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

[dunomapuka]

If that's to be the official name of the continent I'd prefer something much shorter.

Also, I have started a page on the Fisherman AKA Lukpanic languages. Further description 'n' stuff to follow. "Lukpanic" comes from their own name for themselves, which, rather picturesquely, means "the people facing toward the sea."

Though not quite as kickass as "the people whose hills abound in bears."

[Zhen Lin]

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

Sure, why not. Bearing in mind there's an extant corpus that needs to fit, of course.

(Personally, looking back at it, I would rather drop Erhadzy and the associated history, languages etc. But I suppose that's more an aesthetic reevaluation...)

Also, I have started a page on the Fisherman AKA Lukpanic languages. Further description 'n' stuff to follow. "Lukpanic" comes from their own name for themselves, which, rather picturesquely, means "the people facing toward the sea."

I wonder if it could be made to be connected to the Peninsular languages. For instance, there is *larut "ocean" and *piktm "person", which might be made to be cognate with lu and gbagba, say via *lrud and *gʷdəm, with:
*lrud → *lərud → PPI *larut
*pi-gʷdəm (prefixing) → *pigdəm → PPI *piktm
and
*lrud → *lud → LK lu
*gʷdəm-gʷdəm (reduplication) → *gʷam-gʷam → LK gbagba
There's also the PPI verbal ending *-i, which is associated with imperfectives...

[Corumayas]

They're pretty far apart geographically though...


It just occurred to me that maybe the official name of the eastern continent should be in some version of Faralo (like ours are Latin)... maybe a Faralo-ized form ultimately borrowed from an Isles lang?

But it would be good to have something simple to call it for now. Dûhe(h) is as good as anything, even if we eventually decide it's not grammatically or sound-changily correct or whatever.

[Basilius]

If that's to be the official name of the continent I'd prefer something much shorter. :P

That would be my objection, too...

But it would be good to have something simple to call it for now. Dûhe(h) is as good as anything, even if we eventually decide it's not grammatically or sound-changily correct or whatever.

I second this.

Does anybody object? Hey people, if you find it aesthetically ugly, or have any considerations concerning its linguistic plausibility, or can suggest a much better name for the continent - please don't keep silent, or Dûhe is going to stick forever :)

It just occurred to me that maybe the official name of the eastern continent should be in some version of Faralo (like ours are Latin)... maybe a Faralo-ized form ultimately borrowed from an Isles lang?

Fáralo was near-contemporaneous with the attested Isles languages, so it doesn't seem to make so big a difference... (Also, "Ttiruku" doesn't look Fáralo'ized.) I was thinking of designing (the sound changes for) an Isles language to be spoken in a short-lived colony in the Huyfárah area, with Dúyhaq reflected as Tysæ and borrowed to Fáralo as Tiwsæ...

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' [...]

Thank you for pointing to this one!

It seems to suggest that adverbials-as-attributes were indeed allowed in PI and could be put to the right of their head, without any additional marking (like empty verb or somesuch).

However, I doubt if they declined in Mûtsipsa' as you suggest. Unless there is some textual evidence to the contrary, I'd suppose that case markers were attached to the head component, à la the (pseudo-)compound names of Classical Arabic: ʕabdu-l-lāh(i), ʕabdi-l-lāh(i), ʕabda-l-lāh(i)...

Also, I thought of using the word for 'east' (PI hápa, homonymous with 'sun' and 'day') instead of the missing 'land/place', but if we accept Dûhe or Dûheh this is not topical anymore.

[Basilius]

(Now on Ungwikuk)

Though not quite as kickass as "the people whose hills abound in bears."

Anything wrong with this etymology? (Actually, I can change it at any moment.)

I thought the ethnonym was originally a clan name, derived from a toponym referring to a relatively small area, whose original meaning ('hills abounding in bears') had been somewhat obscured in Ungwikuk as spoken around -1000 YP.

Sure, why not. Bearing in mind there's an extant corpus that needs to fit, of course.

OK, I deliberately don't hurry with the lexicon (my original idea was to auto-generate as much as possible). (OTOH I don't believe in accidental convergent evolution, so actually my proposal should be read: *if* you *ever* consider adding or replacing some lexicon of Ōmishima origin, you might also consider Ungwikuk as a/the source, with the available option of reconstructing much of the Ungwikuk lexicon as you wish...)

Personally, looking back at it, I would rather drop Erhadzy and the associated history, languages etc. But I suppose that's more an aesthetic reevaluation...

Is it more about the languages themselves, or perhaps the conhistory/conculture?

The reason why I'm asking: I've seen people stating that the development of civilization in the world of Akana was somewhat slower than on Earth, but with computerization around 2800 YP it doesn't seem really slower... OTOH profound changes in evolving languages (which appears to be the main source of fun with Akana) require a lot of time, realistically...

[Basilius]

I thought of a couple other things. Related to Wikification in the first place.

(1) A ficticious "history of Akana comparative studies". Basically, I would like to be able to refer sometimes to the original sketches of some orphaned languages not as the ultimate authorities but as earlier attempted reconstructions. Just for example, Ran's description of PI could be treated as a pioneer reconstruction by a linguist of Akana whose name was, by mere coincidence, Ran.

(2) An index of typological features found in the languages of Akana. (Particularly interesting would be to have an index of historical changes, a bit later.)

Taking Cedh's Mešmo Koyʔōn as an example - just to illustrate what kind of features could be included in the index.

Phonology

VOT/phonation-based and related distinctions in obstruents: none

Contrastive labial/lingual POA's in stops/affricates: 6 (p t ʦ ʧ k q)

Contrastive labial/lingual POA's in fricatives: 3 (s ʃ ɬ)

Consonants outside labial/lingual POA's: 1/2 (h and ʔ in complementary distribution)

Some less common features of consonant inventory:
- h and ʔ in complementary distribution
- ʔ contrasting with zero onset
- lateral obstruents present (ɬ)
- uvular vs. velar distinction in plosives (k :: q)

Types of onset clusters: stop + y, stop + l (py ty ky, pl tl kl)

Types of coda clusters: s + stop, ʃ + stop, non-distinctive nasal + stop (sp st sk, ʃp ʃt ʃk, mp nt nk)

Quantity distinctions in vowels: present (long :: short, symmetrical)

Contrastive vowel qualities: 4 (i ɛ ɑ o)

Descending diphthongs: Vu, Vi; no quantity distinction, forbidden in closed syllables

Some less common features of vowel inventory:
- absence of u

Tonal contrasts: none

Stress position: contrastive

- - - -

Verb template:
(emphasis) // polarity/mood // (classifier) // verb stem // voice // aspect/number

Some less common features of verb morphology:
- negative mood
- verb classifiers: valency-filling/reducing, patient/theme/instrument/time/location (roles not contrasting)

- - - -

Noun template:
case // number+definiteness // noun stem // (possession)

Etc. etc.

Does it make any sense?

[Cedh]

A typology index sounds like a good idea. Just start such a page if you want to.

Does anybody object? Hey people, if you find it aesthetically ugly, or have any considerations concerning its linguistic plausibility, or can suggest a much better name for the continent - please don't keep silent, or Dûhe is going to stick forever

I prefer Dûûha', but Dûhe would be fine with me. The Fáralo-ized versions Duha or Duhe don't look better than the Mûtsipsa' versions though. Tysæ is not bad either, but I suspect the Fáralo might change the final vowel into either [E] or [a] because iirc they natively have unstressed /æ/ in closed syllables only.

[Zhen Lin]

They're pretty far apart geographically though...

Well, we have IE languages from India up to Scotland, so it's not that implausible...

Also, I thought of using the word for 'east' (PI hápa, homonymous with 'sun' and 'day') instead of the missing 'land/place', but if we accept Dûhe or Dûheh this is not topical anymore.

Aha! That's where Erhadzy efeí "west" comes from. I couldn't remember the exact etymon, but now that I see it, I think it's probably something like "place where the sun sets".

Personally, looking back at it, I would rather drop Erhadzy and the associated history, languages etc. But I suppose that's more an aesthetic reevaluation...

Is it more about the languages themselves, or perhaps the conhistory/conculture?

The language mainly, but points have been raised about the history as well.

[Corumayas]

Does anybody object? Hey people, if you find it aesthetically ugly, or have any considerations concerning its linguistic plausibility, or can suggest a much better name for the continent - please don't keep silent, or Dûhe is going to stick forever

My only objection really is that we don't have any evidence that postpositions can appear with a bare noun root like that. In the corpus the noun is always inflected for case; which is somewhat important actually, in that the case can be changed to modify the relationship indicated by the postposition.

I was thinking of designing (the sound changes for) an Isles language to be spoken in a short-lived colony in the Huyfárah area, with Dúyhaq reflected as Tysæ and borrowed to Fáralo as Tiwsæ...

It seems like such a language would likely be related to Legion's Isles lang, the one we don't know anything about yet...

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' [...]

Thank you for pointing to this one!

It seems to suggest that adverbials-as-attributes were indeed allowed in PI and could be put to the right of their head, without any additional marking (like empty verb or somesuch).

However, I doubt if they declined in Mûtsipsa' as you suggest. Unless there is some textual evidence to the contrary, I'd suppose that case markers were attached to the head component, à la the (pseudo-)compound names of Classical Arabic: ʕabdu-l-lāh(i), ʕabdi-l-lāh(i), ʕabda-l-lāh(i)...

Well, let me make my case further: Ke'idû'ûs'as is listed as a Class I noun, with no comments about it declining differently than others. (Its presumed head, ke, is class VI, so it's not inflected according to that.) Also, since the case of the head noun affects the meaning of the postposition I suspect that changing it might be avoided.

Also, I thought of using the word for 'east' (PI hápa, homonymous with 'sun' and 'day') instead of the missing 'land/place', but if we accept Dûhe or Dûheh this is not topical anymore.

"East of the sea" is a great name. OTOH there doesn't seem to be a postposition that quite fills the role of "of" in that phrase; nor is there one meaning "beyond" or "across", so we'd have to improvise something. "East via on the sea" (using the Mûtsipsa' NOM + postposition construction for "action through") seems plausible for that meaning, and looks quite nice to me: Hefadû'i'un... but it's not short! OTOH maybe Duu'i'un meaning "(land) across the sea"? (I'm guessing that úy > uu in initial syllables but û elsewhere; I'm not sure this is correct, but it seems to explain at least some of the facts.) "Across the sea" is a pretty vague name though; "east across the sea" is much clearer...

They're pretty far apart geographically though...

Well, we have IE languages from India up to Scotland, so it's not that implausible...

True; but IE languages also fill most of the space between. These families seem to be isolated at pretty much opposite ends of the continent.

I guess we could create a story of their ancestral migrations, and it could be an interesting one if we decide to do that. But it seems like it would have to precede the spread of the Westerners, since the Fishermannic people (at least) are already building cities before the Westerners get there. That would put it before -3000. Such an early expansion, over such a great distance, and without the advantage of the (newly domesticated) horse that let the Westerners move so quickly... it all seems unlikely to me.

Also, though admittedly not much has been set in stone about their cultures yet, what there is doesn't suggest that either group crossed high mountains and vast grasslands to get where they are now. They both seem fairly local in outlook, settled in smallish coastal cities, and very much oriented to the sea rather than inland. They don't sound like great continent-crossers to me.

That said, if a really good back-story was devised for it, I'd go along with them being related. (That would probably involve some other related langs scattered along the migration route, too...)

[Zhen Lin]

Or, we could do it anyway and leave it as one of the great anthropological mysteries of Akana.

More seriously, I suppose there doesn't seem to be a good reason for there two be two related oceanic cultures on opposite ends of the continent, at least not without positing events of cataclysmic scale (say, an inland sea suddenly vanishing).

Is much happening in the south of Peilaš?

[Basilius]

My 5 copecks...

I don't think such a scenario needs anything cataclysmic.

IMHO horses are more-less irrelevant, and the earlier the migration started the better. In 1000 years one can cross a continent by feet... mmm... try to count how many times :)

What can prevent such a migration is (for example) the fact that the territory to be crossed is not uninhabited.

But a lot depends on population densities and economic specializations. For example, the migrants could specialize on fishery (which agrees with their later coastal orientation), and the idea of total control over territory might be still a bit foreign to their deer-hunting neighbors if they lived in small villages, like, 200 km apart from each other. Little ground for conflict, gradual spread along rivers... It appears much easier to transport utensils by water routes, even if you already have horses, wheeled carriages and roads, and much more so if you haven't :)

The starting point could be far in the South. Then the Westerners came, with an economy implying much denser population and quite different attitude to control over potential pastures, and ousted everybody who could not fight mounted.

The linguistic grounds for this setting is a different question, though.

[Basilius]

In the corpus the noun is always inflected for case; which is somewhat important actually, in that the case can be changed to modify the relationship indicated by the postposition.

I suspect a misunderstanding here; I meant a lexical derivation model rather than an alternative syntactic construction. Something like Αντί-μηλος in Greek or Trans-caucasia in Latin.

Indeed, it's unattested, but I hoped it didn't conflict with anything attested :)

[dunomapuka]

Let's not worry about Greenbergian amusements with Lukpanic/Peninsular until I put up more info on Lukpanic, which I am designing utterly a priori.

[Basilius]

Well, let me make my case further: Ke'idû'ûs'as is listed as a Class I noun, with no comments about it declining differently than others. (Its presumed head, ke, is class VI, so it's not inflected according to that.)

Good point.

Your observation may suggest that the inner structure of such formations was already somewhat obscured in Mûtsipsa', so they were treated as single words. Which makes me wonder what was the normal way to construe "adverbials-as-attributes"...

Also, since the case of the head noun affects the meaning of the postposition I suspect that changing it might be avoided.

It seems that what affects the meaning of a postpositional phrase in PI is not the case of the head noun specified by such phrase but the case of the noun governed by the postposition; in other words, the case of dúy-uys and not of káy-i.

"East of the sea" is a great name. OTOH there doesn't seem to be a postposition that quite fills the role of "of" in that phrase; nor is there one meaning "beyond" or "across", so we'd have to improvise something. [...]

Actually, my idea was rather something like 'East (which was) before the sea', i. e. hápa(-yi) dúyuys háq or somesuch. Its Mûtsipsa' reflex could be borrowed into Fáralo as e. g. He(i)fadəse(i) (with ə for û or ûû, and -sh- simplified)... not especially short, again :(

But, an option I haven't thought of so far... The PI for '(the) east of (the) sea' is, literally, just dúy hápa. Reflexes of this one must be short enough :)

It seems like such a language would likely be related to Legion's Isles lang, the one we don't know anything about yet...

IMHO geographic proximity does not necessarily imply close linguistic kinship; the people in question could represent e. g. an earlier wave of migration.

So there is still an option of modifying whatever PI name we like to make its sound as mellifluous as we may wish :) E. g. dúy hápa -> Tiwsaffa... mmm?

[Corumayas]

It seems that what affects the meaning of a postpositional phrase in PI is not the case of the head noun specified by such phrase but the case of the noun governed by the postposition; in other words, the case of dúy-uys and not of káy-i.

Oh, you're right of course. Sorry.

Actually, my idea was rather something like 'East (which was) before the sea', i. e. hápa(-yi) dúyuys háq or somesuch. Its Mûtsipsa' reflex could be borrowed into Fáralo as e. g. He(i)fadəse(i) (with ə for û or ûû, and -sh- simplified)... not especially short, again

(For what it's worth, the vast majority of Mûts. > Fáralo loans show u for û...)

But, an option I haven't thought of so far... The PI for '(the) east of (the) sea' is, literally, just dúy hápa. Reflexes of this one must be short enough

Isn't that a possessive construction? I'm not sure it can be used with a locative meaning... OTOH it does make a nice simple name (Duuhefa in Mûtsipsa', I think).

IMHO geographic proximity does not necessarily imply close linguistic kinship; the people in question could represent e. g. an earlier wave of migration.

So there is still an option of modifying whatever PI name we like to make its sound as mellifluous as we may wish E. g. dúy hápa -> Tiwsaffa... mmm?

Sure, that's true.

[Radius Solis]

Actually, I like the idea of a short, one-morpheme name that extends a common noun, adjective, or the like into a placename. In Mûtsipsa', good candidates could be:

Hefa (east) - like we divide the real world into the East and the West
Pitih (under) - short for "under the horizon" (c.f. "Down Under" for Australia)
Mwa (mother) - turning "mother" into a proper noun for the motherland

[Cedh]

I'd like to announce that I have completed the Kuyʔūn Tsinakan text. In the grammar, only the sections on clausal syntax and on derivation are still missing, although there will probably be some minor adjustments to other sections (for instance I'm planning to include a comparative example sentence in the dialects of all three cities).

Comments and critique are welcome!

[Corumayas]

I like it! The idea of these Dāiadak outposts deep in the Xōron Eiel is cool. And so is the grammaticalization leading to the verbal classifiers and voice markers, and the lexicon... nice work!

(I'm not sure how much sense using the name "Meshi" makes, since by this time the heyday of the Meshi is 2000 years in the past; they might not still be around, and if they are they'd probably be using a different name by now-- or at least a soundchanged one. Though I guess it's not that much weirder than modern "Hungary"...)


On an unrelated note: I'm wondering about Akana's paleogeography-- which could become important (in terms of biomes) as we start to work on the eastern continent et al. (btw, shall we go with Dûheh, or Duuhefa/Hefa, or one of the other suggestions...?).... I think I can see how all the continents used to fit together, but I'm not sure how long ago they broke up. Longer ago than Earth's continents, more recently, or about the same? Any ideas?

[Zhen Lin]

I'll be posting a minor update of Proto-Peninsular soon. Mostly just adding ideas for grammatical features, e.g. the option of having Austronesian alignment:
thuriarxa mā fiksa (ham) tankapixrisku. The fisherman caught (a) fish.
fiksa (ham) mā thuriarxa tankapipsisku. The fish was caught by a fisherman. / A fisherman caught the fish.

(I really ought to be working on my homework though. Ah well, procrastination.)

(2) An index of typological features found in the languages of Akana. (Particularly interesting would be to have an index of historical changes, a bit later.)

We can do this partly on CALS.

[Basilius]

Ehm. Well. So.

It seems I've designed another Isles language: Affanonic, spoken by Affanons (grammar, lexicon).

The story is kinda typical (of me, at any rate). You may remember that I mentioned that the name of the original homeland of the Isles people(s) could sound like *Tysæ or *Tiwsaffa or somesuch in some Isles language with appropriate historical phonetics. My mistake was that I failed to stop thinking of it immediately, but started to mull over the sound changes instead. At some point I imagined it would be a nice by-project: a simple and conservative Isles lang, to be finished in three evenings or so.

The trouble is that it didn't agree to my plot. It insisted that it wanted sophisticated declensions, a lot of forms ending in -CjVCjVCjV, and a tricky sound development scenario (leading, among other things, to penis for 'star').

Anyway, it seems that I've finished documenting what I've discovered about it so far.

The original idea was that Affanonic was spoken around -200 YP in a short-lived colony somewhere on the Huyfárah coast, and that its speakers had been ousted or assimilated by the hayday of Fáralo. I don't know if Affanons had any colonies in other places, but it seems that their ancestors were the earliest westward migration wave of the Isles peoples and were the first to contact the inhabitants of Peilaš. If Affanonic doesn't fit in the history of Huyfárah, it can be moved somewhere else.

And, in case anybody wondered, the genuine Affanonic name for the Isles proto-homeland is Tysaffai. Because the dictionary form of nouns is now nominative rather than bare stem. I don't know what it could become if borrowed into (archaic?) Fáralo.

[Radius Solis]

And, in case anybody wondered, the genuine Affanonic name for the Isles proto-homeland is Tysaffai. Because the dictionary form of nouns is now nominative rather than bare stem. I don't know what it could become if borrowed into (archaic?) Fáralo.

Possibly something like T*safay, where * might be any of u, i, or uy. If Zomp elects to borrow it, in whatever form, I'd be in favor of that as our general working name for that continent, in keeping with our general tendency to use Fáralo names. If not then the form as it stands in your own language will certainly suffice too.

It's good to see another Isles lang! Quality work, I like how it turned out. Affanons has a feel all its own, but with an unmistakable Isles flavor. Will you be offering a text sometime? I'd like to see a sample.

[Basilius]

Quality work, I like how it turned out. Affanons has a feel all its own, but with an unmistakable Isles flavor. Will you be offering a text sometime? I'd like to see a sample.

Thank you! I am still missing a few lexical items necessary to finish "The North Wind and the Sun". The first sentence as it looks right now:

Affaini Rirkjun Fø fy Affamoi hus mimyti,
One day the North Wind and the Sun disputed (the following),

y sanjun tsera sunol nini,
who was ( = is) stronger than the other,

fy hus [agree (v. pf.)],
and agreed (on the following),

tsjutsuhun nyral jonommos munnawis ri sanjun tsera titsjuma reffis,
he who (would) remove, more quickly than the other, the clothes from a traveller (who was) passing by,

pipys [victor=ACC] piffjuti jawa.
(he) should ( = will) be called the victor.

('The following' renders the cataphoric use of ri.)

I could cheat and make a litteral translation from PI, but the Affanonic syntax seems already flexible enough to take up the challenge of translating the less adapted English version.

Possibly something like T*safay, where * might be any of u, i, or uy.

Why not -iw-, BTW?

If Zomp elects to borrow it, in whatever form, I'd be in favor of that as our general working name for that continent, in keeping with our general tendency to use Fáralo names.

Exactly my plot :) That is, the first message about the NE continent came to the Fáralo people with the Affanons, and then the Fáralo people spread the news.

[Basilius]

Interesting. I've got used to think that the Proto-Isles as described by Ran is the proto-language for the Isles group in the strict technical sense. That is, everything in the daughter langs has a source in the protolanguage and has developed in the natural way (with the obvious exceptions like borrowings etc.). I sort of ignored Ran's own words about his description relating to a particular "dialect".

Ok. Just of curiousity, I compiled a list (probably incomplete) of morphosyntactic features that looked difficult to explain, based on all of the available descriptions of the daughter langs. I made abstraction of all the potential issues in historical phonetics, which seem to be much more numerous.

The list is copypasted below. And it makes me think that my original vision of the development of the group was not correct. At any rate, it would be much easier to imagine that at least some features in each daughter language have no source in Ran's PI but reflect some peculiarities of a closely related but different dialect of "Broad Proto-Isles".

This can be perceived as a problem, but there seems to be another option. It could be a pretext to start a new fascinating conlinguistic game. That is, to check the list carefully, trying to explain away as much as possible, and use the rest as the basis for reconstructing an older protolanguage. Call it Proto-Proto-Isles, Archaic PI, Proto-Broad-PI, or whatever.

What do you people think?

The list:

- - - - -

Ppãrwak
* "Noun marker" e // i. e. nominal group marker; is obligatorily used after case endings but not after postpositions
* Nominative ending -(i)p.
* Accusative = bare stem ( = attributive form); in principle, can be explained phonetically, but this will require an exception for postpositions looking like mergers of the PI accusative ending with a PI postposition
* Suffix of substantivized verbs -(i)p (attested in the sample text).
* Position of adjectives: can be placed after the head noun but before the nominal group marker -e.
* Tense distinctions in subjunctive: the subjunctive marker -ppwak can be added to the stem of any tense (rather than only to the lexical stem <= non-sensory past)
* Gender (and formality) markers in personal pronouns (difficult to derive directly from 'daughter' and 'son' for the respective PI phrases would mean 'my daughter' etc. rather than 'I who am daughter' etc.)

Zele
* SVO as the default WO
* OVS in "passive" clauses
* Genitive (historically = bare stem) can be put after its head noun
* Dative ending < háq 'in front of'? (attached to bare stem?)
* Ablative ending < hapaq 'behind'? (attached to bare stem?)
* (Prefixed) "go denotes that the noun is the object of a relative clause."
* Appositive marker zhi is indeclinable (clearly from nom. of the PI essive/equative marker)

Mûtsipsa' (presumably, also Kozzaŋ Fasa?)
* Adjectives must agree with their head noun in case and number (and in declension class???)
* Adjectives are mostly prepositive, but can "be placed anywhere in a clause".
* Adverbs "can technically go anywhere in the clause".
* Postpositive negation particle of unclear origin.
* Exception to V-last WO rule: in passives, the agent in instrumental (or benefactive) must be placed after the verb
* The conjunction 'and' when connecting clauses must be in nominative (rather than a form based on bare stem)
* Position of 'and' connecting two nominal groups: "cart horse and" for 'cart and horse'
* Conjunctions other than 'and': "the dog the meat ate the wolf or" for 'the dog or the wolf ate the meat'
* Postpositions with nominative and benefactive
* Postpositions put before a pronoun (found in the sample texts)

Máotatšàlì
* WO: (strictly?) V-first
* Lexical derivation patterns based on metatony
* Tense and mood system (tone-based distinctions in particular)
* Prefixed aspect markers
* Tone-based distinctions in prefixed aspect markers
* Prefixed intensity markers in adjectives/adverbs

[Corumayas]

If it interests you, do it!

At first glance, none of these features seem to be shared by multiple languages. I imagine some could be the result of influence from neighboring or substrate languages; e.g., Máotatšàlì's word order et al. could've been influenced by Nualis.

I'd like to see the phonology issues you've found, too. Maybe some hints for subgroupings in the Isles family could be found there...?

[Basilius]

If it interests you, do it!

Thank you for encouraging me :)

I think this needs to be approved by the community: in fact, my proposal amounts to admitting that what we considered the protolanguage of an important group is not exactly *the* protolanguage (i. e. closest common ancestor).

Besides, it looks like too big a piece of job for one person. And I suspect that somebody can find an easy explanation where I can't imagine any verisimilar scenario.

At first glance, none of these features seem to be shared by multiple languages.

In fact, a few very important ones are. Postpositive epithets of nouns, various arguments placed after the verb, non-canonical case choice with postpositions, subjunctive marker freely added to all tenses...

I imagine some could be the result of influence from neighboring or substrate languages; e.g., Máotatšàlì's word order et al. could've been influenced by Nualis.

Probably, I ought to provide some hints about the point I'd like to arrive at.

Suppose two languages contacted, say, around -500 YP, and exchanged a lot of loans. What was the phonetic form of such loans in the donor language? What were the modifications necessary to adapt them to the phonetics of the recipient language? Which derivation patterns were productive by the moment of contact, and which already marginalized? I'd like to know how such questions can be answered.

A reference to substrate &like does work in the sense "this is not completely unexplainable", but IMHO it doesn't help to understand the specific sequence of events that could lead to the attested state of affairs.

For example, what intermediary stages would you propose for the Máotatšàlì WO? I mean, not considering anything based on speakers' conspiracies ("Those linguistic geeks nuts drive to, Japanese do like head-final word order use us let") :)

Maybe some hints for subgroupings in the Isles family could be found there...?

This task is much more laborious. At a glance, Máotatšàlì tones look most problematic; together with the unusual WO development, they may suggest that it could descend from the most deviant dialect of "Broad PI".

Otherwise I can only point to a few issues with Ppãrwak (actually, I have an unfinished text documenting them). For example, it preserves stem-final consonants in all forms of verbs, including those based on PI non-sensory present (where stem-final consonants were ousted by -s in Ran's PI). I considered devising a complex scenario (with intricate sound changes and analogical levellings) to explain this, but now I'm inclined to think that it was simply a feature of the respective dialect of "Broad PI".