
Image
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by lsd »

For myself, I conlang for here and now, that is why I need pronounceable sounds... by myself (it's an idiolect...)
For conlangers of other planets, are you planning new IPA with sounds that don't exist on earth...
User avatar
J_from_Holland
sinic
sinic
Posts: 217
Joined: 19 Mar 2015 17:19
Location: On this forum

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by J_from_Holland »

Barmish has over 3000 words in the dictionary, but I didn't make an etymology for them. Now I'm trying to reverse-engineer a proto-lang for Barmish (Proto-Thyllic), but this is going painfully slow, especially with the naturalistic irregularities I'm adding (see example). Do you have any idea how to speed up this process?

Example:
The comparative of both uo (nice, fun) and uhng (pleasant, fine) is ůpůr, because in Old Barmish, both words were relatively neutral variations of the same adjective which would get the same comparative. Over the years, the comparative stayed the same but the neutral adjectives got their positive meanings and gained sound symbolism. "U" is pronounced /y/ and it's the "formal" vowel in Barmish: when addressing someone formally, the resulting words and endings tend to have a lot of /y/'s.
A few years, I posted about Bløjhvåtterskyll. That's Barmish nowadays, and it's quite different from back then.
:nld: :mrgreen: | :eng: [:D] | :deu: [:D] | :fra: [:P] | :ell: [:$] | :nor: [:$]
User avatar
Pabappa
greek
greek
Posts: 577
Joined: 18 Nov 2017 02:41

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Pabappa »

J_from_Holland wrote: 03 Mar 2020 20:11 Barmish has over 3000 words in the dictionary, but I didn't make an etymology for them. Now I'm trying to reverse-engineer a proto-lang for Barmish (Proto-Thyllic), but this is going painfully slow, especially with the naturalistic irregularities I'm adding (see example). Do you have any idea how to speed up this process?
Unfortunately I can only advise you that this is going to be a long slog. If you want to do etymologies and have them be naturalistic, you're going to have to handle each word individually and make sure the proto-language stands on its own. I spent years converting the 6000 words I had generated for Poswa and Pabappa into 6000 words that descended from a shared parent language. For the most part, I looked for roots in the parent language that were "close to" the ones that I got from the word generator both in meaning and in sound, but prioritizing meaning over sound. In some cases, I found no close matches on sound and just went with semantics, meaning I was creating a wholly new etymology. This was actually easier than trying to match both sound and meaning, but I tried to match the sound even so because I had become attached to some words that had been in the language for a long time.

This is very labor intensive, but one good thing is that sometimes new ideas show up on the way .... e.g. the words for "ribbon" and "to offer, propose" merged in Poswa, so now I have a single polysemic root with both meanings. Another word pair that merged is "line" and "smooth", so now I have it as "line; smooth boundary". Of course, not every potential merger is semantically feasible, which means that quite often I have to go back and generate yet another etymology, meaning it takes even longer. I should also add that in most cases, when two words merged, I said that they were originally one word all along and that the meanings split within the daughter language. This is more naturalistic than carrying down two homophonous roots for thousands of years, though both situations occur in the wild.

Depending on what your strategy is, you may or may not see opportunities for merging and splitting words like this. I got the mergers because the parent language was pure CV and because there were a lot of sound changes along the way, so two very different CVCV + CVCV compounds quite often would merge into the same result in the end. My evolution time from the parent language to the present day is 7,000 years, possibly more than what youre looking at. But if your proto-language was already highly differentiated in its lexicon ... i.e. not many similar roots to start with, you might not see quite as much of this going on. Polysemy by sound change collision is a fun diversion and not necessary to make a good language, but if you can find the time, polysemy in general is common in languages and you can play with semantic shifts once you've got your lexicon in the state you want it.

I should add that my languages lack loanwords by design....not *every* word in your language needs to come from its parent language, unless you are also striving for a perfectly pure language with no loanwords.
Kavunupupis, šiŋuputata.
When I see you pointing at me, I know I'm in trouble. (Play)
User avatar
J_from_Holland
sinic
sinic
Posts: 217
Joined: 19 Mar 2015 17:19
Location: On this forum

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by J_from_Holland »

Thanks for the comprehensive reply!

Something that you seem to do differently is using a word generator. I specifically prohibit myself from using a word generator for Barmish. The only time I used it was to generate test vocabulary for the proto lang, to test the sound changes, but those words didn't come into the actual protolang unless they happened to result in actual Barmish words.

In terms of meaning, i choose everything manually, and the syntax of the protolang (Thyllic) is much more analytical than the syntax of Barmish itself.
A few years, I posted about Bløjhvåtterskyll. That's Barmish nowadays, and it's quite different from back then.
:nld: :mrgreen: | :eng: [:D] | :deu: [:D] | :fra: [:P] | :ell: [:$] | :nor: [:$]
User avatar
KaiTheHomoSapien
greek
greek
Posts: 641
Joined: 15 Feb 2016 06:10
Location: Northern California

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by KaiTheHomoSapien »

I've never really been able to make naturalistic etymology. That seems beyond my abilities. I of course employ derivational affixes in regular patterns and I will sometimes derive words from each other that are related lexically, but I can't explain the deep etymology of every word I come up with, especially since I often come up with new words randomly. A sequence of sounds just comes to me when I think of the real-world object or feeling or action that I'm trying to name. And those can't possibly have a realistic etymology since they're more or less random. Kudos to anyone trying to do that, though. I think that's one of the hardest aspects of natural language to replicate.
Image
User avatar
Sequor
sinic
sinic
Posts: 438
Joined: 30 Jun 2012 06:13

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Sequor »

KaiTheHomoSapien wrote: 06 Mar 2020 19:29 I've never really been able to make naturalistic etymology. That seems beyond my abilities. I of course employ derivational affixes in regular patterns and I will sometimes derive words from each other that are related lexically, but I can't explain the deep etymology of every word I come up with, especially since I often come up with new words randomly. A sequence of sounds just comes to me when I think of the real-world object or feeling or action that I'm trying to name. And those can't possibly have a realistic etymology since they're more or less random. Kudos to anyone trying to do that, though. I think that's one of the hardest aspects of natural language to replicate.
As with anything else in conlanging, it's mostly a question of spending dozens upon dozens of hours creating complications with words... You'd need to at least create a proto-language, probably backwards in time, if not also a whole environment of languages around your language throughout time to borrow or calque words from, all while working out the origins of all your words one by one.

- Change all the meanings of this word, very drastically (Latin capere 'to grab sth' > Spanish caber 'to fit [inside a place]')...
- Change all the meanings of that word, less drastically (tenēre 'to hold sth' > tener 'to have sth; to have [to do sth]')...
- Switch which is the basic meaning of that other one (augurium 'the profession or act of bird divination; (sometimes) an omen obtained through bird divination' > agüero 'a (usually bad) omen; (rarely) the ancient and medieval practice of bird divination')...
- Retain yet that other word just as it was 2000 years ago even though it was a very colloquial word at the time (panticēs 'belly' > la panza 'belly')...
- Merge this other pair into one word while retaining both meanings (somnus 'sleep', somnium 'dream' > el sueño 'sleep; dream')...
- Borrow this one (dēdicāre 'to problaim sth, dedicate sth' > dedicar 'to dedicate sth')...
- Borrow that one but only use it very metaphorically (focum 'hearth' > el foco 'point of attention, focus; light bulb; epicentre of an earthquake')
- Borrow that other one but use it very specifically (albēdō 'whiteness' > albedo '(in physics) radiation reflected by a body, particularly light', conclāvem 'lockable room' > conclave 'the conclave i.e. the room in the Vatican where the Roman Catholic cardinals trap themselves until they choose the new Pope')
- Calque that one (suprā dīctus 'said above' > susodicho 'said above')...
- Keep that word but add an affix to reinforce its length (apem 'bee' > *ap-iculam > abeja, generāre 'to beget sb' > *in-generāre > semi-learned engendrar 'to beget sb; produce [a project]')...

...Besides other possibilities, like clipping and compounding.

I don't really think it's hard in terms of effort within one hour, but it does take many hours to get anywhere...
hīc sunt linguificēs. hēr bēoþ tungemakeras.
yangfiretiger121
sinic
sinic
Posts: 337
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 »

The reworked version of my roleplay setting's main language has three diphthongs, [ɐi̯ ɐʊ̯ oi̯], of which [ɐi̯ oi̯] become [ɐj oj] word-internally while prevocalic. Is word-internal, prevocalic [ɐʊ̯ → ɐw] plausible, even though its second component is [ʊ̯] rather than [u̯]?
Alien conlangs (Font may be needed for Vai symbols)
User avatar
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by sangi39 »

yangfiretiger121 wrote: 14 Mar 2020 14:05 The reworked version of my roleplay setting's main language has three diphthongs, [ɐi̯ ɐʊ̯ oi̯], of which [ɐi̯ oi̯] become [ɐj oj] word-internally while prevocalic. Is word-internal, prevocalic [ɐʊ̯ → ɐw] plausible, even though its second component is [ʊ̯] rather than [u̯]?
I would say it is.
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
yangfiretiger121
sinic
sinic
Posts: 337
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 »

Thanks.

That same language once had whistled and labialvelarized sibilants (cover symbol S), such as [sᶲ sʷ], in complimentary distribution based on vowel backness (all back vowels are rounded). With a vowel system of /ɐ e i o ʊ/, are [Sʷo Sʷʊ → Sᶲo Sᶲʊ] plausible?
Alien conlangs (Font may be needed for Vai symbols)
User avatar
Omzinesý
mongolian
mongolian
Posts: 4079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Omzinesý »

Omzinesý wrote: 10 Mar 2020 18:39

Code: Select all

	Control		Noncontrol 
Nonpast	SG1 tok-a-j	|	tokk-u-j
	SG2 tok-a-p	|	tokk-u-p
	SG3 tokk-a 	|	tok-u
	PL1 tokk-a-si	|	tok-u-si
	PL2 tok-a-psi	|	tok-u-psi
	PL3 tokk-a	|	tokk-u
	--------------------------------------------
Past	SG1 tokk-a-j	|	tok-u-j
	SG2 tokk-a-p	|	tok-u-p
	SG3 tok-a 	|	tokk-u
	PL1 tok-a-si	|	tokk-u-si
	PL2 tokk-a-psi	|	tokk-u-psi
	PL3 tok-a	|	tok-u
This is my verb paradigm in Ivka.
It is complex enough, but it is very regular. The only markers are Person suffix, Control marker vowel, and consonant gradation.
How could I make it less regular?
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by sangi39 »

Omzinesý wrote: 21 Mar 2020 10:43
Omzinesý wrote: 10 Mar 2020 18:39

Code: Select all

	Control		Noncontrol 
Nonpast	SG1 tok-a-j	|	tokk-u-j
	SG2 tok-a-p	|	tokk-u-p
	SG3 tokk-a 	|	tok-u
	PL1 tokk-a-si	|	tok-u-si
	PL2 tok-a-psi	|	tok-u-psi
	PL3 tokk-a	|	tokk-u
	--------------------------------------------
Past	SG1 tokk-a-j	|	tok-u-j
	SG2 tokk-a-p	|	tok-u-p
	SG3 tok-a 	|	tokk-u
	PL1 tok-a-si	|	tokk-u-si
	PL2 tokk-a-psi	|	tokk-u-psi
	PL3 tok-a	|	tok-u
This is my verb paradigm in Ivka.
It is complex enough, but it is very regular. The only markers are Person suffix, Control marker vowel, and consonant gradation.
How could I make it less regular?
The easiest option I can see is to drop word-final /p/, having it cause word-initial gemination of the following word. It would make the present SG2 form identical to the past SG3/PL3 form, and the present SG3/PL3 form identical to the past SG2 form in isolation, but the second person form might, for example, cause a following object noun or pronoun to have a geminate initial.

You could have /aj/ and /uj/ collapse go /e/ and /i/ respectively word-finally as well, which actually might be an even easier way [:P]
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Well, for a start, that's already pretty irregular, with that (infuriating no doubt to learn!) sadistically-inverted gemination thing going on. And remember, "better" doesn't have to mean "sadistically irregular".

To make it even more irregular, you've basically got three options: suppletion, sound change, or analogy. The first uses different morphemes in some forms; the second replaces agglutinative suffixed with fusion; and the third complicates one paradigm by introducing bits from another. [and remember: irregularit

For now, let's keep things simple, and just use a little bit of simple sound change.

1. stress goes on rightmost heavy syllable, or else on rightmost syllable if none are heavy
2. u>y, o>u
3. unstressed vowels umlaut (here, u>o before a, a>e before i, and y>i before i), while stressed vowels break (here, u>ua before a, and a/y > ai/yi before i)
4. Palatalisation before e and i
5. stops assimilate to following fricatives
6. unstressed medial vowels drop, where this does not create a cluster of more than two consonants
7. unstressed final verbs reduce to schwa
8. stressed moves to first syllable
9. stressed vowels before single consonants lengthen
10. geminates simplify; /tS/ and /ts/ reduce to the fricatives
11. palatalised /k/ > /ts/, palatalised /s/ > /S/; kS > tS, but tsS > ts
12. ai, yi > ej, ij; ua > wa
13. coda /j/ strengthens to /k/
14. y > i
15. schwa > i
16. final consonants drop
17. Cw > Cr

And just a bit of analogy, to disambiguate: the noncontrol forms borrow final -a in the present singular 2nd from the control forms, to disambiguate 1st and 2nd; the nonpast control 3rd singular borrows the long vowel from the noncontrol to disambiguate from the plural; but in the noncontrol past the opposite happens (the noncontrol borrows the long vowel from the singular); and the long vowel spreads from the control past 3rd to the 2nd (or, possibly from the present 2nd).
Oh, and then they realise that the plural ends in -Si, so spread that to all forms in the noncontrol paradigm. /tu:kiSi/ and /tukiSi/ are then very similar, so marge as the long vowel form.

And finally, 18. long vowels break

Yielding:

Code: Select all

	Control		Noncontrol 
Nonpast	SG1 tauke	|	tuki
	SG2 tauka	|	tuka
	SG3 traiki 	|	twaki
	PL1 toseSi	|	twaSi
	PL2 taukeSi	|	twakiSi
	PL3 traki	|	twakiSi
	--------------------------------------------
Past	SG1 toke	|	twaki
	SG2 tauka	|	twaki
	SG3 tauka 	|	twaki
	PL1 tosi	|	tusiSi
	PL2 tokeSi	|	twakiSi
	PL3 tauka	|	twakiSi
Sufficiently irregular?
Last edited by Salmoneus on 21 Mar 2020 16:28, edited 4 times in total.
User avatar
Pabappa
greek
greek
Posts: 577
Joined: 18 Nov 2017 02:41

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Pabappa »

if it were me:

/aj uj/ merge as /e/, so the 1st person singular goes to /tokke~toke/ in present and /toke~tokke/ in past. then by analogy it becomes /toke/ in present and /tokke/ in past.

Im assuming by control you mean whether an action was voluntary? if so, then the 1st person singular no longer has that distinction, and so the speakers will use the plurals for the non-control forms, and thus everybody will say the equivalent of "oh no! we just dropped your phone!!" instead of explicitly taking the blame on their own.

but this is just a wild idea. the way you laid it out so regularly to begin with makes me think you would want the volition distinction, if thats what it is, to stay. in which case i think phonological changes, rather than semantic changes, are the best way forward. e.g. just from the looks of it i like Sal's idea, which wasnt on the screen when i started typing this post.
Kavunupupis, šiŋuputata.
When I see you pointing at me, I know I'm in trouble. (Play)
User avatar
Omzinesý
mongolian
mongolian
Posts: 4079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Omzinesý »

Thank you all!

Yes, the conjugation is not simple, but it can be described by few rules.

All of your answers are basically diachronic.
This lang is just synchrinic (and borrows very much from Northern Saami). But maybe there could be an old past suffix that, beside causing consonant gradation, merges with the person markers. Some synchretism could de well.

@Pabappa
Control is not just volitionality. I got it from a Salishan language. It is described in Ivka thread.
My meta-thread: viewtopic.php?f=6&t=5760
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Omzinesý wrote: 23 Mar 2020 15:37
All of your answers are basically diachronic.
This lang is just synchrinic
Well, if you're not interested in a diachronic/naturalistic language, then you can do whatever you feel like, can't you? Why ask us for ideas? Why not just, I don't know, roll dice?
User avatar
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by sangi39 »

Hmmmm, yeah, if you're not worried too much about diachronics (I was assuming that you meant "how would you take this, and make it more irregular in the future), then you can do more or less anything you want in terms of added irregularity. Like, maybe the first person forms follow their own distinct pattern, or maybe the control non-past forms take a different set of person/number suffixes, or perhaps the past tense don't distinguish person, but instead mark animacy or gender (to steal from Russian).
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
yangfiretiger121
sinic
sinic
Posts: 337
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 »

My setting's Phoenixfolk language has [o̞ ɯ] and an extensive set of palatalizations. Are [fʲo̞ fʲɯ → fø̞ fy], for example, plausible?
Alien conlangs (Font may be needed for Vai symbols)
Post Reply