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

[Legion]

Cool stuff, Radius!

[Radius Solis]

Sorry for the wall-o-text, but I think it's a good idea to put NT on a solid foundation. Don't read if this doesn't interest you.



I've been back-deriving some proto-Talo-Edastean words (just the original nouns so far) to see how it works out, and various issues have arisen. It's not TOO bad, although it's quickly becoming clear that pTE was even uglier than Ndak Ta. And there's far, far too much /N\/ all over the place compared to the frequency of /N/, and there's too little NT /k kw w/ for what you'd expect from their reflecting mergers of multiple pTE sounds. But no matter... none of that is actually unrealistic. And perhaps the relative paucity of both /q/ and /k/ contributed to their loss of contrast.

Going by the previously listed changes only, a good 2/3 or so of the words remain the same as in NT. That's probably fine, given the shortish time period involved. But now I'd like to address a few other little issues...

First, the syllabic nasals should be dealt with. Sal's Andagg strongly implies that there were vowels here in pTE that dropped in NT to create the syllabic nasals. As well, stop+nasal clusters are a bit of a mess. Voiceless stops were supposed to be pronounced [?] before nasals, and Naidda reflects this, but Faralo and Adata do not. So I think this should be a sound change occuring late in NT's history, which did not spread to all corners of the Ndak empire. Lastly, voiced stops assimilated before nasals into geminate nasals, as in the compound runnak from rud-naka, but only medially; so this change will have to come before the change that created the word-final, syllabic versions of these clusters.

Henceforth:

7. *dn, *bm, and *gN assimilate to [n:], [m:], and [N:]

8. Unstressed vowels between an intervocalic stop and a final nasal were lost, after which POA assimilation applied and the nasal became syllabic. I don't see any counterexamples in the list, but some may turn up (anything ending in, e.g., VtVn) - if they do I'll have to modify this.

9. Unstressed initial vowels were syncopated, before intervocalic consonants and before nasal+stop clusters. Initial nasals then became syllabic before stops. I think this can account for all the wordforms seen:

mbontai "seed" < *ambóntai (while untai < untai)
baus "ox" < *abáus (there should be some cases like this)
oslók "forget" < *oslók (cluster /sl/ blocks syncope)
enánda "feel" < *xenánda (initial consonant blocks syncope - therefore this change must have been before x-loss, and this entire set of changes 7 - 9 needs to occur before change 1 in the previous list)

10. Voiceless stops became glottal stops before nasals, but only in Kasadgad and the lower Aiwa (during the Akan dynasty of the Ndak empire). Thus it could be seen as a change in late NT or in early Naidda, either works.





Side note: I've only just now, after all this time, noticed a certain infelicity in the original Ndak Ta lexicon: "spear" is given as tselô. NT did not have nasal versions of /o u/. Adata's reflex treated it like any other final nasal vowel (adding -n), so I can't ignore it; I'm just going to call it an irregular word. Odd though it is, occasionally a language will have a "phoneme" that appears in only one or two words, typically borrowings or ideophones. (IIRC Finnish has geminate /h:/ in exactly one word; English similarly has the initial cluster /vr/ only in "vroom".) So now we need a nearby language with /o~/ for this to be a loan from! :8


EDIT:

I have decided to add in a sporadic change: fronting of qw/kw to p. Only in a half dozen or so words. But there had to be more dorsal sounds in pTE than this. The entire original NT verb list has a mere ten instances of /k(w)/.

[dunomapuka]

What are your thoughts about Tlaliolz at this time? Also, do you think this will gel nicely with Proto-Xoronic? Or will a clear picture of Proto-Macro-Edastean elude us?

Also, I would like your Naidda sound changes plz.

[Radius Solis]

PME... I don't know. That's going to be a tough one to try to reconstruct, if it's even possible. Proto-Xoronic, as given, really fits the bill for what I first suggested: something where the relationship isn't hard to spot but where pinning down more than a few (of maybe even any) specific sound correspondences would be difficult.

I'll write up what I can on Tlaliolz soon and post it; I will definitely need help constructing a full sound change set that gets the right results, if anyone wants there to be one.

Naidda sound changes, oy. There's been a lot of dialect interference in Naidda's history: words borrowed or influenced from other dialects where a slightly or moderately different set of changes applied, or applied in a different order. And the language hasn't been standardized or codified since Ndak times, so there's been no strongly prestigious dialect to dominate the rest. So there are multiple forms present for a number of words, even within the same town. The dictionary is designed to reflect this by representing only the most common forms for the Delta region as a whole rather than a definitive word list for any one location. For a very few words I even intermixed some early Faralo sound changes!

The practical upshot: I'll post what I used for the basis of Delta Naidda, in a few hours when I have time to make it digestible and a bit prettier, but be warned that maybe only 2/3 of the dictionary can be reproduced with it exactly.

_________________________________________


Naidda SCs

The list, now in its nth revision, is still incurably messy. But my notes and .sc file are way worse, to the point I'm not sure I haven't missed a minor change or two here. Also beware chronicity: the first three sections should be taken as running roughly concurrently (this creates no conflict), with the final section taking place afterwards.


1. medial
in this case "medial" means that there's at least one vowel later in the word

mb nd ŋg v ɾ ɲ / medially
mp nt ŋk ŋkw mv nʒ ɲ ɲ / medially
b d g v ɾ j / V_V
p t k b d ʔ / V_V
p t k v ʒ j / in medial clusters, except after s
ts ʃ / medially
w bw mbw kw 0 vw mw gw / medially
g j / d_
iV uV jV wV / _
sj dj tj lj rj ʃ dʒ tʃ ʎ ʒ / _
p t k tʃ b d g dʒ / V_V
b d g gw dʒ p t k kw tʃ / s_
gw > w


2. initial

mb nd ŋg m n ŋ / #_
mp nt ŋk p t k / #_
ts tʃ / #_
bw mbw v mw / #_
ŋ ɲ / #_
V > 0 / #_sC

3. final

pm tn kŋ tsn ʔm= ʔn= ʔŋ= ʔn= / _#
bm dn gŋ m n ŋ / _#
[+obstruent] 0 / _#


4. everywhere

stress regularizes to 1st syllable

i e a o u ai au e ɛ a o o i u / in open syllables when not nasalized (in NT)
i e a o u ai au ɪ ɪ ə o u i ɒ / in closed syllables or nasalized (in NT)

ə ɒ / stressed
a ɪ ɒ ə / unstressed
u o / _#

V 0 / if 2nd syllable in a 3+ syllable word, most of the time, unless stressed

ŋw w / _
r g ŋ ʒ ʒ ɲ / _[j i I e] or _
s ʔ / V_V and V_[+nasal]
ŋ n / except finally
C: V: C V / _

j i / V_
l 0 / _[r s w]

r > ɾ / except intervocalic

if anyone notices important problems or lacunae in this list, please tell me, thanks!

[Corumayas]

The original intention was something like the Gangetic plain, but more temperate, largely because snow was known:

Also note that Puoni and Naidda have words for olive and olive press, and Puoni has names for citrus fruits, appropriate for the warm climate I thought the area had. Though I could change those if needed

Ah, ok. These are somewhat in conflict, since citrus fruits don't do well if there's a frost. Possible resolutions:

-The citrus fruits Naidda speakers know about are simply grown further south.

-Lasomo gets snow but Kasca doesn't. (This might make sense if we rotate the Aiwa valley so that Lasomo is significantly further north than Kasca.)

-The Aiwa valley occasionally gets a little snow, but not much. In other words, it's right at the border between snow territory and citrus-growing territory (like northern Florida I guess).

-The climate gets warmer between Ndak Ta and Naidda times.

-Akanaran citrus fruits are more frost-resistant than their earth equivalents.

Also, as I understand it, olives require a Mediterranean climate. So either olives aren't appropriate for Kasca, or Akanaran olives grow in much wetter conditions than earth ones.

Here is a preview of my current map revision, with some tilting of Siixtaguna, some reshaping of the island coastlines, and a slight rotation of the whole continent.

Not bad. Maybe we could compare some other versions, if you have time to make some... I'd like to see one with less tilting of Siixtaguna but more rotation of the whole continent.

proto-Talo-Edastean

Nice stuff so far. (I too would like to compare it to Proto-Xoronic, but with the wiki still down that's not possible....)

Naidda SCs

I can't quite get "Naidda" from these... it seems to come out "Naizha" instead. I'm not sure if this is dialect mixing or if the changes are slightly off.


The wiki has been down for days. Does anyone know what's going on with it?

[Cedh]

Here is a preview of my current map revision, with some tilting of Siixtaguna, some reshaping of the island coastlines, and a slight rotation of the whole continent.

Not bad. Maybe we could compare some other versions, if you have time to make some... I'd like to see one with less tilting of Siixtaguna but more rotation of the whole continent.

Here you go:


This is the original continent outline, rotated c. 20° clockwise, with the islands from the more recent version. I think the eastern regions of the continent look quite okay, but the Western sphere ends up very far north. Maybe we should use something in between. I've also been considering reshaping most of the western areas rather drastically, mostly for tectonical reasons, to the end of a) increasing the east-to-west distance at the latitude of the Rathedān somewhat, and b) creating more room for the western civilizations. Comments please...

Naidda SCs

I can't quite get "Naidda" from these... it seems to come out "Naizha" instead. I'm not sure if this is dialect mixing or if the changes are slightly off.

There are two reasons for this: Firstly, the order of these three changes:

b d g &rarr; v ɾ j / V_V
p t k &rarr; b d ʔ / V_V
p t k &rarr; v ʒ j / in medial clusters, except after s

If the last of these changes is applied to only the first plosive in a cluster, I get /najta/, but there's no further change afterwards to flap the /t/. And even if we add one, there's a further possible obstacle in the 4th SC section with the following palatalization rule, which I'd personally tend to extend to /4/ quite naturally:

r g ŋ &rarr; ʒ ʒ ɲ / _[j i I e] or _

The wiki has been down for days. Does anyone know what's going on with it?

I guess we've been using it too much, as the error message mentions a bandwidth limit. I predict that the wiki will work again on May 1st. But maybe we should ask Neek how much bandwidth he has and whether upgrading or moving to a wiki of our own would be an option...

[Radius Solis]

No, you won't get "Naidda" from those changes. It's the Southern dialect version of the name.

-----

I'm not sure how much frost Kasca should get, but the sea being nearby is a moderating influence that would tend to minimize hard freezes. There is probably a warm oceanic current running right up the coast. And I know from experience that oranges can tolerate a few frosts per year; there are orange trees all over Tucson where I live, and not a winter goes by without the temperature dropping into the mid-low 20s once or twice and milder frosts a dozen or so times. Every spring bajillions of oranges still ripen (like right now - it's fun to throw or stomp on the fallen ones ). Lemons and limes don't do nearly so well, though.

Olives probably don't grow in Kasca proper - the delta's way too wet - but I was hoping they'd grow in Kuaguatia ("south Kasca"), which is both further south and further inland where it's drier and somewhat warmer.

[Corumayas]

This is the original continent outline, rotated c. 20° clockwise, with the islands from the more recent version. I think the eastern regions of the continent look quite okay, but the Western sphere ends up very far north. Maybe we should use something in between. I've also been considering reshaping most of the western areas rather drastically, mostly for tectonical reasons, to the end of a) increasing the east-to-west distance at the latitude of the Rathedān somewhat, and b) creating more room for the western civilizations. Comments please...

I'd be in favor of that... our continent is quite narrow in the middle, and in my opinion it could really use more steppe/plains for nomadic cultures to develop in. What would happen if we rotated the east of the continent but left the west in place, filling in the gap with steppe...?

I'm not sure how much frost Kasca should get, but the sea being nearby is a moderating influence that would tend to minimize hard freezes. There is probably a warm oceanic current running right up the coast. And I know from experience that oranges can tolerate a few frosts per year; there are orange trees all over Tucson where I live, and not a winter goes by without the temperature dropping into the mid-low 20s once or twice and milder frosts a dozen or so times. Every spring bajillions of oranges still ripen (like right now - it's fun to throw or stomp on the fallen ones ). Lemons and limes don't do nearly so well, though.

Right. There's a lemon tree in the backyard where I'm living now, in fact (near San Francisco)... but apparently it doesn't freeze here often. Mandarin oranges and tangerines grow in South Korea too. So if Kasca's climate is borderline, oranges could still grow there. That might imply that it's a little south of where it is on the original map, but it needn't be a lot... say, around 35° N instead of 40°. So it might resemble the Carolinas or mid-coast China...?

Olives probably don't grow in Kasca proper - the delta's way too wet - but I was hoping they'd grow in Kuaguatia ("south Kasca"), which is both further south and further inland where it's drier and somewhat warmer.

I don't know... olives are notoriously picky about where they'll grow, and all the major olive producing regions have Mediterranean climates (according to Wikipedia 95%+ of olives are produced in the Mediterranean region itself). In the US they're apparently only grown in California and Arizona (around Phoenix)... Wikipedia says "Olive cultivation has also been attempted in the south-eastern states, especially in South Carolina, Florida and Mississippi" (which sounds like it wasn't successful... but no source is given and I can't find any more information online). On the other hand, it turns out that olives are now being grown in patches of New Zealand; maybe rain shadows help create some drier microclimates there.

There's still the possibility that Kuaguatian olives are different from ours, too.

EDIT:
Eureka! Some varieties of olive do grow in Florida. So it looks like you're good after all... I withdraw the objection.

[Radius Solis]

The wiki has been down for days. Does anyone know what's going on with it?

I guess we've been using it too much, as the error message mentions a bandwidth limit. I predict that the wiki will work again on May 1st. But maybe we should ask Neek how much bandwidth he has and whether upgrading or moving to a wiki of our own would be an option...

Neek is also about to go to boot camp (he's joining the military) if he hasn't already, and may be unavailable soon or already. This leaves the future of the entire KQ in limbo, and my recommendation is to either pool some funds for a host for our own wiki, or else talk to the Frathwiki folk and ask nicely if we can put a ton of Akana stuff there in return for a modest donation or two, and if Frathwiki can handle the traffic. I'm somewhat in favor of the latter option, as installing and managing a wiki is potentially messy and difficult, especially if none of us have appropriate tech savvy.

Either way, my intention as soon as the KQ comes back up is to save the wikicode for every Akana article that currently exists. Just in case.

[zompist]

For what it's worth, MediaWiki is pretty easy to use and reliable. I installed it myself for Almeopedia and have had no problems with it. (Except for spammers.) The MediaWiki pages have good instructions.

You need a database too; Lore already had one and set that bit up for me.

[Legion]

A little something I've been thinking about: to bring some unity between the various classical edastean languages, that is, propose some macro-grouping, which I think is still faisible since, besides Adata and Faralo, most other languages are still at stage where we can work around some stuff in it.

Here's what I have in mind:

Western Edastean:
-Adata
-Ndok Aisô


Eastern Edastean:
-Naidda
-Faralo
-Komeyech (replacing Dhaleglo, which is now obsolete, on the Dawaim islands)


Isolate?:
-Qedik?

Western Edastean common features:
-Evolution of a third row of occlusives.
-Loss of the article.
-Preservation of a more complex verbal morphology (moods, passive voice)
-Shift to SVO word order.

Eastern Edastean features:
-Simplification of the verbal complex, compensated by developpement of an auxiliary verb system.
-Preservation of the VSO word order.
-Preservation of the articles.
-Consonantic mutations (Faralo and Komeyech).
-Inflected prepositions (Naidda and Komeyech).
-Enclitic pronouns (Faralo and Naidda)


Thoughts?

[dunomapuka]

Western Edastean common features:
-Evolution of a third row of occlusives.
-Loss of the article.
-Preservation of a more complex verbal morphology (moods, passive voice)
-Shift to SVO word order.
Thoughts?

Yes, I've been having similar thoughts about Ndok: SVO, no articles. Not quite sure what I'll do with the verbs, but yeah, they'll probably work similar to Adata. Furthermore, Ndok will share the innovation of plural prefixes from determiners. Actually, I think it might end up with two plurals: one for "a lot" and one for "a few."

My current draft of Ndok Aisô also has occasional consonant mutations, for example: mpóisôxeu "fish" --> na-fóisôxeu "fish" (pl.)

[Radius Solis]

Additionally:

* Faralo and Naidda both have polypersonal marking.

* They also both distinguish two sibilant articulations, while Adata and (I think) Ndok Aiso only have one.

[dunomapuka]

uhh guys

disaster. this is what the wiki is re-directing to now:
http://paul.serenityis.net/suspended.page/


God, this really is the "cursed relay."

[Corumayas]

And it was working fine just a few minutes ago. Arrgh.

[Radius Solis]

Um. Oh boy.

Well, I just asked around and it seems Neek is indeed off in basic training, and thus pretty much unreachable. I'm sending an email to the hosting co, serenityis.net, to see what if anything can be done (e.g. if they'll accept third-party payment of Neek's bill, which I will do if it isn't too big).

[Legion]

Additionally:

* Faralo and Naidda both have polypersonal marking.

* They also both distinguish two sibilant articulations, while Adata and (I think) Ndok Aiso only have one.

Eh, Komeyech also has two sibilant articulations too

[schwhatever]

Side note: I've only just now, after all this time, noticed a certain infelicity in the original Ndak Ta lexicon: "spear" is given as tselô. NT did not have nasal versions of /o u/. Adata's reflex treated it like any other final nasal vowel (adding -n), so I can't ignore it; I'm just going to call it an irregular word. Odd though it is, occasionally a language will have a "phoneme" that appears in only one or two words, typically borrowings or ideophones. (IIRC Finnish has geminate /h:/ in exactly one word; English similarly has the initial cluster /vr/ only in "vroom".) So now we need a nearby language with /o~/ for this to be a loan from! :8

I've been interested in joining this project for a while now, so I'd be happy to develop a neighbor language for such influence. Would a modern descendent be of interest as well (as another isolate, perhaps?). Are there any areal issues that I should be wary of?

PS: I won't be able to contribute as much as I'd like until later in May. Bloody AP tests... grrrm.

[Cedh]

My intention as soon as the KQ comes back up is to save the wikicode for every Akana article that currently exists. Just in case.

I have a complete copy of the Akana stuff on my local wiki, the latest synchronization dating to April 23, 00:36 CEST. The only things I remember someone changing after that are some minor corrections on the Habeo languages page. Adding this to the FrathWiki per copy&paste would take about an hour; using the import function would be much faster but require wiki admin rights. For future expansion it might be better though to create a wiki of our own. Let me suggest to discuss this issue on #Akana. For those who haven't been there yet: the server is irc.sorcery.net.

I've been interested in joining this project for a while now, so I'd be happy to develop a neighbor language for such influence. Would a modern descendent be of interest as well (as another isolate, perhaps?). Are there any areal issues that I should be wary of?

Welcome!

I'd suggest we ascribe NT tselô to dialectal variation, or else a Miwan/Ngauro/Tlaliolz borrowing. I don't think we should add yet another substrate language so early in history while we have not yet described the established ones. As for these, Miwan and Ngauro are in Corumayas' domain as he is establishing the diachronic framework for the Eige-Isthmus family, but we don't have a description of Proto-EI yet. Tlaliolz rests on the upcoming Proto-Talo-Edastean sketch by Radius, and should be formed to be a plausible influence for Puoni, of which, sadly, only a lexicon is currently available. But this might be a good project.

I don't think there's much room for another descendant of Ndak Ta itself, except maybe further upriver in the lower Xōron Eiel (where Komeyech was originally meant to be placed, but there we already have Meshi (of which nothing is known except that it belongs to the Eige-Isthmus family and forms part of a dialect continuum with Miwan), and several Xoronic/Habeo languages by pocketful_of_songs.

Probably the best option would be to engage in one of the areas where we have obvious holes. For instance:

- Tlaliolz, as indicated above.

- In order to continue with the history of Athalē and Lasomo after c. 500 YP we could use a Hitatc language for the lower Milīr valley, namely Ktacwa and/or a close relative of that. I'm currently struggling with the syntax for Proto-Hitatc and could use some help there; when that is done, there is a solid proto-language to start from. I do have some general ideas for the sound changes of the western Hitatc languages, but only very little is set in stone yet (1. Western Hitatc changes PHt. *m_0 n_0 N_0 N\_0 into fricatives; 2. WHt. has *pi fi mi > pL fL mL / _V at some point in its history, 3. Ktacwa (c. 200 YP) is fairly clustery and contains the words xastn "to admire", yorot "to change" and šilok "to consider" which were loaned into Pencek, and 4. ktacwa is a reflex of intermediate *xitatc wan). You're more than welcome to join me in this.

- If you want to be able to work rather freely, but from a solid basis and in a region of relative historical significance, you could derive a northern descendant of Fáralo. It would probably have some Feråjin and/or Doroh influence, but there is currently not much info on these.

- In the west of the continent we have the excellent Proto-Western with no completed descendants so far. The only caveat here is that the area is rather far off, and the time depth until descendants other than Gezoro (being worked on by kodé) make regular contact with the Eige valley cultures is likely to exceed three millennia.

- In the east of the continent we are in need of a Proto-Núalís-Takuña language. I've recently suggested I could adapt my Proto-Madhainic for that, but if you would like to place a completely new a-priori language there, just go ahead. As Takuña (we have four words so far) and Núalís (only the name is known) both date to c. -500 YP, PNT should be spoken before -1500.

[Radius Solis]

That is a true relief, cedh. Thank you for having the foresight.

For future stability I agree having our own host would be superior; but for speed of having stuff available again, frathwiki is already set up.

Let's compromise, for now: we put a few of the most important pages up on frathwiki (like Akana, Empire of Athale, and Timeline) for use as a reference and something we can point newer people to, while we decide what to do about a long-term solution.

Meanwhile, losing the KQ is still a blow to the whole ZBB, and not just for the Akana content. So when I see what the hosting company says about it, I will post a thread in NOTA to seek public consensus on what to do.

--------

Incidentally... I do already have another NT daughter that's been half-sketched and sitting in my files for months now. Tentatively called Dixttak. It's a variant spoken in a northern mountain valley (until its speakers were overrun by another group), with no known descendents. Rather than posting a sketch, I was considering using it for a little mini-game within the relay, just for fun: A manuscript turns up in the imperial archives in Huyfarah. Scholars believe the language to be Edastean, and have worked out a rough translation. Here it is. How much of the language and its diachronics can you reconstruct? But Dixttak is low priority compared to many other things, and may not see the light of day for months or maybe ever.

[Cedh]

Let's compromise, for now: we put a few of the most important pages up on frathwiki (like Akana, Empire of Athale, and Timeline) for use as a reference and something we can point newer people to, while we decide what to do about a long-term solution.

Done. It doesn't look too pretty because the templates do not work properly with the default stylesheet, but I'd suggest we leave the code alone for now, until we've come to a decision.

[Zhen Lin]

Oops, I completely forgot I had sysop on FrathWiki. But I doubt we need to import the full history.

Presumably you have the CSS we need? I can add it to the stylesheet.

[dunomapuka]

Well, let's not let this snag bring the project to a halt. Legion, I'm curious to know how Komeyech is coming along. Bringing the Dagaem Islands into focus will be cool. I myself must report that Ndok Aisô remains worse than sketchy still. I have been busy with schoolwork.

Has anyone been IRC chatting? Should we schedule something?

[Radius Solis]

If we can schedule a chat time, I will be there.

In other news, Spack is re-examining Qedik and a revised/better description may result.

[Legion]

Boy12 > Komeyech is progressing well, I think I'm almost done with the morphology, I'll start the syntax soon.

Rad > yay for Qedik!