Conlang relay [relocated] (aka "The Cursed Relay")

[Basilius]

Yawurenyi in the world of Akana?! A frightening scenario :)

I am curious about Zwirwi. Really!

[Radius Solis]

Yawurenyi in the world of Akana?! A frightening scenario

I am curious about Zwirwi. Really!

No actual Yawurenyi, of course. I have seen nothing of the language, ever, beyond its name and the other two names that are in Corumayas' signature. Without knowing anything else about it, these three words define a clear and strongly-flavored phonetic aesthetic, and I had the thought to see what Proto-Isles might look like if moved in that direction. (I'm always playing with things like this.)

But then I noticed that what I would do, was nicely compatible with the first half of Zele's sound change history - so that immediately suggested a sister language. The first half of mine, which is comparable to Zele's, goes like this:

V1 > V2 / _ħV2
haplology reducing double reduplications to single reduplications
d > l / except initially
j > Z / _i
g > j / _i
g > h
sporadic medial syncopation
p t k > v d g / medially except after /s ħ ?/
s > S
ħ > s / _[p t ts k]
ħ ? > h / _[m n l w j], #_
ts dz > s z
ħ > R

(subject to revision)

The second half has various vowel and diphthong changes, merger of /l/ and /R/ into a new /r/, creation of some geminates, some vowel epenthesis, and glottal stop deletion - among other things I haven't finished deciding on yet, like what happens to the accent system.

[Basilius]

Thanks!

The second half has various vowel and diphthong changes, merger of /l/ and /R/ into a new /r/, [...]

I don't think I've ever seen [r] ← [R]... it's usually the other way round.

[the duke of nuke]

Awesome stuff, awesome indeed Zeluzhia is starting to emerge into the light!

I agree with Basilius that /R/ > /r/ seems a little odd; wouldn't /l R/ be more likely to simply merge as /R?

More significantly, do we feel that Zele needs a new description? And if so, how much can it be changed from the current version? I can certainly fire up the SCA and fill in the lexicon, but if there's a PI lexicon in .csv form or similar that would speed matters up a bit - does one exist?

[Basilius]

More significantly, do we feel that Zele needs a new description?

I think someone should PM brandrinn. I would really appreciate his comments on the origin of the relativizing go-, for example. Also, on the history of conjugations (some forms seem to be quirky, phonetically). Or any other interesting stuff he may keep the records of, or remember :)

And if so, how much can it be changed from the current version?

I'd prefer it to be conservative, i. e. with stuff mostly added rather than removed/modified.

That is, we might treat the current version as any other incomplete description; some pieces may need reinterpretation (e. g. I doubt that Zele "passive" should be properly termed so) or completion, but the material should be treated as quoted correctly (unless we find obvious typos or somesuch).

In fact, I've been curious about some bits since the first time I saw the description. For example, this passage in one of the sample texts (with PI glosses):

Code: Select all

qúquq-piwta{q>ms} háy qí
“hufusoas re i. 
You are calling not for my benefit,

piwta{q>ms} túq tuyyih-uys
pusoas tsu tsezhires.”
you're calling so that you can eat!'

The second part looks easy, and means litteraly '[You] call your eat(ing)'. By analogy, it is tempting to analyze the first one as '[You] don't call my good'. However, it seems that its syntax is different, for i 'good' doesn't carry the accusative marker. In fact, it seems that the adjectival stem i is construed adverbially here: '[You] don't call [...] good-ly'; and then it is tempting to analyze re 'I' as its argument: '[You] don't call (for-me (good-ly))'.

In other words, I suppose a development by analogy that altered the syntax of adjectival stems when used adverbially.

Substantive use:

qí(-yi) 'good = what is good, goodness'
háy qí(-yi) 'My good = what is good for me'

Adverbial use:

qí 'good-ly = in a good manner'
háy qí 'in a manner that is good for me' ← the innovated construction

It seems that the description itself doesn't mention such constructions, but I'd like to preserve pieces like this (i. e. find an interpretation for them) rather than correct them.

I can certainly fire up the SCA and fill in the lexicon, but if there's a PI lexicon in .csv form or similar that would speed matters up a bit - does one exist?

I think the wiki code should not be difficult to convert to .csv by global search-and-replaces :)

I've e-mailed you an RTF version (with table) which I made from what Ran's original description had (before wikification) - it must be easier to convert; I haven't checked if it's identical to the current one.

[Radius Solis]

In a small phonology there's much more flexibility for big sudden POA changes than in a larger phonology, especially with sonorant consonants. Given this, and the auditory similarity of [r] and [R], and the likelihood that contact languages have only a coronal rhotic, and that there are no other uvular consonants to reinforce it, I really don't think R > r should be a problem.

[dunomapuka]

More significantly, do we feel that Zele needs a new description? And if so, how much can it be changed from the current version? I can certainly fire up the SCA and fill in the lexicon, but if there's a PI lexicon in .csv form or similar that would speed matters up a bit - does one exist?

To make this easier for us, I'll be putting the Zele grammar up on the wiki in its current form, with some minimal reformatting.

[Mbwa]

I know I am behind on this. I haven't forgotten, but I don't think there is a rush for a Lukpanic language right now, is there?

[Radius Solis]

I would really appreciate his comments on the origin of the relativizing go-, for example.

I may have figured this out.

It looks like go- is a prefix on the modified noun, while relative clauses precede the noun, so the order is RC go-NOUN. Since it goes between the two, it may have originally started as part of the relative clause instead of part of the nominal system. And I think I know where: among the clause-final discourse particles of Proto-Isles is a *kiw which marks "assertion". /go/ would be one of several possible outcomes of the particle, depending on the verb, but it could have been generalized to all.

Of course assertion-marking and subordination-marking do not seem very similar, but honestly, all of us participants of the original game pulled tricks like this without explaining them, due to being pressed for time. And it's not completely implausible anyway, just interesting.

[the duke of nuke]

I've had a look at the noun declensions and I think I understand them pretty well now. The nominative and accusative are straightforwardly inherited from Proto-Isles and the genitive is from the root (adjectival) form. The locative, dative, and ablative are derived from the postpositions *qas, háq, and *hapaq, which I think Basilius had already spotted.
It seems that the apocope and possibly also medial voicing processes are a little more complicated that they first appeared. The former apparently depends on the length of a word.

It turns out that to decline a noun fully you actually need to know both the nominative and the genitive, although if you know the nominative you know the declension type (as described in the original grammar). I'm going to write a new description of the nominal morphology with examples, and then start looking at verbal morphology.

[Basilius]

In a small phonology there's much more flexibility for big sudden POA changes than in a larger phonology, especially with sonorant consonants. Given this, and the auditory similarity of [r] and [R], and the likelihood that contact languages have only a coronal rhotic, and that there are no other uvular consonants to reinforce it, I really don't think R > r should be a problem. :)

I'm not sure. For example, looking at the allophony of a language like Hawaiian, I notice two things:

- allophones that would look like the result of a whole chain of SC's in a less minimalistic phonology;
- phones that look like optional retention of foreign pronunciation in loan words.

The latter hardly has much to do with historical SC's, while the former doesn't *revert* the direction of more familiar SC's...

I would really appreciate his comments on the origin of the relativizing go-, for example.

I may have figured this out.

It looks like go- is a prefix on the modified noun, while relative clauses precede the noun, so the order is RC go-NOUN. Since it goes between the two, it may have originally started as part of the relative clause instead of part of the nominal system. And I think I know where: among the clause-final discourse particles of Proto-Isles is a *kiw which marks "assertion". /go/ would be one of several possible outcomes of the particle, depending on the verb, but it could have been generalized to all.

Of course assertion-marking and subordination-marking do not seem very similar, but honestly, all of us participants of the original game pulled tricks like this without explaining them, due to being pressed for time. And it's not completely implausible anyway, just interesting.

Very interesting idea indeed!

BTW, one of the sample texts features a discourse particle, ko, which I tentatively identified with PI káws, but PI kiw is certainly a better candidate.

[Basilius]

A bit unexpectedly, I started an Epic Treatise on the Historical Morphology and Phonology of Zele...

Join the fun :)

(It's all because boy #12 wikified the grammar, so I re-read it, etc... you know how such things happen...)

[dunomapuka]

I know I am behind on this. I haven't forgotten, but I don't think there is a rush for a Lukpanic language right now, is there?

Nope. Have you figured out where you want to put your language?

[Mbwa]

Doanu.

[the duke of nuke]

Alces has been developing a new language (Wendoth) that he wanted to put in Akana and my suggestion was putting them in Tuysáfa.
Western Tuysáfa looks good because:
- it has only one language described, which is Proto-Isles
- there should be plenty of languages from the Mediundic and possibly Primundic families, as well as Proto-Isles' Ultimundic relatives
- it later becomes significant with the expansion of maritime powers to the west and south
- Wendoth shares some typological similarities with Proto-Isles, so it's a good candidate for a neighbour with which areal features were shared
- Alces wanted his language family to split up after the speakers' homeland became a desert, and central western Tuysáfa is a fairly suitable environment

We're discussing (in this thread) the scenario that the speakers of Wendoth were the invaders that displaced the speakers of Proto-Isles. Here's what Alces says in the matter:

Region 2 on your climate map looks like an ideal place, then.

So the scenario could be something like:
- -3000 - Pre-Wendoth is spoken as described in the grammar.
- -2500 - Wendoth speakers take up agriculture from the east. By this time at least they are in Region 2, i.e. central Tuysáfa west of the mountains.
- -2100 - A very harsh century, with frequent droughts, begins. Time of Wendoth as described in the grammar.
- -2000 - Most Wendoth speakers have left their original home due to the droughts, and perhaps their agriculture contributes to the problem too.
- -1900 - Wendoth speakers have reached the southwestern tip of Tuysáfa encouraging the Proto-Isles speakers to leave their home.

I don't see any problems with this, apart from possibly some timing details. What do the other members of the project think?

[Basilius]

In case there are people who haven't been watching...

It appears that the Invaders who forced the speakers of Proto-Isles to leave Tuysáfa now have a name: the Wendoth.

Now you are warned. It's not too late yet to post your objections, corrections, amendments, etc. :)

[Radius Solis]

Awesome. It will be interesting to see what Wendoth vocabulary the Islesians carry with them and how they borrow the great variety of sounds in Wendoth.

In other Isles news, the sound changes and name for Kiizwaye (= "mountain Zele") are finalized and the morphology is looking pretty solid. After some indecision because I thought it was boring at first, I decided the most realistic nominal inflection paradigm, given my SCs, would result in four declensions directly reflecting the four possible coda consonants of proto-Isles (0, *s, *h, *q). Two of which make extensive use of ablaut - as do all the verbs. Beyond that, it's surprising how much the results still resemble Zele, even though 2/3 of the sound changes are not shared. Similar sound inventories and spelling rules have a lot to do with it, but the morphology works out to be quite similar too, even after liberal applications of analogy and sandhi in hopes of making it more distinct. Nevertheless I think I'll run with it, though I do not expect to put up more than a minimal sketch comparable in size to Zele's.

[Basilius]

Awesome. It will be interesting to see what Wendoth vocabulary the Islesians carry with them and how they borrow the great variety of sounds in Wendoth.

In the first place, we need to understand which of the Wendoth dialects contacted with PI, and what that dialect looked like.

For me the most fascinating feature of Wendoth is the enormous potential diversity of its daughters: the three vowel phonations (lax/creaky/breathy) differently landing on the adjacent consonants can produce tons of different forms basing on perfectly regular SC's alone.

In other Isles news, the sound changes and name for Kiizwaye (= "mountain Zele") are finalized and the morphology is looking pretty solid. After some indecision because I thought it was boring at first, I decided the most realistic nominal inflection paradigm, given my SCs, would result in four declensions directly reflecting the four possible coda consonants of proto-Isles (0, *s, *h, *q). Two of which make extensive use of ablaut - as do all the verbs. Beyond that, it's surprising how much the results still resemble Zele, even though 2/3 of the sound changes are not shared. Similar sound inventories and spelling rules have a lot to do with it, but the morphology works out to be quite similar too, even after liberal applications of analogy and sandhi in hopes of making it more distinct. Nevertheless I think I'll run with it, though I do not expect to put up more than a minimal sketch comparable in size to Zele's.

I replied to this part on Akana Forum.

[Corumayas]

Siŋmeasita Lukpanic is up. There'll probably be more later: I'd like to describe the morphophonology in more detail, come up with some dialectal words/word usages, and maybe do a sample text too.

[Mbwa]

Okay, I have really forgot about/put off this thing for a while. Honestly, I don't know if I will work on it again. It's no big deal, I bet, just a little Lukpanic language, but I feel kinda bad about giving up. If anyone wants, for whatever reason, I could post the sound changes I have so far, but it's not much.

[Corumayas]

Okay, I have really forgot about/put off this thing for a while. Honestly, I don't know if I will work on it again. It's no big deal, I bet, just a little Lukpanic language, but I feel kinda bad about giving up. If anyone wants, for whatever reason, I could post the sound changes I have so far, but it's not much.

It's nothing you should feel bad about; Akana has conlangs that have been unfinished/on hold for years.

But if you're not going to do any more with it, I'd say you might as well post what you've got. Maybe someone else will be inspired to carry on with it at some point.

[roninbodhisattva]

I've tried to drop in on this project so many times in the past, and just have not been successful. But I still would like to give it another go. I really don't know where to start, and how to do that, though.

[Cedh]

I've tried to drop in on this project so many times in the past, and just have not been successful. But I still would like to give it another go. I really don't know where to start, and how to do that, though.

More than anything else you need to be willing to say or write something. There's plenty of space on Akana, and even in the Edastean sphere there's still much room for conworlding. If you come up with something that's realistic and interesting, and does not conflict with something that has been established already, the odds are pretty good that it will fit in. If you don't know something or don't understand a specific situation from what is written on the wiki, just ask! If you want to make a daughterlang from one of the existing languages, choose a parent and go ahead (except that Ndak Ta and Proto-Western should probably not have another direct daughter because there are so many of them already). If you have an interesting conlang or conculture of your own that you would like to place in Akana, we're likely to find a suitable spot for it. And in case you want to help one of the "senior participants" with a shelved project, that should be possible too.

Which part of Akana interests you most? Language, culture, history, economics, geography? A specific region, time period, or language family?

[Cedh]

I'd like to announce that I've mostly completed the morpho(phono)logy description of my long-awaited vowel-harmonic Fáralo daughterlang spoken on the eastern Ishtmus: Cəssın!

[Basilius]

Very impressive!

Is linni 'country' an Affanonic loan? (If not, it's an interesting coincidence...)