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

[TzirTzi]

TzirTzi >
A question... as I understand, PNT (or somewhat more recent) vowels developed to Takuña as follows: /ii uu/ > [iː uː], /ia ua/ > /ii uu/ (phonetically [ɪj ʊw] or [ej ow]). That is, a bit different from what comes to mind first Perhaps, this point needs some explication...

What actually happens in my sound change rules is something like:

PNT /ii uu iu ui ia ua ai au aa/ [i: u: iw uj i@ u@ 6j 6w a:]
[i@ u@] -> [e: o:] -> [eI oU] -> [Ij Uw]
[6j 6w] -> [E: O:] -> [{E QO] -> [ae Qo] -> [aj Aw]
[iw uj] -> [ju wi]
Takuña /i: u: a: ii uu iu ui ai au/ [i: u: a: Ij Uw ju wi aj Aw]

Although not quite in that order, and mixed in with other changes. I see it as a stage of an ongoing raising process by which many Takuña vowels are merged as /i: u:/ (rather like the very long term tendency in Greek for all vowels to slowly collect as /i/).

Not really related... I came across a real-world orthography that uses ñ for [ŋ]: the official Romanization of Kazakh. It is much less used than cyrillics presently, but the official website of Government of Kazakhstan still supports a version in such orthography.

Well... Nice to know that someone other than Tolkien has used it . Still, not the nicest of romanisations...

[Corumayas]

This makes sense (though I wouldn't expect very close sisters of Mutsipsa' et al. in Ttiruku-- I'd guess that all the western Isles langs might form a genetic subgroup).

What if we try to *discover* such groupings first, based on purely linguistic things? Like, earliest sound changes that happened to each language? It would be a lot of fun if we discover that Mutsipsa' 's closest relative is e. g. Zele

You're right, of course that's the way we should do it.

Since earliest dialectal splits must have taken place still on the NE Continent, and then all the migrations passed (consecutively?) through the bottleneck of Ttiruku...

Probably. On the other hand, it's possible that at least some of the northwestern Isles speakers crossed the sea directly from the NE continent, rather than from Ttiruku. The description of the arrival of the Mûtsinamtsys on their islands sounds to me like they got there via a long sea voyage, rather than skirting the coasts.


TzirTzi: For what it's worth, I'm pretty sure there are Núalís loans in Máotatšàlì (some of which were then borrowed in Mûtsipsa'),and probably some Takuña ones in Mûtsipsa' (some of which might've been borrowed in Máotatšàlì). Hopefully this reconstruction project will sort those out...

In the interest of that, it would be good if someone could work out etymologies for Máotatšàlì too. I'll see what I can do over the next couple of days.

[Basilius]

In the interest of that, it would be good if someone could work out etymologies for Máotatšàlì too. I'll see what I can do over the next couple of days.

I continue immersing in Isles history anyway, and my idea about palatalizing pharyngealization was based specifically on Máotatšàlì data, so probably I must have a try. OTOH cedh seems to have already done something, and I know close to nothing of Máotatšàlì's neighbors.

It promises to be complicated, though. Thus, I notice some words that were perfect homonyms in PI but become differentiated by tone in Máotatšàlì (e. g. àmá 'day' vs. àma 'sun', kazàš 'hand' vs. kazáš 'year', kiomi 'moon' vs. kiomí 'month'). Other examples seem to point to some derivation pattern, e. g. lyzæ 'actor' - lyzæ` 'middle singer (tenor/alto)' - lýzæ` 'high singer (soprano)' - lỳzæ` 'low singer (bass)'; or: mutsí 'think' - mutsì 'believe' - mùtsì 'trust', but the grammar doesn't mention such patterns. Some other stamements ("the first vowel in adjectives is always low or mid tone" &like) make me feel that it's simply a voluntaristic treatment of contrastive tonality...

What to do with all that?

[Radius Solis]

What to do with all that?

Mark all the things you can't figure out "unknown" and deal with them afterwards, maybe. Sometimes if you set aside the hardest parts and work on the easier things, later on the harder parts will seem clearer. Either that or we simply won't be able to solve the mysteries, but that's okay, there are unsolved mysteries in the development of some real languages too. And of course we could always retcon an explanation - for instance, the tonal distinctions arising by picking up a productive tonal alternation from a contact language.

As for nearby languages, what you see on the wiki and grammar is all we know.

[Corumayas]

I've wikified the Máotatšàlì lexicon, with the more obvious etymologies filled in. As Basilius observed, there are derivational tone alternations; I'm not sure where they come from (possibly tone replaces reduplication in some words?), so I've marked the words I take to be derived with a tilde in the etymology. (Sometimes the word with the more basic meaning seems to be the derived form, if I'm right about the regular correspondences...)

[Basilius]

I've wikified the Máotatšàlì lexicon, with the more obvious etymologies filled in.

Nice job! In fact, there don't seem to remain any PI etymologies lacking (except perhaps kagæ` 'relative' ~ kakan 'brother'; maybe searching the Grammar will reveal something to add).

As Basilius observed, there are derivational tone alternations; I'm not sure where they come from (possibly tone replaces reduplication in some words?), [...]

I thought of this, too, looking at pairs like 'sea' :: 'lake'. It all appears to be an undocumented derivational model in PI, then. And supports my analysis of reduplicates being prosodically different, based on the Ppãrwak reflexes.

Unfortunately, the Grammar does not seem to provide any decisive examples reflecting the reduplicative forms (the gV-/(x)V-prefix may be one... or not).

There are lots of split reflexes in segmentals, too: waniš 'star' vs. wænìš 'stage light', kæ`pùs 'extract' vs. kæpyš 'remove', perhaps miogí 'water' vs. miokì 'blue', mypà 'clothing' vs. mupà 'clothe' (I think the latter is hardly < munna; also, I don't think words with u instead of y are necessarily loans - cf. the alternating forms of accusative marker, -ỳs / -wùs).

And of course we could always retcon an explanation - for instance, the tonal distinctions arising by picking up a productive tonal alternation from a contact language.

Or, still better, a couple borrowed suffixes/segmental alternations ultimately reflected as tones...

[Corumayas]

I've added some more words and etymologies to Mûtsipsa' too (function words and philosophical terms). There's still some obscure compounds in both langs which might involve PI roots, I think.

About the split reflexes in Máotatšàlì: it could also be the result of dialect mixing, or just simple innovation (maybe by analogy with a few inherited minimal pairs, somehow).

It's apparent that high tone in Máo. results from a following /?/; the source of low tone is harder to identify, but it looks like /h s/ and maybe /m/ might be involved.

[Radius Solis]

One challenge I'm facing with Mûtsipsa' right now is determining what noun classes all those philosophical terms belong to. Class 1 is all consonant-final nouns so that much is easy, but the other six classes are all vowel-final.

[Corumayas]

The etymologies should help with the ones that are compounds; e.g., if I'm right that kfasa is "spirit house", it should be class VI (the same as psa "house").

Of course that only works if you can identify the second element of the compound.

(Some words in the philosophy section had their nominative suffixes attached, so that could be another clue.)

[TzirTzi]

Hey - just to say that I'll be mostly disappearing from the ZBB for a few weeks to concentrate on university work, so if there are any issues with akana wiki then feel free to email me rather than posting here

[Radius Solis]

Good luck with that TzirTzi, and when you return the time may be ripe for some conworlding in the Siixtaguna region, for which your input would be desirable.

[Basilius]

Probably. On the other hand, it's possible that at least some of the northwestern Isles speakers crossed the sea directly from the NE continent, rather than from Ttiruku. The description of the arrival of the Mûtsinamtsys on their islands sounds to me like they got there via a long sea voyage, rather than skirting the coasts.

(The context.) I thought of that once more...

Either I fail to locate the passage referred to, or it doesn't sound so unambiguous to me. And then...

A leap from the NE continent directly to the northern islands would look like the Age of Discovery beginning somewhere in Neolithic... (I haven't figured out yet how to measure dustances in cedh's map, though... cedh, can you help?)

And the starting point would be located somewhere near the NW corner of the continent, which (combined with the migration to Ttiruku and Zeluzhia) suggests that previously the Isles speakers had spread all along the Western coast. Which makes me wonder how big the nation who ousted them had to be...

Also, it seems to imply transporting a few thousands of people (with children, cattle, utensils, etc.) in just one trip... Either that, or establishing a route connecting the two continents, which existed for a prolonged period but was then forgotten for obscure reasons.

For sure, there's the Terrestrian example of Polynesians, but their most spectacular moves (like, from Tahiti to Hawai`i) seem to have followed a chain of resettlements with the distances gradually increasing. The geography of Akana doesn't seem to provide a setting for this - without involving Ttiruku, that is. Mmm?

* * *

On a different topic. I meditated a bit over the timeline proposed by cedh here.

My suggestion for the beginning of a timeline:

c. -20000 to -15000: Humans first cross the Ttirukũ island chain to reach the NE continent. Zeluzhia is colonized soon after. (This is early enough so we won't have to worry too much about making languages of the two eastern continents appear related to each other, and late enough so eastern cultures don't have too much of a head start given their favourable environment)

c. -12000 to -8000: A second wave of migration from Peilaš to the NE.

c. -5000: A third wave of migration. ((The languages of this stock remain demonstrably related) About the same time, agriculture is first invented on Peilaš, but the migrating peoples are still hunter-gatherers.

c. -2500: Agriculture is invented independently on the NE continent (probably on the southern coast opposite NE Zeluzhia).

c. -2000: Agriculture spreads to the western tip of the NE continent (Proto-Isles culture).

c. -1800: Rise of the first major empire of the NE continent, conquering roughly the western third of the landmass. Proto-Isles peoples flee offshore and establish themselves on various islands in the Ttirukũ chain, bringing agriculture with them.

Lately, I catch myself at using (in my internal speech) the following names for the language groupings representing the three waves:

Primundan
Mediundan
Ultimundan

Do these look oK? My non-native intuition about English may fail to warn me of some unpleasant association... (And, perhaps Primundic etc. would sound better?)

Tangentially related: it seems to me that people didn't like my term "the Insular Pipe" (proposed here). Any suggestions for an alternative naming?

I considered Wikifying something on these matters, so I thought I'd better ask people first...

[Radius Solis]

There might be plenty of small islands between the NE continent and the Ttymytan islands, good enough for easy island-hopping but too small to show up on the world map. One of my map-redraw proposals from a year or two back actually showed a whole little Northern Polynesia in this region of ocean, for exactly this purpose, but I dropped that map proposal (for other reasons).

Lately, I catch myself at using (in my internal speech) the following names for the language groupings representing the three waves:

Primundan
Mediundan
Ultimundan

Do these look oK? My non-native intuition about English may fail to warn me of some unpleasant association... (And, perhaps Primundic etc. would sound better?)

There's nothing wrong or unpleasant with those names. But you're right that -ic would sound better.

[dunomapuka]

Radius: If you're not doing Fishermanese, I'd like to take over, if there are no objections. Because I feel like creating a language from scratch. What, if any, are the constraints on the project? We're just looking for some vocabulary, right?

[Radius Solis]

Awesome! Yeah, there aren't many constraints on the Fisherpeople's language. We're looking for some vocabulary both generally, and in the specific sphere of urban life... all else was an excuse to try out some diachronic ideas. Those have now found another home.

The original template called for a two-stage setup, first a mother language and then, a millennium later, a group of dialects descended from it, with most of the Western borrowings and the "focus" time being the latter period. The later people were to be spread among several smallish cities, somewhat Minoan-based in culture and architectural style. You are free to keep all, some, or none of the above, as you see fit.

[dunomapuka]

Awesome! Yeah, there aren't many constraints on the Fisherpeople's language. We're looking for some vocabulary both generally, and in the specific sphere of urban life... all else was an excuse to try out some diachronic ideas. Those have now found another home.

The original template called for a two-stage setup, first a mother language and then, a millennium later, a group of dialects descended from it, with most of the Western borrowings and the "focus" time being the latter period. The later people were to be spread among several smallish cities, somewhat Minoan-based in culture and architectural style. You are free to keep all, some, or none of the above, as you see fit.

I'll stick with the plan. What I think I'll do is a Swadesh list, some urban and sea-related vocabulary, and anything else that strikes my fancy. Are they supposed to be literate?

[Radius Solis]

I don't know, because I hadn't yet figured out exactly what time period they'd be in. Choose as you prefer.

[Corumayas]

Looking over the Mûtsipsa' page again, I just realized that a couple more of the philosophers have names that look like they must be Takuña even though it's not specified. Also, several of the terms Takuña philosophers use lack good Isles etymologies, or have uncertain ones, and just might be Takuña words. Here's a list:


Takuña names:

Sútupaj
Rutawká
Ná'ápíru (?)
Ahuñipá (?)


possible Takuña words:

nuduuhasihi 'epiphany, enlightenment'
ifiisana 'the spirit world' -- almost certainly Takuña, since it's said to be a traditional concept in their beliefs

pudusa'a 'leafy flesh'
pudumasi 'bloody flesh'
sa'usi 'wood, biological matter'

(the above three seem to share some non-Isles roots)

pinetu 'wind, air'
huude 'stars' (maybe Mûts. < 'wind-son')
a'usode 'planets'
paniimuki 'calendar'


There are probably other Takuña loans in Mûtsipsa'; there are still 40-odd words without good etymologies. But atm it's difficult to sort them out from the loans from Thokyunèhotà and so on. The same is true for Máotatšàlì and Núalís loans; there are even ~17 apparent loans that are shared between Máotatšàlì and Mûtsipsa', which could probably come from any of those three sources...


Legion, if you see this and have time, it'd be enormously helpful if you could look over the Máotatšàlì and Mûtsipsa' lexicons and identify any words that you recognize as coming from Thokyunèhotà.

[Basilius]

Hi!

I have a question for people who work on/with the Isles langs.

I miss a name. A toponym, to be more specific. "The NE continent" sounds like a temporary ersatz, and expressions like "NW of the NE continent" look ugly.

So I tried to invent something based on PI.

The variant I liked most so far is the following one.

Dúyhaq = dúy 'sea' + háq 'in front of, before' = (literally) 'what is before sea', (lexicalized already in PI as) 'starting point of a maritime travel', (later re-interpreted as) 'the former overseas homeland'.

This will imply postulating a derivation pattern based on postpositions attached to substantives without case endings (with meanings like 'what is in...', 'what is behind...' etc.). AFAIK this pattern is otherwise unattested as of now, so potential problems must be weighed first.

Also, I thought of using the Mûtsipsa' reflex of the above, but it'll probably end in an apostrophe, which I found inconvenient... Besides, I don't understand what should be the reflex of the first vowel (zero as in tssi < tzuysi 'well'? û as in tûk < tuygim 'to stop'? ûû as in dûûxu < dúykuq 'stream, river'?) and the final consonant (glottal stop as in he' < háq 'in front of' or hafatsa' < hapatzaq 'to burn'? zero as in mimki < mimkiq 'water'? h as in 'itesah < qí taysaq 'good comparison'?).

What do you people think? Do you have any alternative suggestions?

[dunomapuka]

Good idea. That place needs a real name -- how about Dûûha for the Mutsipsa form? I like that better than using the PI form.

[Corumayas]

It looks like it should maybe be Dûhe', actually; cf. dûke island < dúy-káy 'sea-mountain' (where dúy by itself > duu), and he' 'in front of' < háq. PI á regularly gives e, in any case.

Another option, if we wanted to avoid the bare stem + postposition construction, would be to construct a whole noun phrase: something like dínti dúyuys háq 'farmland in front of the sea' might be good. As a Mûtsipsa' phrase it'd be diinti duu'ûs he', but I'm not entirely sure what changes to apply to make it an old compound. Maybe something like Diintituu'ûse'?

[Zhen Lin]

I'm liking how Kuyʔūn is turning out. Ungwikuk also looks interesting, if very unusual with rules like [[ss]] > /n/...

[Basilius]

Another option, if we wanted to avoid the bare stem + postposition construction, would be to construct a whole noun phrase: something like dínti dúyuys háq 'farmland in front of the sea' might be good. As a Mûtsipsa' phrase it'd be diinti duu'ûs he', but I'm not entirely sure what changes to apply to make it an old compound. Maybe something like Diintituu'ûse'?

I thought of such compounds, too. Unfortunately, it won't decline correctly (and fixing this will require a more profound revision than just adding a derivation pattern, IMHO).

OTOH the PI grammar does not specify the word order with postpositional (and other adverbial) phrases used as attributes of nouns. In fact, it is not clear if such constructions are allowed at all, which I feel to be an issue... If we decide that they were indeed allowed and could be both head-initial and head-final, then another option would be dúyuys-háq-dínti, with case markers suffixed in the usual manner.

What I didn't like with this option: we don't know a suitable PI lexeme with meanings like 'land' or 'country' (or just 'place'); reflexes of dínti seem to display more specialized meanings in both Mûtsipsa' and Máotatšàlì. It looks more like a consequence of the briefness of the available description than like a peculiarity of limited vocabulary of the language itself (which does not appear especially Creolish in other respects). Sure, Ran gave us a carte-blanche for fleshing it out, but I think we should understand the structure and the (pre-)history of PI much better before we start adding new roots.

BTW, with my proposed derivation there is a similar issue of component ordering: there is a potential alternative like háqduy.

Also, we could opt for a Ppãrwak reflex (after all, we already have the Ppãrwak-looking "Ttiruku"): dúyhaq > Túuhap (not sure about the intervocalic -h-, though) or háqduy > Háptuu.

Or else invent a new Isles lang to obtain tastiest sound possible :)

[Basilius]

Ungwikuk also looks interesting, if very unusual with rules like [[ss]] > /n/...

Yes, it's supposed to look crazy :)

At some moment all word-initial resonants got devoiced, with *l ultimately yielding /s/; *l and *n were already in near-complementary distribution at that moment, and there were no phonemic voiced obstruents. (Another typological peculiarity: immediately before the initial devoicing the system lacked bilabial obstruents altogether.)

Precipitating things a bit.... The whole project is centered on one idea: a lexical derivation system which is very rich and flexible even for natlangish standards and provides for lots of compact coinages to express rather complex notions. (Something I often miss in historical conlangs, BTW.) Everything, including alternations, class agreement with prepositions etc., and ultimately even ergativity and auxiliary-based conjugation is actually motivated by word formation.

In short, it's a language to borrow from :) I am planning to make it an open project as soon as I outline some skeletal features.

BTW, Zhen Lin, I kinda have a personal message for you which you may choose to take as a joke or not :) - While I didn't think of that when I was looking for the name (whose original meaning, as it seems, would be something like 'people whose hills abound in bears'), now I find Ungwikuk and Ōmishima suspiciously similar-sounding... I thought Ungwikuk was spoken somewhere in the eastern half of the NE continent, in case anybody wondered :)

[Corumayas]

I thought of such compounds, too. Unfortunately, it won't decline correctly (and fixing this will require a more profound revision than just adding a derivation pattern, IMHO).

Hmm. PI nouns with final /?/ are Class IV in Mûtsipsa'; so I think it could be interpreted as a stem Diintituu'ûse, with the NOM form Diintituu'ûseh.

It turns out there is a parallel place-name, actually: the main island of Mûtsinamtsys is Ke'idû'ûs'as (n.I) < káy-i dúy-uys qas 'mountain by the sea'. (Which suggests something like Diintidû'ûsheh...)

I pretty much agree with the rest of your post. Actually, head-final phrases would make more sense typologically; so maybe Duu'ûse'diinti? But the above example seems to be head-initial, so maybe not.