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

A forum for all topics related to constructed languages
User avatar
sangi39
moderator
moderator
Posts: 3328
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

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

Post by sangi39 » 22 Sep 2019 20:10

spanick wrote:
22 Sep 2019 17:56
When starting with a proto language, how do you typically go about developing case marking? Most cases seem pretty intuitive to me when derived from prepositions but there’s a few I’m not sure about, specifically Nominative, Accusative, Ergative, and Absolutive.

To be clear, I know that I may not need to make some of those, but languages like Latin do have nominative case marking, so I’m just curious about how that comes to be. Those four cases are also just the ones I’m curious about, I don’t necessarily intend to have all four in one language.

I really don’t ever make proto languages from which to derive more naturalistic conlangs, but I’m giving it a go. So, for those of you that do, how do you do it? Do you derive case markers or do you just randomly assign them in the proto lang?
Now that's a question.

On the one hand, you could just hand-wave it and say that the nominative, accusative, etc. were present in the proto-language. This is the case in Proto-Indo-European, Proto-Semitic, Proto-Uralic, I think (the locative cases in Finnish, from what I can remember are built on older locative cases in a similar manner to that seen in Tsez, but the syntactic cases come straight from Proto-Uralic), and it seems to be the case in Proto-Turkic as well (I was hoping the definite accusative case in Turkish was an innovation, but it seems to be inherited).

I tried to see if there was any thought on where, for example, the PIE nominative *-s and accusative *-m came from, but beyond "*-s might have originally have been an ergative or agentive case", it's mostly just taken as "original", i.e. it's been there so long no-one can see where it came from.

I do this with my proto-languages. Proto-Sirdic's cases are just... there, as are the cases in Proto-Skawlas.


As to where they could come from... hmmm... I think I've seen a suggestion that nominative/agentive cases could come from an older vocative case, or demonstratives, the emphasis of "it is this X that does Y" indicating agency, and then its use becomes more and more frequent. I seem to recall an idea that the nominative case might also come from deverbal nominalisers (eventually the suffix is taken to be separate from the noun, and dropped when the noun is further declined).

The accusative could come from an older dative, especially in a language that previously had secundative alignment (the recipient of a ditransitive verb is marked in the same way as the patient of a transitive verbs, as opposed to indirective languages where the patient and the theme are marked in the same way, with a distinct case for the recipient).
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: 1743
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus » 22 Sep 2019 23:04

Ergatives are often formed from genitives or instrumentals.

User avatar
Creyeditor
mongolian
mongolian
Posts: 4518
Joined: 14 Aug 2012 19:32

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

Post by Creyeditor » 23 Sep 2019 00:07

Okay, I don't have much too add, but maybe some additional perspectives.
Structural case marking can develop from differential case marking, These are situation where only some kind of arguments are marked with case. In differential object marking for example, it is often the case that only animate objects are marked with an accusative-like case. These new case markers often develop from adpositions like dative "to", allative "to" for differential object marking. For differential subject marking you sometimes find subject marked with an instrumental "with". Sometimes these are already cases but semantic and not yet structural.
The other way you can get case marking is from nominalization. Subjects of nominalized clauses are often marked the same as possessors. If these nominalizations become finite clauses, the possessor marking can stay and the subjects keep their genitive marking.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
sangi39
moderator
moderator
Posts: 3328
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

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

Post by sangi39 » 23 Sep 2019 00:27

Salmoneus wrote:
22 Sep 2019 23:04
Ergatives are often formed from genitives or instrumentals.
<immediately runs to Google>

Now that is pretty interesting!

I've found a couple of sources I need to read through when I get time*. Completely forgot about ergativity in Indo-Iranian languages, which would have been a good shout [:P]

*Areal dimensions in case syncretism: Ablatives and genitives by Michael Noonan and Elena Mihas, and Instrumental prepositions and case: Contexts of occurrence and alternations with datives by Ludovico Franco and M. Rita Manzini, look like good starts.
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.

User avatar
Pabappa
sinic
sinic
Posts: 285
Joined: 18 Nov 2017 02:41
Contact:

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

Post by Pabappa » 23 Sep 2019 00:53

this may not be a helpful reply but it might be interesting to see *why* and how each case develops into an ergative. e.g. i can see 2 paths for the genitive...
1) " boy-GEN (his) knife" starts out meaning "the boy's knife", but somehow the speakers start generalizing the GEN as the form you use whenver someone possesses something, and then you get "boy-GEN (his) knife carry" and from there it ceases to be solely a genitive marker.
2) the passive route .... "boy-GEN sleep-3P" ... the sleeping of the boy, initially used only for patient/experience type verbs, butlater extended to all types of verbs, so e.g. "boy-GEN girl-ACC kiss-3P", means the boy is kissing the girl, and this is an active verb.

For instrumental i think its just same as saying in English "oh! i just got bitten by a horsefly!" ... or more likely for inanimate objects eg. "my chair just crushed my toe". this would be in a language that previously only allowed inanimate objects to interact with humans by means of indirect cases like the instrumental.
Sorry guys, this one has the worst sting.

User avatar
spanick
roman
roman
Posts: 1104
Joined: 11 May 2017 01:47
Location: California

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

Post by spanick » 23 Sep 2019 01:56

Thank you, everyone, for the responses. They’re very helpful! Not only is it good to know where these cases come from if I decide to go these route, but it’s also nice to know that it’s ok for case markings to simply already exist in the Proto language. This question has somewhat slowed the development of this protolang but hopefully it will be all the better for it.

User avatar
spanick
roman
roman
Posts: 1104
Joined: 11 May 2017 01:47
Location: California

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

Post by spanick » 23 Sep 2019 01:57

Thank you, everyone, for the responses. They’re very helpful! Not only is it good to know where these cases come from if I decide to go these route, but it’s also nice to know that it’s ok for case markings to simply already exist in the Proto language. This question has somewhat slowed the development of this protolang but hopefully it will be all the better for it.

User avatar
Zekoslav
sinic
sinic
Posts: 313
Joined: 07 Oct 2017 16:54

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

Post by Zekoslav » 23 Sep 2019 10:17

Indo-European accusative usually has allative/illative functions as well (e.g. Latin eō Rōmam). I fancied that they might be original and that the accusative function developped later, with animate > masculine, feminine nouns but not inanimate > neuter nouns, similar to how Spanish uses a, normally a dative preposition, for personal accusatives.
Languages:
:hrv: [:D], :bih: :srb: [;)], :eng: [:D], :fra: [:|], :lat: [:(], :deu: [:'(]

A linguistics enthusiast who would like to make a conlang, but can't decide what to call what.

- Tewanian languages
- Guide to Slavic accentuation

User avatar
Omzinesý
runic
runic
Posts: 2707
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

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

Post by Omzinesý » 23 Sep 2019 18:04

Pabappa wrote:
23 Sep 2019 00:53
this may not be a helpful reply but it might be interesting to see *why* and how each case develops into an ergative. e.g. i can see 2 paths for the genitive...
1) " boy-GEN (his) knife" starts out meaning "the boy's knife", but somehow the speakers start generalizing the GEN as the form you use whenver someone possesses something, and then you get "boy-GEN (his) knife carry" and from there it ceases to be solely a genitive marker.
2) the passive route .... "boy-GEN sleep-3P" ... the sleeping of the boy, initially used only for patient/experience type verbs, butlater extended to all types of verbs, so e.g. "boy-GEN girl-ACC kiss-3P", means the boy is kissing the girl, and this is an active verb.

For instrumental i think its just same as saying in English "oh! i just got bitten by a horsefly!" ... or more likely for inanimate objects eg. "my chair just crushed my toe". this would be in a language that previously only allowed inanimate objects to interact with humans by means of indirect cases like the instrumental.
I've lost the track of where the discussion begins. But the normal source of genitive-ergatives is from clausal possession.

Germanic and Romance "perfects" develop from a transitive possessive verb: "I have the dishes washed." => "I have washed the dishes."
If the language has an intransitive clausal possession construction, the same route ends up to ergatives.

Finnish doesn't have such perfects but it could.
"Minulla on tiskit tiskattuna."
SG1-ADESS COP dishes wash-PASS.PTCP
'I have the dishes washed.'
=>
*"Minulla on tiskattuna tiskit."
* SG1-ERG cop dishes wash-PASS.PTCP
*'I have washed the dishes.'

Salmoneus
MVP
MVP
Posts: 1743
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus » 23 Sep 2019 21:58

In many language, the ergative-genitive appears to come from nominalisations. That is, the "verb" used to be, in effect, a gerund.

So, to say "I don't like you because you ate my cat", you effectively say "I don't like you because (of) your eating my cat". Gerunds usually have ergative syntax cross-linguistically, so these clauses become ergative. Then the ergative construction escapes out of these limited circumstances and becomes the default for a whole range of associated situations (eg, the past tense, or conditionals, etc). In some languages it then becomes the default in all clauses.

User avatar
Void
sinic
sinic
Posts: 283
Joined: 15 Aug 2016 16:15

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

Post by Void » 26 Sep 2019 21:36

How do I go about creating a naturalistic vigesimal base numeral system, with a vestigial duodecimal base system in place on numbers lower than, say, 100?

Right now, I have a relatively primitive number set for the numbers 1 to 12 (7, 8, 9, and 11 are compounds, but w/e), 20, 100, 144, and 1000. The duodecimal and vigesimal tens (i.e. 20, 60, 80; 24, 36, 48, 60, 72, 84, 96, 108, 120, 132, 144) are also "primitive" (formed through special constructs).

So most numbers smaller than 200 are constructed with a vigesimal system, e.g. 150 is (144 + 6) or 50 is (48 + 2); numbers between 200 and 1000 are constructed using a mixed system, i.e. the hundred-set (240, 360, 480, ..., 960) of the duodecimal system is used, as well as any number that can be formed by adding the numbers 1 - 12 (I'm not sure how to word this one, but basically, 240 - 252, 360 - 362, 480 - 492, ..., 960 - 972); otherwise, the vigesimal system is used, e.g. 253 is (200 + 40 + 13) rather than (252 + 1).

Now I have a feeling that this system is just a clusterfuck and would never actually occur in a natural language, so my question(s) is: how can I make this more naturalistic? And if the answer is "you can't," then should I just stick to a true duodecimal (or vigesimal) base, and leave it at that?
Rjimâyin

User avatar
sangi39
moderator
moderator
Posts: 3328
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

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

Post by sangi39 » 26 Sep 2019 22:20

Void wrote:
26 Sep 2019 21:36
How do I go about creating a naturalistic vigesimal base numeral system, with a vestigial duodecimal base system in place on numbers lower than, say, 100?

Right now, I have a relatively primitive number set for the numbers 1 to 12 (7, 8, 9, and 11 are compounds, but w/e), 20, 100, 144, and 1000. The duodecimal and vigesimal tens (i.e. 20, 60, 80; 24, 36, 48, 60, 72, 84, 96, 108, 120, 132, 144) are also "primitive" (formed through special constructs).

So most numbers smaller than 200 are constructed with a vigesimal system, e.g. 150 is (144 + 6) or 50 is (48 + 2); numbers between 200 and 1000 are constructed using a mixed system, i.e. the hundred-set (240, 360, 480, ..., 960) of the duodecimal system is used, as well as any number that can be formed by adding the numbers 1 - 12 (I'm not sure how to word this one, but basically, 240 - 252, 360 - 362, 480 - 492, ..., 960 - 972); otherwise, the vigesimal system is used, e.g. 253 is (200 + 40 + 13) rather than (252 + 1).

Now I have a feeling that this system is just a clusterfuck and would never actually occur in a natural language, so my question(s) is: how can I make this more naturalistic? And if the answer is "you can't," then should I just stick to a true duodecimal (or vigesimal) base, and leave it at that?
Hmmmmm, I think what you could do is:

1-12 are unique
13-19 are formed through addition
20, and multiples thereof up to 220 are formed through multiplication
240 (12*20), 2880 (12*12*20), 34560 (12*12*12*20), etc. are higher bases or

I would expect to see an extremely mixed system beyond 240, but I suppose you could have unique numbers for pure multiples of 12 (24, 36, etc, and so on), but I'd guess at that point, the language is either going to end up base-12 or people are just going to give up on using those numbers.

You could probably have 400 as the next highest base up instead of 240, so something like (12+7)*20, but that's a bit less mixed than you seemed to be looking for.
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: 268
Joined: 17 Jun 2018 03:04

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

Post by yangfiretiger121 » 28 Sep 2019 18:22

Vlürch wrote:
22 Sep 2019 15:53
yangfiretiger121 wrote:
22 Sep 2019 15:19
Naturalistically speaking, is Bäïkal, which was meant for religious communications at first, a dialect of Common or its own language due to the changes?
That's entirely up to you within the context of the conworld/conculture.

A lot of people consider the Sinitic languages to be dialects of the same language even though they have very little mutual intelligibility due to different phonologies, lexical differences and even some different grammar, etc. while a lot of people consider Serbian, Croatian, Bosnian and Montenegrin separate languages even though they have a high degree of mutual intelligibility. Some Finns consider Karelian merely a dialect of Finnish, and they can be mutually intelligible to a high degree (because there's a dialect continuum (or something like that) and the Karelian language is being absorbed into the Karelian dialects of Finnish (where it isn't being replaced by Russian, that is)), but it can also be totally unintelligible: there was once some documentary on TV about old people in Russian Karelia without subtitles, and at least I only caught a word here and another there...

So, basically, it's an entirely political distinction. You could have something equivalent to the real world controversies regarding dialect vs language within your conworld/conculture if you want. Usually I'd say it's nationalists that want everything to be lumped together as one language while less nationalistically inclined people don't really care or think it should be up to the speakers to decide how they want their language/dialect to be seen. One word that's sometimes used in ambiguous cases is isolect.
Okay. Thanks. That makes sense.

As the Skaran Empire encompassed the home system's four habitable planets by conquering one of the others and reaching peace agreements with the other two, is it natural for the conquered planet's language to be completely excluded from the stellar language's formation?
Alien conlangs (Font may be needed for Vai symbols)

User avatar
Vlürch
sinic
sinic
Posts: 300
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

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

Post by Vlürch » 29 Sep 2019 15:52

yangfiretiger121 wrote:
28 Sep 2019 18:22
As the Skaran Empire encompassed the home system's four habitable planets by conquering one of the others and reaching peace agreements with the other two, is it natural for the conquered planet's language to be completely excluded from the stellar language's formation?
Considering we haven't yet had any interplanetary wars or even extraterrestrial colonies, I don't think that kind of circumstances can really be predicted realistically. As far as I know, "like on Earth, but with bigger distances so everything is exaggerated" tends to be the default way to go about it. So if the conquerors had a negative view on the conquered planet's language and excluded its speakers from positions of influence, I don't see why it wouldn't be naturalistic for its influence on a lingua franca (assuming the "staller language" is a lingua franca) to be minimal, or even completely absent since the distances between planets tend to be bigger than the distances between countries.

However, languages in the real world tend to not be immune to influences from other languages considered "inferior" by the conquerors; for example, both Swedish and Russian have some loanwords from Finnish (although much less than Finnish has from them), Persian and Arabic have loanwords from Turkic languages (but again, the (southern) Turkic languages have more loanwords from Persian and Arabic than the other way around), English has loanwords from Hindustani (but again, Hindustani has more from English), etc. etc. etc.

If the conquerors had very little contact with the conquered (and vice versa), then it would logically seem naturalistic for the languages to not influence each other a lot. If the conquered planet was just a vassal or something and not fully integrated into the empire and they mostly kept to themselves (by choice or by force), then it would make more sense for their language to not influence the lingua franca. If the lingua franca is a conlang even conically, then it having no influences at all from the conquered planet's language would make sense if the conquerors have a negative view on the language.

As for how entire planets could have only one language on them, that's another matter entirely... but I guess it's possible, even if almost certainly a single language spoken across the entire surface of even a small planet would at first develop into a dialect continuum and, with time, the dialects would become mutually unintelligible and their speakers would consider them separate languages. Maybe if they're extremely advanced in terms of technology and everyone is very networked and there are no internal conflicts, it could be plausible?

Salmoneus
MVP
MVP
Posts: 1743
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus » 29 Sep 2019 23:40

yangfiretiger121 wrote:
28 Sep 2019 18:22
As the Skaran Empire encompassed the home system's four habitable planets by conquering one of the others and reaching peace agreements with the other two, is it natural for the conquered planet's language to be completely excluded from the stellar language's formation?
History is more complicated than this.


In the rise of Rome, Latin displaced the languages of three 'civilised' cultures: Etruscan, Punic, and Greek, in each case through military conquest.

Etruscan gave Latin many loanwords, but otherwise rapidly became extinct.

Punic continued to be spoken in rural areas of its homeland, but otherwise became extinct very quickly, leaving very little trace on Latin, if any.

Greek, however, continued to be the lingua franca of half the Empire (and eventually all the Empire once the west was lost), gave many loanwords to Latin, and was spoken by all educated native Latin-speakers as a second language.

EDIT: Is this a Farscape fanfiction, btw? Or a Dr Who fanfiction?

yangfiretiger121
sinic
sinic
Posts: 268
Joined: 17 Jun 2018 03:04

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

Post by yangfiretiger121 » 30 Sep 2019 11:28

Thanks guys. The "circumstellar language" is meant to be a lingua franca to connect four other planetary lingua francas. Thus, I may use a "sphere of influence"-type model, meaning the closer to the home planet another planet is; the more (still relatively little) influence its lingua franca has on Imperial Creole. This way, the home system's Standard dialect remains relatively stable over time while the more distant dialects may grow to sound like different languages altogether as with real world Yi in China.

And, Sal, it's a roleplay setting, not a fanfic.
Alien conlangs (Font may be needed for Vai symbols)

User avatar
Void
sinic
sinic
Posts: 283
Joined: 15 Aug 2016 16:15

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

Post by Void » 30 Sep 2019 12:06

sangi39 wrote:
26 Sep 2019 22:20
Hmmmmm, I think what you could do is:

1-12 are unique
13-19 are formed through addition
20, and multiples thereof up to 220 are formed through multiplication
240 (12*20), 2880 (12*12*20), 34560 (12*12*12*20), etc. are higher bases or

I would expect to see an extremely mixed system beyond 240, but I suppose you could have unique numbers for pure multiples of 12 (24, 36, etc, and so on), but I'd guess at that point, the language is either going to end up base-12 or people are just going to give up on using those numbers.

You could probably have 400 as the next highest base up instead of 240, so something like (12+7)*20, but that's a bit less mixed than you seemed to be looking for.
I think I'll just stick to a duodecimal system, then. I'm no mathematician and so developing a complex numerical system isn't really my number one priority, now that I think about it. I'll probably get sick of calculating complex numbers in the future and instead of scrapping the system, I'll just scrap the language - better avert that disaster now.
Rjimâyin

Salmoneus
MVP
MVP
Posts: 1743
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus » 30 Sep 2019 12:39

yangfiretiger121 wrote:
30 Sep 2019 11:28
And, Sal, it's a roleplay setting, not a fanfic.
It's just that your Skaran Empire automatically think of the Scarran Empire and/or Skaro.

yangfiretiger121
sinic
sinic
Posts: 268
Joined: 17 Jun 2018 03:04

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

Post by yangfiretiger121 » 30 Sep 2019 17:04

Salmoneus wrote:
30 Sep 2019 12:39
yangfiretiger121 wrote:
30 Sep 2019 11:28
And, Sal, it's a roleplay setting, not a fanfic.
It's just that your Skaran Empire automatically think of the Scarran Empire and/or Skaro.
Ah. Okay.

Additionally, I have a feeling I'm planning too much of the language because Old Common (Fmënïk [ˈfmɛ.nɪk], FMNK) and Church Common (Bäïkal [ˈbɑɪ̯.kal], dialect code BKL) were spoken when plate armour was around in the setting and were recombined into Early Middle Common (Knäüsäv [ˈknɑʊ̯.ʃɑɱ]; KNSV) 185 years ago (setting equivalent of 1834). Considering the aforementioned, (a) has the language had enough time to develop into Late Middle Common (Ypnæq [ˈy.pnæ̰s͡k]; YPNQ) for a modern-equivalent version of the setting, and (b) will the language have had the time to develop into New Common (think Modern/New English) by the setting's 26th century for Starfinder? The spellings and pronunciations for the languages' names are as they were when the languages were spoken.
Alien conlangs (Font may be needed for Vai symbols)

holbuzvala
sinic
sinic
Posts: 216
Joined: 01 Jan 2017 14:03

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

Post by holbuzvala » 01 Oct 2019 08:35

would it be strange to have a vowel inventory wherein only some vowels can be long?

My inventory is as follows:

Code: Select all

SHORT
i y    u
 e ø   o
  a
 
LONG
iː    uː
   aː
   
DIPHTH
ai
au
ay

Post Reply