The Allosphere

[Nortaneous]

(For now, this will just be collecting information from the old thread.)

The Macro-Vengic families are spoken in the northwest of the Large Continent. They are divided into five subfamilies. The main distinguishing features between these are the reflexes of Proto-Macro-Vengic *N and the regular stress pattern.

The original Macro-Vengic stress system involved mostly-predictable penultimate stress, with stress-attracting long vowels. Of the five subfamilies, Vengic preserves the original system, Hluic and Zotic regularize to penultimate stress, Narngic shifts to initial stress, and the situation in Whatic is more complicated; it seems that Proto-Whatic developed unstressed vowel apocope during a period of confusion of stress, and later became averse to open syllables and added reduplicative material or /h/ to the ends of some vowel-final words.

Innovation in the pronoun system is also common, especially since the first- and third-person pronouns tend to become similar.

A common feature of Macro-Vengic is a distinction between 'minor' and (reduplicative) 'major' forms of verbs. This distinction undergoes various semantic developments.

Here are some cognate sets.

Code: Select all

Proto-Yia       Proto-Vengic   Proto-Hluic     Proto-Zotic     Proto-Narngic   Proto-Whatic    
*ʔaβeŋi         ʔβɪŋi          ʔbæɲ            ʔawəɲ           ʔœyɲ            ʔaheɲ
*mekaa          mkaa           mɣa             miɣa            mek             ---
*hoteŋa         theŋa          tteŋ            ðeŋa            hodəŋ           wətaŋat
*nakasa         ŋkasa          ŋgas            nɣasa           nagas           nakas
*fukee          fuke           kke             huɣe            fuk             fəkeh
*cĩ             ciŋ            sæ̃              cĩ              cĩ              ceŋ
-------------------------------------------------------------------------------------------
                V'eng          Hlu             Zot             Narng           What
                ʔβɪŋ           ɓɤɲ             ʔɔɔɲ            ʔœyɲ            ʔjhiəj
                mka           mɯ              ɲɤɤ             mek             ---
                tɦɛŋ           ndɯɲ            zeŋɔ            ʔodn            mtŋɯʔn
                ŋkah           ŋgaɬ            nɤɤjh           narʔ            ŋkarh
                fke            ŋgi             huu             huk             pkeh
                ciŋ            ɬe              sẽ              sĩ              ceŋ
-------------------------------------------------------------------------------------------
                Vynyi          Gyitha                                          Ahemai
                vəɲi           ciə̃                                             ji
                ngo:           ɦɯa                                             ---
                thaŋa          ə̃tẽ                                             bləʔài
                ---            ə̃ŋaθ                                            nəʔài
                ke:            ə̃ge                                             wʔè
                cin            siə                                             sa

Vengic is divided into two subbranches: Eastern Vengic, represented solely by Vynyi, and Everything Else.

Vynyi is a highly conservative language, which generally preserves post-stress syllables, but instead has been influenced by ??? (Central Montane? If so, probably Deghuric subgroup) and developed agglutinativity, a decently large case system, and, true to Central Montane form, a bewildering variety of essive and lative verbalizers.

Vynyi reflects *N as a palatal nasal (or /mj/ before /u/), with retroflexion on the following vowel.

The phoneme inventory of Vynyi is as follows:

/pʰ p b tʰ t d tʃ dʒ k g/
/f s ʃ ɬ x h/
/β z l j ɣ/
/m n ɲ ŋ/
/a e ə o i u/ + nasality, retroflexion, and length

The singular pronouns are:
1SG. ao, preserving PMV *aqo
2SG. ur, preserving PMV *ru without the d- element common in Everything Else
3SG. kyun (animate) / kyaan (inanimate), innovative forms from a circumlocution *coʔV kuyaa/kuyu mangV "he who/that which one sees" -> "that person/thing" (cf. V'eng co', n-kya, mang)

V'ëng is not the most significant of the Everything Else languages, except from the future-historical standpoint: it develops into Zzyx-wqnp /ⁿdẑ̩wɒ̀ɴ/, the language of the Empire of Zzx-zzyx /dzə̂ⁿdẑ̩/. But it is a fairly typical Everything Else language.

/p t s~tʃ k ʔ/ <p t c k ʼ>
/b d z~dʒ/ <b d z>
/f s̺ x h/ <f s x h>
/m n ŋ/ <m n ŋ>
/β r j ɣ gʟ/ <v r y g l>
/a ɛ ɔ ɪ ʊ i u/ <a e o ë ö i u>
There is an additional falling tone, from h- -h, transcribed as [ɦ] for some reason.

Here is a fragment from the Tower of Babel:
Yëʼ e s yah gö ru s vëng mka.
[jɪʔ e sjaɦ ɣʊ ɾu fɪŋ mka]
yëʼ e s=yah gev ru s=vëng mka
world of 3POSS=whole PST have 3POSS=language one
Now the whole world had one language.

ʼA theng gö ʼoc leg, ʼa xë s re e Sinar, ʼa sitsir ga.
[ʔa tɦɛŋ ɣʊ ʔɔʃ gʟɛɣ ʔa xɪ zɾɛ ɛ ˈsinað ʔa sitsið ɣa]
ʼa h-teng gev ʼ-oc le~g, ʼa xë s re e Sinar, ʼa sir~sir ga
then COLL-man PST ADV-east move~MAJ, then find plain of Shinar, then settle~MAJ there
As people moved eastward, they found a plain in Shinar and settled there.

Theng veng zi, "Hë ca vʼa rig, hë ca pʼyuz." Myan rig öng nha, myan dur öng ntë.
[tɦɛŋ βɛŋ dʒi hɪ tʃa ʔβa ɾiɣ hɪ tʃa pʼjuʒ mjan ɾiɣ ʊŋ nɦa mjan duð ʊŋ ntɪ]
h-teng veng zi, hë ca ʼ-va rig, hë ca ʼ-byu~by. mya~my rig öng nha, myan dur öng ntë
COLL-man say REFL, COLL JUSS 1-make brick, COLL JUSS 1-bake~MAJ. use~MAJ brick against rock, use~MAJ tar against mortar
They said to each other, “Come, let’s make bricks and bake them thoroughly.” They used brick instead of stone, and tar for mortar.

The singular pronouns are conservative, aside from the d- element in the 2SG:
1SG. a' ( -> Zzyx-wqnp wqp)
2SG. dru ( -> Zzyx-wqnp trvp)
3SG. a ( -> Zzyx-wqnp wq)

to be continued

[Nortaneous]

Another family is Macro-Amqolic. It is divided into two subfamilies: Amqolic, consisting of the languages Amqoli and Rengni, and the divergent language Deghuri.

The basic phoneme inventory of Macro-Amqolic is very large:
/pʰ pˠʰ tʰ tˠʰ tsʰ tsˠʰ tʃʰ tʃˠʰ kʰ kʷʰ/
/p pˠ t tˠ ts tsˠ tʃ tʃˠ k kʷ/
/bʰ bˠʰ dʰ dˠʰ dzʰ dzˠʰ dʒʰ dʒˠʰ gʰ gʷʰ/
/b bˠ d dˠ dz dzˠ dʒ dʒˠ g gʷ/
/sʰ sˠʰ ʃʰ ʃˠʰ xʰ xʷʰ/
/s sˠ ʃ ʃˠ x xʷ/
/zʰ zˠʰ ʒʰ ʒˠʰ ɣʰ/
/z zˠ ʒ ʒˠ ɣ/
/m n/
/mˠ nˠ/
/l lˠ r rˠ j w/
/ɗ/
/h/
/a ə i u/

The consonants written with velars were actually realized as clusters with a following velar element: a plosive in the case of unaspirated consonants, a fricative in the case of aspirated ones, and a velar nasal in the case of the nasals. This system is preserved despite the loss of the aspiration system, which occurred everywhere, albeit with some compensation. Deghuri debuccalized all aspirated fricatives and lengthened high vowels before voiced unaspirated consonants, but otherwise simply dropped aspiration; Rengni had some tonal developments; and Amqoli simply lost the system without compensation. The labiovelars are shifted to implosives everywhere, and are only reconstructed because in Deghuri the labiovelar element is lost when preceding a consonant, and in Proto-Amqoli-Rengni they shift following PMA schwa to *o.

Amqoli has palatalized *k *g to ch j before front vowels, and merged them as q before back vowels. All k g are from cluster resolution, mostly *kl *gl. It has also, along with Rengni, lost schwa and developed a complex five-vowel system, but native disyllabic roots cannot contain all possible vowel sequences: high vowels can only be followed by themselves, o can (except when preceded by certain consonants) only be followed by u, and mid vowels cannot appear in second syllables at all. The single PMA implosive has become a palatalized lateral.

A characteristic Amqoli development is the shortening of roots due to widespread syncope, resulting in the bipartite division of inflectional classes by whether or not syncope has applied. The complex consonant system has partially collapsed, with *Px- giving rise to a new aspirate series.

Morphologically, Amqoli is characterized by the loss of core case marking and the development of a proliferation of plurals and causatives.

Deghuri preserves the complex consonant series and, as implosive labials, the labiovelar plosives; it has also shifted PMA *a *e to aa a, and innovated a series of long high vowels. Its main phonological innovation is the loss of sonorant clusters.

There is some morphosyntactic restructuring due to contact with Kangshi. The Kangshuic mode system has been, generally speaking, adopted, and a distinction between alienable and inalienable possession has developed. The baroque locative system of Amqoli has been pared down somewhat, if not fossilized, and possibly there is split ergativity originating from earlier participial constructions.

Also as a result of contact with Kangshi is the loss of *l and shift of initial *r to the implosive dh. The length system is mostly reflected in quality: a i u are mid vowels whereas aa ii uu are high. This has the hilarious result of reversing high and mid vowels in borrowings from Amqoli, because borrowing patterns were conventionalized before this happened.

The contrastive stress system of PMA was lost, mostly without compensation; stress in Deghuri is ultimate unless the penultimate vowel is long.

(edit: this should be redone. evidently the Kangshuic language that Deghuri would be in contact with is something called Oahshw, which has an inventory of p t k ʔ (pʰ) tʰ kʰ ɓ ɗ ɠ m n ŋ s h w r j, and lots of tones)

Rengni remains to be worked out. Presumably there is a great deal of influence from Narngic, an otherwise insignificant branch of Macro-Vengic. Probably the q-relativizer of Amqoli and (maybe) Deghuri either fails to develop or is lost, and relative clauses work completely differently.

[Nortaneous]

Not quite a language family, but a notable area of linguistic convergence both abstractly and geographically similar to our Paleosiberia, is Haruan, consisting of the unrelated languages Kett, Arve, Sestmag, and Kannow, and probably some others. (If the shared conworld thing is still happening, this would be a good place for something, although it's as far as you can get from Tsalaysia.)

Notable features include the cod-Austronesian alignment that was popular on the ZBB about a decade ago, a two- or three-way evidentiality system, word order that places the verb close to the beginning of the sentence (Arve has strict VOS emerging from the fusion of topic pronouns with the auxiliary verbs; Kett and Sestmag have not undergone this development, and instead are vaguely V2; and Kannow has fairly simple VSO), an elaborate aspect system (clearly a late development in Kannow, and formed with auxiliaries), a general lack of nominal inflection other than the topic marker, benefactive constructions, and noun incorporation or N-V compounding. There is, however, variance in person marking: Kannow and Arve can mark up to four arguments on the verb complex, but Kett and Sestmag lack it entirely.

As for phonology, all languages except for Kannow have a strong and unpredictable stress accent. A fortis-lenis contrast (in Kett, Arve, and Sestmag) and a system of rounded plosives (in Kannow and Kett) are reconstructible; however, in Arve, the fortis-lenis contrast gave way to a three-way length contrast on vowels, and in Kett, *tw *kw have become a bilabially trilled affricate and a bilabial implosive, or a labial-alveolar and a labial-velar, depending on the dialect. (It is possible that labial-alveolars and labial-velars were their original values.) Fortis consonants may appear initially and finally, as shown by Written Arve satt (an auxiliary) and the names of the languages Kett /ket:/ and Sestmag /ɕ:maˤ/.

In terms of lesser phonological developments, *st > ɕ is shared by Arve and Sestmag; Arve, in fact, eliminated clusters altogether, and in doing so greatly expanded its consonant inventory.

[Nortaneous]

It would probably be wise to elaborate on the voice systems of Macro-Amqolic.

PMA had the following affixes:

Voice:
*-pi- passive
*-lu- causative
*-ri- reflexive

Tense:
*-y- ~ *-i- past
*-ay- past 2 (possibly there was an aspect distinction?)

There was also a deverbal noun of action in *-ngew-, which is relevant in Deghuri.

In Amqoli, the causative system has been extended with an affix descended from the verb *mgéde, creating a distinction between intentional and unintentional causatives:

mtaq zhim chuphotsmod
[mtaʁ ʒĩ tʃuˈpʰotsmod]
rock 1S.ACC 3INAN.L-fall_down-CAUS2
I tripped on the rock (= the rock made me fall down)

zha njaryo da bqele phagrul
[ʒa ⁿdʒaˈjo da ˈbʁeje pʰaɣˈruɫ]
1S.NOM wife DAT dishes-PRE 3F.T-wash-CAUS1
I'm having my wife wash the dishes

This doesn't exist in Deghuri: (note the general loss of final short vowels, and the loss of *l in Deghuri)

matāk am šupkassu (-ts-l- > -ts-s- > -ss-)
[mɤˈtak ɤm ʂopkɤsˈso]
rock 1S.ACC 3INAN.L-fall_down-CAUS

ā mārāšaɣ bɣīm dapākkū
[a maˈɾaʂɤʁ bʁim dɤpakˈku]
1S.NOM wife-1S.POSS dishes-GEN 3F.T-wash-CAUS

On the other hand, Amqoli lost the passive and innovated a generic pronoun:

bqeli du mpagru (m-h- > mp; *p > h)
[ˈbʁeji du mpaɣˈru]
dishes 4 3INAN.L-wash
the dishes are being washed

Whereas Deghuri preserves it:

bɣī mapākkup
[bʁi mɤˈpakkop]
dishes 3INAN.L-wash-PASS

The Amqoli reflexive-reciprocal is cognate to the Deghuri reciprocal. Deghuri has no reflexive, and instead has an innovated middle voice, similar to the archaic English 'passival':

Borju thagluri (l~r dissimilation)
[ˈbordʒu tʰaɣˈɫuji]
Borju 3M.T-wash-REFL
Borju is washing himself

Būrju Dhišāv sapākkuri
[ˈburdʐo ɗeˈʂau̯ sɤpakkoˈɾe]
Būrju Dhišāv 3M.T-wash-RECIP
Borju and Renxau are washing each other (lol gay; do note that this is an Enze name, adapted with initial r > implosive dh)

(The Amqoli reflexive can also take the reciprocal use, but you'd feed it a NP with a conjunction: so you'd translate the above Deghuri sentence as Borju Reshab sha thagluri, with sha 'and'.)

As for the Deghuri middle voice:

Būrju pākkuyūm sadha
[ˈburdʐo pakkoˈjum sɤˈɗɤ]
Borju wash-PTCP-GEN 3M.T-be

Unlike Kangshi, Deghuri doesn't have an antipassive; you would use a generic object instead. (Then again, Kangshi doesn't have generic objects - right, H13? - so maybe this is a bad idea. This seems like the sort of difference that could persist, however, especially due to continued contact between Deghuri and Amqoli, which likes its generics.) One relevant difference between Deghuri and Amqoli here is that Deghuri doesn't preserve the partitive, so to reintroduce the object you would instead use the genitive.

[Hallow XIII]

If the shared conworld thing is still happening

Of course it is! Now, what did you want to know? Oh, Kangshuic. Let's see...

Okay this is way overdue and it won't be very satisfying but I'm going to try and get the big elements of Kangshuic diachronics out of the way here. So, first things first, the approximate shape of Proto-Kangshuic. Phonologically, Proto-Kangshuic had a consonant inventory of approximately something like this:

Code: Select all

p t     k q
b d 
  s   ʃ x
  z   ʒ ɣ
m n     ŋ
w r l y

It allowed a maximum syllable of CCVC and had seven tones, which were partly distinguished by pitch and partly by phonation (breathy/modal/glottalized). I am significantly less certain on the vowel inventory, which may have had a length distinction or contained diphthongs, but probably one of them. In any event the pure vowel qualities were probably something like /i ɯ u e o a/, but until I've finalized all the sound changes, who knows.

Morphosyntactically, the broad strokes of PK are something like this: PK was a moderately synthetic language with about an equal amount of prefixation and suffixation. Person marking, that was already weakening, was prefixal, as were certain markers of mood and nonfinite markers, while the bulk of TA marking happened suffixally. The language was split-ergative and had a nominal case system of five items: Absolutive, Genitive-Ergative, Accusative(-Allative), Dative(-Comitative) and Locative. There was already a noun class system, derived from a system of numeral classifiers, which is continued in most branches of Kangshuic. The strongly variable morphosyntax in the various descendants is a legacy of the more clitic-like nature of the class markers in PK. There was some amount of nonconcatenative morphology: some combinations of suffixes, due to contraction of complex codas in PPK, produced a marked stem change, and the third-person object marker was a tone change.

The categories distinguished in the verb are person marking for both agent and patient (agent marking is identical to S marking for all verbs), aspect (unmarked, progressive-imperfective, completive, inchoative), tense-mood (unmarked, past, remote-irrealis) and two participle forms that are otherwise unmarked (approximately present and past). Voice was indicated either through simple detransitivization or through periphrastic constructions.

Obviously, there were many changes from this to descendant languages. The Kangshuic family divides itself into around four branches, although the identity of the branches with actual genealogical points of divergence is by no means clear and several isoglosses transcend branch boundaries. In the absence of geographical information other than some MS Paint maps, I will refer to them as Kangshuic A, B, and C, as well as Pirka, a single language that forms a branch of its own.

Pirka is distinguished from all other Kangshuic languages by the simple fact that it did not undergo Kangshuic final loss, a sound change affecting all other families in which final obstruents and nasals were lost, with fricatives vanishing altogether, all stops merging into the glottal stop and final nasals leaving behind vowel nasality. Pirka, however, almost fully retains the original PK inventory of finals, modulo its global loss of the voicing distinction. However, the distribution of finals is significantly different from PK, depending largely on whether the syllable tone was originally glottalized or not. Pirka also only has two tones, merging the original seven first to four, with two binary distinctions of high vs. low and glottal vs. modal, and later losing the glottalization distinction with a concomitant consonant split. It shares the innovation of aspirated stops from fricative-stop sequences with large parts of Kangshuic A and B. Its vowel inventory is /i ĩ e ə ɛ a ã ɔ o u ũ/.

Morphosyntactically, Pirka is extremely innovative, having lost basically all inherited verbal morphology and being highly isolating compared to other Kangshuic languages. The noun class system in Pirka is expressed through determiners that obligatorily accompany nouns, and unusually for Kangshuic languages express a definite-indefinite distinction. The case system is completely gone and has been replaced through various serial verb constructions, like in large parts of Kangshuic A.

Kangshuic A is the best-attested and most widely spoken branch of Kangshuic. Old Kangshi, ancestor of Court Kangshi and several related dialects, is the earliest (and in continuity with the modern forms, basically only) written Kangshuic language, and already shows several distinctly Kangshuic A features, especially in phonology and morphology. Kangshuic A has full loss of final consonants, reflecting their influence in vowel quality and tone, and all languages in the branch have undergone a change of *w > b, *y > ɟ, *r l > dʐ. Old Kangshi reflects these as /b z ʐ/, respectively. The vowel inventories of Kangshuic A languages are generally quite large, including generally six to eight monophthongs, with at least one back unrounded vowel, and a large array of diphthongs. Significant parts of modern Kangshuic A languages also underwent a sound change in which alveolar sibilants were lateralized, with Court Kangshi one of the notable exceptions.

Morphosyntactically, Old Kangshi still preserves the PK case system, although especially oblique case information is often redundantly encoded by verbs even in early texts. Modern Kangshuic A generally does not, although the Ergative has become a topic marker in some of them. The verbal paradigm is generally completely restructured, since PK suffixation generally either fossilized as stem change or was lost entirely in favor of serial verb constructions. Court Kangshi is an example of both developments, indicating the categories of version and subjunctive by various forms of stem change, with principal mood marking occurring by a realis clitic. TAM, however, is handled by derivation or serial verbs. The loss of the case system and person marking on the verb generally also made word order less variable than it was in older stages of the language.

Kangshuic B and C are both more scattered and internally varied branches. The principal change affecting Kangshuic B was widespread vowel epenthesis, followed in many cases by loss of final vowels, creating a distinctively unkangshuic CVCVC(V) word pattern. They generally preserve cases, but have innovated a much more regular verbal TAM paradigm. Kangshuic C languages, being the farthest from the core Kangshuic area and in contact with both other Qoic as well as Zotic languages, make heavy use of phonation distinctions at the expense of tone, and preserve the front rounded vowels that result from fronting before alveolar finals. Morphologically, they are both more heavily head-marking and prone to univerbation than other branches of Kangshuic, including consistent verbal person marking, often a preservation of some verbal aspect-pairs and the nonfinite forms of PK, but also an innovation of new distinctions.

I'll have to go into more detail on each of these in separate posts.

Well, that was more than two years ago, so let's get on those separate posts, shall we.

[Hallow XIII]

So, Kangshuic C. As mentioned above, Kangshi itself is probably not in contact with Deghuri! This is because it has to spawn A Civilization Spreading Religion To The Tsi, and it is probably hard to do that if you are in mountains that are very far away from the sea. Admittedly the Mongols are Vajrayana Buddhists now, but.

Anyway, what is probably in contact with Deghuri is Kangshuic C. Kangshuic C shares the derivation of its tone categories with pretty much every other branch of Kangshuic, which is... not immensely complicated, but still somewhat involved.

Proto-Kangshuic-Pirka (= Pre-Proto-Kangshuic but I no longer like that name) had exactly one suprasegmental, which was a glottalization distinction on syllables. We will call glottalized syllables type A, and the others type B. The rest of the tone generation depends on initial and final consonants, as well as preinitials. Proto-Kangshuic-Pirka had epenthetic vowels between the consonants of initial clusters, which meant that those consonants generated tone as well. For the purposes of Kangshuic C, the important thing is that voiced obstruents spread breathy voice. The tones were then generated as follows:

(Where C = voiceless obstruents G = voiced obstruents, N = nasals, 0 = everything else; and X = glottalized, H = breathy)

Code: Select all

A/CV, A/GV, A/NV, A/0V, B/CV, B/GV, B/NV, B/0V -> 55X 35X 55X 55X 33 11H 11 11
in -C -> 55X 33X 55X 55X 33 11H 11 11
in -G -> 53X 33X 53X 53X 31H 11H 31H 11H

and -0 and -N are same as open syllables.

Since only breathiness can float from preinitials, and only on nonglottal syllables, we are left with the following tone categories:

Code: Select all

55X 35X 33 33H 31H 11 11H

The non-breathy low tones are dragged up from 33 11 -> 55 33, leaving the following seven categories for Proto-Kangshuic C:

Code: Select all

1 (55), 2 (55X), 3 (35X), 4(33H), 5 (31H), 6 (33), 7 (11H)

.

As far as the consonants are concerned, the shifts are relatively simple: the palatal sibilants merge into the alveolar ones.

The vowels and finals behave in a spectacularly irregular fashion, generating enormous vowel inventories in almost every Kangshuic C language, which we will go over in the next post.

[Hallow XIII]

So, Oahshw! Which I just discover should be Eahswa. Or similar. Anyway.

Eahswa [ʔeah³³sɯə⁵⁵] has the following maximal syllable structure: C(j)(w)V(j,w)(ʔ,h)

Diachronically, initial /b d g/ shifted to /pʰ tʰ kʰ/ when not in a cluster and /p t k/ otherwise; simultaneously, initial /p t k/ were imploded to /ɓ ɗ ʄ ɠ/ in type B syllables. /q/ became a glottal stop, /z/ became /r/, /l/ merged into /n/ and /x/ merged into /h/. Dialectally, /pʰ/ also shifted to /h/ and velars palatalized before /j i/, leaving the following maximal inventory of initials:

Code: Select all

p t (c) k ʔ (pʰ) tʰ (cʰ) kʰ ɓ ɗ (ʄ) ɠ m n (ɲ) ŋ s h r
<p t c k ʔ ph th ch kh b d j g m n ñ ng s h r>

The medials /j w/ are spelt <y v>; in the sequence /jw/ they are realized as [ɥ] <ü>, or [ø̯] <ö> before mid vowels.

The vowels did a whole lot of things. The ~standard dialect~ has:

Code: Select all

ɒ ø y ə i u e o ɯ æ ɒə ao ea eə eo əɯ øə øy oə iə uə ɯə
<a ö ü i u e o w ä aǎ ao ea eǎ eo ǎw öǎ öü oǎ iǎ uǎ wǎ>

...Distributed in various ways around the various syllable types.

Of the diachronic seven tones, the ~standard~ retains four; it merges tone 2 into tone 3 and tones 4 and 6 into 5.

Some words:
^Bdo:d "black" -> ⁷thöă
^Andas "blue" -> ³täh
^Bšaŋ "sky" -> ¹saǎ
^Bmo:k "earth" -> ⁵mouʔ
^Brtaj "be born" -> ¹tea
^Bžbwŋ "between" -> ⁷pwă
^Apa "man" -> ³pă
^Bzdup "make" -> ⁷tviʔ
^Bto:ds "serve, service" -> ⁵döeh

Grammatically, almost all of the inherited morphology has been lost. New person markers have been innovated from older pronouns, and auxiliary verbs have replaced the inherited tense and aspect markers. However, the old mood markers are retained.

Syntactically... I need to figure out the syntax. Hit me if I haven't posted example sentences by the end of January.

[Nortaneous]

Man, syntax, who even knows syntax.

So! Hlu.

Hlu is, of course, a member of the Hluic language family. The characteristic feature of Hluic is the regularization of PMV stress onto the penultimate mora -- so *nakasa 'a god' > ⁿgahl, but *kusaa 'two' > kla (~ Gejaehl kwa). This is the same as the regular outcome of Vengic, but Vengic contains irregularities that the other branches do not, which can be reconstructed as words with irregular stress.

Other characteristics of Hluic are:

- *s *c *z > PH *θ *s *y
- *N > *r
- Palatalization of velars preceding *i *y, followed by umlaut of high vowels to schwa when followed by *a, and umlaut of *a to schwa when followed by a high vowel
- Lenition, prior to unstressed vowel loss, of intervocalic *t *k to *ð *ɣ
- Assimilation of many clusters resulting from unstressed vowel loss to geminates, which in Proto-Hluic could appear word-initially
- Assimilation of nasal + plosive or fricative clusters to prenasalized stops
- Lowering of nasal vowels, which are then denasalized in everything but Gyitha
- Dissimilation of *e to *æ (merging with *ĩ) when followed by a palatal
- Occasional reflection of pretonic rounded vowels as -β-

Here are some Leipzig-Jakarta list items in Proto-Macro-Vengic and Hlu, with V'eng (Vengic) and Vynyi (Eastern Vengic) cognates where available:

Code: Select all

PMV     Defn.   PHluic  Hlu     V'eng   Vynyi
qata    father  ʔað     ʔɔl     ʔat     ---
teŋa    man     teŋ     ɗɯɲ     teŋ     tama
mekaa   one     mɣa     mɯ      mka     no:
kúsaa   two     kʷθa    kβɔ     kʊs     kuso
mẽtu    fire    mað     mɔl     mɪ̀n     məndu
(d-)ru  2SG     ru      yu      dru     ru:
terũxu  root    trɔx    tɔx     trux    tsormu
Naxu    egg     rɤx     yɯ      drè     ɲərxu
nakasa  a god   nɣaθ    nɯhl    ŋkà     ---

(Note the loss of *ŋ in Vynyi -- this is probably due to Amqolic influence.)

Hlu is divided into two major dialects: Gejaehl and Bor. Bor Hlu is overall more phonologically conservative: it retains *r (vs. Gejaehl *r- > y-, *-r > 0), the distinction between Proto-Hlu *a and *ɔ (vs. Gejaehl *a > ɔ), *uy (> Gejaehl βi), and clusters of a plosive + *l (Gejaehl *l > 0 / C_). On the other hand, it's lost verbal number agreement and regularized the system of inalienable possession, merges *e into *i, and has *ly > y.

Presumably there are also differences in the aspectual system or something.

[gestaltist]

This was surprisingly fun to read.

I have one question. How did this romanization come about?:

Zzx-zzyx /dzə̂ⁿdẑ̩/

[Nortaneous]

I have one question. How did this romanization come about?:

Zzx-zzyx /dzə̂ⁿdẑ̩/

Zzx-zzyx has nine vowels: /a ɒ e ə o i u ɿ ʮ/. /a e o i u/ are written with the usual letters. /ɿ/ is written <y>, as in Yi. This leaves /ɒ ə ʮ/, which are written <q> (cf. X-SAMPA <Q>), <z> (cf. Natqgu <z> /ə/), and <v>. Then there are tone letters, -p -x -t, exactly as in Yi. Mid tone is unwritten. Prenasalization is marked by doubling the consonant.

In zzx, <z> has to be a vowel, so the syllable is z-z-x, onset-vowel-tone. In zzyx, on the other hand, there's a vowel letter, so the syllable has to be zz-y-x.

There's a "normal" orthography, which uses a four-vowel analysis: /a ə ɿ ʮ/. So [ɒ e o i u] are analyzed as /wa yə wə yɿ wʮ/. Possibly there will be diachronic change in the dialects, as opposed to the classical language, that makes this more reasonable.

Anyway, here's the Seven Kill Stele in Hlu, V'eng, and Zzyxwqnp.

Gejaehl Hlu:

Hnuex thu kueny xihmun hmahliphuthukla.
hnuex thu kueny xi=hmun hma-hli-phu-thuk-la
heaven thing myriad man=DAT BEN.3-HABIT-create-DAT-3SG
/nɯç tʰu kɯɲ çim̥un m̥ɔɬipʰutʰuklɔ/

Yahlyah i xi ddoe hnuexhmun ji choenila.
yahlyah i xi ddoe hnuex=hmun ji choe-ni-la
nothing POSS man for heaven=DAT recompense NEG-COP.HABIT-3S
/jɔɬjɔh i çi ɗɤ n̥ɯçm̥un ⁿɟi cʰɤnilɔ/

Thueysago. Thueysago. Thueysago. Thueysago. Thueysago. Thueysago. Thueysago.
thue-y-sa-go
kill-PL-JUSS-1PL.INCL
/tʰɯjsɔⁿgo/

Bor Hlu:

Khoe hnuex thau kueny ddoe xi saq phula.
khoe hnuex thau kueny ddoe xi saq phu-la
DEF heaven thing myriad for mankind nourish create-3SG
/kʰɤ n̥ɯç tʰau̯ kɯɲ ɗɤ çi saʔ pʰula/

Roahlyoah xi ddoe hnuex ji dduela.
roahlyoah xi ddoe hnuex ji ddue-la
nothing man for heaven recompense be_at-3SG
/rɔɬjɔh çi ɗɤ n̥ɯç ⁿɟi ɗɯla/

Dueysago. Dueysago. Dueysago. Dueysago. Dueysago. Dueysago. Dueysago.
n\thue-i-sa-go
2\kill-PL-JUSS-1INCL
/ⁿdɯjsago/

There are a few things going on here.

Gejaehl, unlike Bor, doesn't use definite articles for proper nouns. The verb complex is a little difficult: I'm not sure how to gloss thuk, but it marks the presence of a dative, which in this case is also a benefactive. Part of the Hathic influence in Hluic is the complete loss and later redevelopment of all ditransitive structures, since ditransitives are foreign to Hathic; the extensive use of ditransitives is one of the syntactic differences between Gejaehl and Bor.

It would be possible to follow the text more literally, like the Bor translation below, and replace xihmun hmahliphuthukla with ddoe xi hmaxak hmahliphula in_order_to man BEN.3-feed BEN.3-HABIT-create-3SG, but it wouldn't be idiomatic, and infinitives do have to be marked for the benefactive. (Gejaehl xak isn't related to Bor saq -- they're from *hazaka and *caqo respectively.)

The structures of alienable possession are also different. Both Gejaehl and Bor have suffixes of inalienable possession, derived from the (old) pronouns (-laeq, -yu, -la) but alienable possession in Bor works on the Hathic model, using a simple locative verb -- "nothing is at man (with which to recompense heaven)" -- whereas Gejaehl preserves the Macro-Vengic model of a possessive preposition and a copula -- "nothing belonging to man exists (with which to etc.)".

I've translated "kill" as being in the 1PL.INCL. In Bor, the ancient second-person nasal prefix has been extended to this; in Gejaehl, it has not.

(I wonder how widespread the Rau (Hathic) monotheistic worship of Qapi Mongkoush is, and if it's notably widespread, whether or not it'd spread into Hluic territory.)

Hluic is SOV, but SVO is probably original for Vengic. This is probably because I got confused, but I'm not about to invalidate the entire corpus, nor do I understand diachronic syntax whatsoever, so the real reason is Amqolic influence gradually shifting it toward SOV typology.

V'eng:

Kömthag zë rʼang za chaʼ e theng.
Kömthag ze~i rʼang za chaʼ e theng
Kömthag create\MAJ myriad for nourish of mankind
/kʊmtǎɣ dʒɪ ʔɾaŋ dʒa cǎʔ ɛ tɛ̌ŋ/

Mya s dri e theng zoʼ za mye s Kömthag.
mya s=dri e theng zoʼ za mye s Kömthag
none of=thing POSS mankind COP for recompense of Kömthag
/mja zdrjɛ tɛ̌ŋ dʒɔʔ dʒa mjɛ skʊmtǎɣ/

Ca tʼug tʼug tʼug tʼug tʼug tʼug tʼug.
ca <ʼ>dug
JUSS 1-kill
/tʃa tʼuɣ/

V'eng, unlike Hlu, wouldn't use "heaven" in the way the unusual dialect of English used for translations from Chinese would; presumably in Hlu it's an Amqolic import, since Hathics wouldn't either. But there are, in its pantheon, some deifications of vague daoöid primordial forces, such as Kömthag, who represents the natural world, and Kömdëg, who represents the human one. Natural law theorists in the Hathic or Hluic cultural spheres would speak of it as emanating from their cognate of Qapi Mongkoush or Hnuex, although on Hathenai proper it'd be a little more complicated -- on Hathenai there's nothing that wouldn't be -- but in the Vengic sphere, it'd be split into the domains of Kömthag and Kömdëg. Here we use Kömthag.

The major/minor verb contrast inherited from Proto-Macro-Vengic, which was originally a sort of reduplication, is preserved unusually well in Vengic, and has specialized into a semantic use that is remarkably hard for even native grammarians to define. Gods always take major verbs.

The difference between s and e is a little difficult -- s is more tightly semantically binding, but the two are mostly interchangeable, and there are phonological factors also: e is more likely to be used after consonants and high vowels (which then become glides), and s is more likely after non-high vowels. There's also a fused form es, which can be used to bind nouns to adjectives. This variance is mostly lost in Zzyxwqnp.

Zzyxwqnp:

Kuntqx jje chip yutrqn kqtye chaxmq yep yattznt.
Kuntqx jje chip yu-trqn kqtye chax-mq yep yattznt
Kömthag IPFV create myriad for nourish-NMLZ of mankind

Jjotryp yep yattznt tq kqtye pexjje Kuntqx.
jjotryp yep yattznt tq kqtye pexjje Kuntqx
nothing of mankind COP.IPFV for recompense Kömthag

Chatvx. Chatvx. Chatvx. Chatvx. Chatvx. Chatvx. Chatvx.
chatvx
JUSS.1.kill

The major/minor contrast has been lexicalized. Jussives are a little irregular, since they preserve the old person marking.

If you prefer the more reasonable orthography:

Kwuntwâ nye tyì ywuchwan kwáye tyâma yè yátén.
Nywechì yè yátén twa kwáye pyênye Kwuntwâ.
Tyatû. Tyatû. Tyatû. Tyatû. Tyatû. Tyatû. Tyatû.

[gestaltist]

Thanks for the explanation. It's incredible how different the Zzyxwqnp text looks in the two romanizations.

[Hallow XIII]

Will you look at that, it's almost the end of January. Time for the Qishabei in Eahswa!

⁶ʔea¹saǎ ¹dǎ⁷te³göǎi ⁸pǎ ⁷te⁷nüiʔ ¹ei ¹dao ²ngǎw ¹rǎ⁵buh ²paǎ

⁶ʔea-

cIV-

¹saǎ

sky

¹dǎ-

REAL-

⁷t<h>e-

PL/cVII-

³göǎi

bear

⁸pǎ

thing

⁷t<h>e-

PL/cVII-

⁷nüiʔ

count

¹ei

arrive

¹dao

NEG

²ngǎw

so_that

¹r<s>ǎ-

PL/cI-

⁵buh

raise

²paǎ

human

¹rǎ²paǎ ⁸pǎ ⁴khoǎ ⁷the⁷meah ⁷ngǎiʔ ¹dao ¹ei ⁷the⁴phǎ ⁸kuh ⁶ʔea¹saǎ

¹r<s>ǎ-

PL/cI-

²paǎ

human

⁸pǎ

thing

⁴khoǎ

good

⁷the-

cVII-

⁷meah

one

⁷ngǎiʔ

exist

¹dao

NEG

¹ei

arrive

⁷the-

cVII-

⁴phǎ

give

⁸kuh

return_to

⁶ʔea-

cIV-

¹saǎ

sky

¹ʔaǎh⁸hǎ, ¹ʔaǎh⁸hǎ, ¹ʔaǎh⁸hǎ, ¹ʔaǎh⁸hǎ, ¹ʔaǎh⁸hǎ, ¹ʔaǎh⁸hǎ, ¹ʔaǎh⁸hǎ

¹ʔaǎh

kill

-⁸hǎ

-ANTIP

The first thing to notice here is that there is actually a significant amount of verbal morphology. Eahswa actually has polypersonal agreement (although the affix for a third-person agent is universally zero) inherited in large part from PKP; furthermore, some historical serial verbs have become sufficiently tightly bound to the verb to become syntactically obligatory affixes. This happened to the antipassive -⁸hǎ, etymologically "go".

Second, unlike in Kangshi, noun class markers aren't tightly bound to the noun root. Indeed, the general rule is that a classifier occurs either directly modifying the noun itself or as an ~anaphoric~ (verbal object marker or modifying a dependent phrase), but not both. Similarly, the Kangshi double-classifier construction for inalienable possession does not exist; instead, all non-predicative possession is done either by simple apposition or with a relative clause construction using the verb ¹bea "be with. be together".

A further difference in the noun class system is that while in Kangshi the plural is a distinct noun class of its own, in Eahswa the plural is instead marked on the class marker via consonant gradation. For instance, the class one marker ¹sǎ- rhotacizes to ¹rǎ-, and the class four marker ⁶ʔea- becomes ⁶hea-. Synchronically, these alternations are highly idiosyncratic, but are quite regular historically, deriving from a plural prefix *m- which underwent cluster resolution.

Going from morphology to syntax, Eahswa has a few features distinct from Kangshi. It is a topic-prominent language, which the second sentence provides a good illustration of. Bracketing the big syntactic chunks, we get:
[¹rǎ²paǎ]_TOPIC [⁸pǎ ⁴khoǎ ⁷the⁷meah]_SBJ [⁷ngǎiʔ ¹dao]_MAIN [¹ei ⁷the⁴phǎ ⁸kuh ⁶ʔea¹saǎ]_REL

The topic is fronted and does not interact with the comment clause. Predicative possession is expressed on the East Asian pattern -- "mankind, there is not one good thing". Relative clauses are extraposed when they modify the subject, and are not especially marked for modification, unlike in Kangshi, which uses a nominalizing prefix.

Also unlike Kangshi, Eahswa does not have prepositions. Kangshi has three, derived from verbs; some Kangshuic A languages lose the two less common ones and simply have ni as a generic oblique preposition. Eahswa, and Kangshuic C more generally, keeps the verbal constructions or simply uses juxtaposition.

[Nortaneous]

Well, Deghuri is going to need work.

More information on Hlu should be brought over from the old board.

Sound changes:

More: show

ɲ > r
c > ts
z > y
t k > ð ɣ / V_V
V: > V
k x ŋ > c ç ɲ / _i
ky xy ŋy > c ç ɲ
a$i a$u > ɤ$V (this change is shared with Vynyi)
i$a u$a > ɤ$V (this change is shared with Vynyi; note that {e o}${i u} don't also become schwa)

Stress shifts to the penultimate syllable and multisyllabic roots lose their unstressed vowels.

nP mβ nð ŋɣ > ⁿP[+voice] ⁿb ⁿd ⁿg
f > h
hP > P:
hC > C
ĩ ũ ẽ õ > æ̃ ɔ̃ ã ã
e > æ / _{c ç ɲ}

s ts > θ s

Proto-Hluic > Hlu:

V~ > V
?F > ?P[+implosive]
C[+alveolar]i > Cɯ
ax ɤx > ɤx ɯx
x > h > 0 / _#
t n > c ɲ / _i
æ e > e i
ik ix iŋ ic iç iɲ > ɯc ɯç ɯɲ ɯc ɯç ɯɲ
ek ex eŋ ec eç eɲ > ɤc ɤç ɤɲ ɤc ɤç ɤɲ
ɣV > Vˠ / C_
o u > we wi EXCEPT Vˠ VC[+velar]
w > 0 / C[+labial]_
aˠ eˠ ɤˠ oˠ iˠ ɯˠ uˠ > ɯa æ ɯɤ o ai ɯ u
P: > ⁿP[+voice]
C[-voice] > C[+implosive] / _V
l r > 0 / C_V
C[labial]CV > CV[+round]
ɯa ɯɤ > ɯ

Some grammar notes:

More: show

There are egophoric markers tu (< *tβɯˠ < *otaɣẽ; cf. V'eng tag) and çwi, perfective and imperfective respectively. These are descended from mildly irregular first-person-marked forms of the aspectual copulas: one of the oldest isoglosses in Hluic is the shift of pretonic o- to -β-, and Hlu itself is spoken just to the west of the dividing line. These particles take the second position in the phrase, but may fuse with the verb when appearing directly before it.

For example:

gwiʔ tw-ɤ-ɲɤm. yu tu-nɤ-ɲɤm ɤŋ
1S EGO-PST-drink. 2S EGO-2\PST-drink or
I drank. Did you drink?

gwiʔ tu βuya ɤ-ɲɤm. yu tu βuya nɤ-ɲɤm ɤŋ
1S EGO beer PST-drink. 2S EGO beer 2\PST-drink or

yæ çwi ɠeʄæɬ ɗɯ-la
house EGO.IPFV Gejaehl be_in-3
My house is on Gejaehl.

yæ ɠeʄæɬ ʄa (ɗi-a)
house Gejaehl be_in\PL-3
There are houses on Gejaehl.

(Etymological notes:
* The first-person singular pronoun gwiʔ is descended from a phrase cognate to V'eng (nk)ur e aʔ, roughly "me here", a circumlocution innovated to disambiguate from the third-person pronoun a. This is one of two common ways Hluic resolves this ambiguity; the other (and likely older, since it occurs both in the eastern regions and in Gyitha) is a descendant of Proto-Yia *aʔo for the first-person pronoun and *mɣay, roughly from "this one", for the third-person.
* 'Drink' < PY *(n)inima.
* 'Beer' is probably a late Kharidze loanword from, *buya.)

(Final -t becomes a glottal stop after fronting preceding vowels, after which -s > -t. This makes inalienably possessed nouns somewhat irregular because syllabification: ɓweʔ-yu 'your penis' but ɓoɗæʔ 'my penis'. This doesn't happen with verbs, however.)

Hlu preserves the second-person prefix n-, but has lost and reinnovated all other person-number marking by grafting pronouns onto the verb. (This is drawn from Tangut.)

The first-person marker -æʔ can't coexist with the egophoric particles, so it generally only appears in questions: (note the absence of the past tense marker; this is because of the adverb. tense is ~relative)

ⁿgwiʔ niⁿjwiɠa ɲɤm-æʔ ɤŋ
1S yesterday(ADV) drink-1S
Did I drink yesterday?

Or involuntary actions, which don't take egophoricity:

ⁿgwi yi ɓoɗæʔ ⁿdo hɤk-æʔ
1S because my_penis enormous trip-1S
I just tripped over my enormous penis.

Cf. (note the fusion of tu and yi -- these fusional forms exist for many prepositions)

ⁿgwi twi ɓweʔyu ⁿdo çɤk
1S EGO-because your_penis enormous FUT\trip
I will (intentionally) trip over your enormous penis.

Or, in the second person:
yu yi ɓoɗæʔ ⁿdo n-ɤk-u
2S because my_penis enormous 2-trip-2S
You just tripped over my enormous penis.

(Etymological notes:
* 'yesterday' < *eteNu e zaka 'past day'. ue > uy > wi
* -(ŋ)i = abstract nominalizer; ne-ŋi 'that which is located at'
* fusion of *niɲ ytwiɠa to niɲjwiɠa
* adjectives are intensified by prenasalizing the initial consonant: ɗo 'big' > ⁿdo 'enormous')

The history of the third-person pronouns is complicated. Hluic preserves the old semidistal demonstrative *ʔati > *ʔði, which compounded with *tengaCV and *hamatsi > *tŋa, *hmas to produce the gendered animate pronouns...

ʔði e tŋa > lyekŋa > yɤʔŋa
ʔði e hmas > lyehmas > yehmaɬ

ð > l is regular, ly > y is regular, place assimilation in PN clusters is regular. PN clusters usually become plain nasals in Hlu, but in some other Hluic languages they become geminate plosives, e.g. pre-Gyitha ʔðyekka. In some dialects these are compressed further, to yŋa yhmaɬ.

The third-person marker is -a for verbs that end in a consonant. For verbs that end in a vowel, -i -u -e -o become -ya -wa and low vowels take epenthetic -l-. As usual, this is reused for inalienable possession. Verbs that end in -ɯ and -ɤ (which are always singular) used to take zero-marked third person, but -la has spread by analogy.

Immediate future and far future are distinct. Far future is <y>, immediate future is... a verb that got grafted onto the complex, let's go with li. This means it also takes the person marking -- and remember, ly > y, so for third-person subjects it's ya-.

ⁿgwi tu yɤʔŋa li-ɓæ
1S EGO 3S.M IMF-fuck
I am going to fuck him.

yɤʔŋa kwɤ-la ɗɯ-la ya-ɓæ-la ɤŋ
3S.M ass-3 in-3 IMF.3-fuck-3 or
Are you going to fuck him/her in the ass?

Alternatively, the adpositional phrase can be incorporated, in which case it doesn't take as much person marking; but since the object isn't specified by the possession marking, it has to be made overt:

yɤʔŋa yehmaɬ ɗɯ-kwɤ-li-ɓæ-la ɤŋ
3S.M 3S.F in-ass-IMF-fuck-3 or
Are you going to fuck her in the ass?

The sentence yɤʔŋa ɗɯkwɤɓæla ɤŋ would be interpreted as having a dropped subject: "Are you going to fuck him in the ass?"

(Note that the positions of the noun and the adposition are reversed. This is due to the particulars of serial verb chains and word order in Hathic, although it would be slightly more complicated there. Hluic has Hathic influence, but it doesn't have the mess of focus particles that Hathic does, and doesn't use these chains as extensively. In Hlu, unlike in Gyitha, agglutinative compounding derived from serial verb chains being reinterpreted as one unit isn't obligatory, so either yɤʔŋa kwɤla ɗɯla yaɓæla ɤŋ or yɤʔŋa yehmaɬ ɗɯkwɤɓæla ɤŋ are acceptable; Gyitha doesn't allow the cognate of the first construction.)

In some cases, Hlu preserves plural marking, descended from Proto-Yia -i.

Verbs ending in -k -w -ŋ mutate to end in -c -y -ɲ, with various effects on the vowels: (note the secundative syntax)

ⁿgwiʔ tu tʰuk
1SG EGO run
I am running.

ŋæʔ tu tʰuc
1PL.EXCL EGO run\PL
We are running.

ⁿgwiʔ tu kɯw
1SG EGO swim
I am swimming.

ŋæʔ tu kwi
1PL.EXCL EGO swim\PL
We are swimming.

In some cases, this also causes ablaut:

ⁿgwiʔ çwi ɬɤŋ
1SG EGO.IPFV ferment_a_beverage
I ferment beverages.

ŋæʔ çwi ɬiɲ
1PL.EXCL EGO.IPFV ferment_a_beverage
We ferment beverages.

-ɤ- can become either -i- or -u-. -a- becomes -ɤ-. Most velarized vowel irregularities have been analogized out, but here's one that hasn't:

tʰowæʔ ya çwi yiw ɗæ
hand-1.POSS INSTR EGO.IPFV brick break
I break bricks with my hands.

tʰowæʔ ya çwi yiw ɗi
hand-1.POSS INSTR EGO.IPFV brick break
We break bricks with our hands.

(Etymological notes:
* The first-person exclusive plural pronoun ŋæʔ is from *ŋaako. The first-person inclusive plural pronoun ⁿgo is from *ŋakoNu. This vowel shortening may be regular.)
* *tafuko 'run' > tʰuk
* *keNuβa 'swim' > kruw > kɯw (dissimilation); uy > wi
* *lohiŋa 'ferment a beverage' > lhɤŋ > ɬɤŋ
* *taɣeNa 'destroy, break, split, chop' (NB: metathesis in Vengic to *taNaɣa) > tɣer > ɗæ)
* *dVroβu 'hand' > drow > tʰow; ow > ɤw so this is another irregular possessed noun (cf. tʰɤwyu 'your hand(s)')
* *yibu-ŋi 'bread' > ypʰuɲi; *hazaka 'feed' > çak. I didn't need either of these words for the examples.)

Tower of Babel: (in Gejaehl Hlu, except /r/ and /aj/ are maintained - these would be /j-/~/-0/ and /æ/ in proper Gejaehl)

More: show

niɲɤhli yihl hɤɲ ⁿgɔ ɤ- ru -lɔ
world whole language one PST-have-3

ʔɔ ⁿdɤɲ kʰɤ ʔwes ɤ-le -ʔɯŋ
then mankind DEF east PST-move-go_to

ʔɔ çinɔ yi ɤ- çɤ -hli -lɔ
then Shinar plain PST-find-settle-3

ʔɔ ⁿdɤɲ ɤ- wɤɲ-yi
then mankind PST-say-REFL

yæ ⁿbɔ -wu -sɔ -h -æ
brick make-bake-JUSS-COLL-1

yæ wɔ hnɔ ɤ- myɔ-lɔ
brick instead_of rock PST-use-3

ⁿgɔs tʰu wɔ ⁿdɤy
and tar instead_of mortar

ʔɔ ⁿdɤɲ ɤ- wɤɲ-yi
then mankind PST-say-REFL

hmæ- yo -ⁿbɔ -sɔ -hæ
BEN.1-city-make-JUSS-COLL.1

ɗo li ʔyɔ yæʔɯŋ li hmæ- ⁿbɔ -sɔ -hæ
tower REL sky reach REL BEN.1-make-JUSS-COLL.1

ⁿdɤ chɯ ɗwi niɲɤhli yihl ɔy- tɔw
in_order_to NEG across world whole PASS-scatter

ʔwe ɠom ⁿdɤ li hmɔⁿgu- ⁿbɔ -ɗɯ -sæ i ⁿdɤɲ ɗo yo sweʔ li ɤ- ⁿbyɔ-le -lɔ
but God in_order_to REL BEN.3(2)-make-be_in-NONF POSS mankind tower city see REL PST-down-move-3

ɤ- wɤɲ-lɔ kʰili lihlɔ ⁿdɤɲ ⁿgɔ yoʔ-sæ hɤɲ ⁿgɔ ru -sæ ⁿbɔ -ɗɯ -hlæ-kʰi
PST-say-3 if they mankind one COP-NONF language one have-NONF make-be_in-can-COND

li rɔhl'yɔh i lihlɔ çwi mi- rwæ -hlæ-kʰi
if...then nothing POSS them EGO.IPFV FUT-fail-can-COND

hmæ- ⁿbyɔ-ⁿbo -le -sɔ -yi tu
BEN.1-down-CAUS-move-JUSS-REFL EGO.PFV

hɤɲ i lihlɔ ⁿdɤ li lihlɔ chɯ-wi- -yi -sæ li hmæ- ⁿbo -hwin -sɔ
language POSS mankind in_order_to REL them NEG-understand-REFL-NONF REL BEN.1-CAUS-split-JUSS

ʔɔ ɠom lihlɔ ʔihlirɔrɔ ɤ- ⁿbo -ʔwirɔrɔ-lɔ
then lord them outward_in_all_directions PST-CAUS-scatter-3

ʔɔ lihlɔ yo ɯy hmɔ -ɤ -mɤ -lɔ

[Nortaneous]

It occurs to me that the Amqolic topic/alignment system should be made any sense of. So, how about this.

More: show

- Topics are explicitly marked. Changes in topic are marked with Amqoli -eg; otherwise there is (usually) overt topic marking on the verb.
- There are three noun classes, arbitrarily called T, S, and L. These should become more relevant later, maybe with adjectival agreement or something. Maybe the topic marker will also change based on noun class. If there's no topic marked with -eg, the verb should take topic agreement.
- Topic agreement suppresses ergative and absolutive agreement. I forget why these are ergative and absolutive and they probably should not be but oh well. So if the topic is the agent of a transitive verb, you don't get ergative agreement; and if the topic is a patient of a transitive verb or the agent of an intransitive verb, you don't get absolutive agreement.
- If the topic is neither the agent nor the patient, it's marked with the postpositional case or something. Except this is usually -e, so it is realized as zero when followed by -eg. Oops. Maybe there will be a stress shift.
- Also, ergative markers pull stress one syllable toward the end of the word because why not. Possibly they were all originally long vowels or something. But Amqoli stress is unstable, so we can postulate a lot of analogy somehow or other.

This is all nonsensical bullshit but morphosyntax is fake anyway.

edit: Actually, Tangut has something like the pre-reform Amqoli system, so it might be better to just do that. There's one concord slot, but whether or not it's for the subject or the object depends on the verb.

[Nortaneous]

Here's some historical phonology for the Kett languages, which I may or may not stick with. (This is, of course, a complete departure from the earlier stuff, which was boring.)

The conditions for palatalization and labialization should be more complex, but they aren't yet, oh well. Also they should be more about cluster collapse and less about vowels.

edit: This won't work - it doesn't fit with the Insular forms at all. There should probably be more consonants in the proto-language.

More: show

So, there's a proto-language:

/p t ts k/
/s ɕ h/
/m n ŋ/
/l r j w/
/a e i o u/

First, there's lenition in clusters:
i u > 0 / P_P unless stressed
clusters where C2 is a coronal > a geminate coronal (except for Cl Cr)
p k > w / _P _F _N
t > j / _P _F _N

Then, a rounding contrast develops in the stops and nasals:
m p > ŋʷ kʷ / {o u}_ _{o u}
m p > mʲ pʲ / _{e i}
ŋw wŋ > ŋʷ
wk kw > kʷ
jm mj > mʲ
mʲ > nʷ
jp pj > pʲ
pʲ > tʷ
s > sʷ > tʷ (or s > tθ > tf > tʷ, or whatever)
[These clusters also apply through a fricative, except /s/ palatalizes when immediately adjacent to /j/ - so wsm > tʷŋʷ, jsm > ɕnʷ, jps > tʷtʷ, etc.)

The rounding contrast is reinforced by the loss of /o/:
l r h > lʷ rʷ hʷ / o_ _o ! _j
o > we / #_ ! _Cʷ
o > e

This leaves an inventory of:
/p t tʷ k kʷ/
/ɕ h hʷ/
/m n nʷ ŋ ŋʷ/
/l lʷ r rʷ/
/j w/
/a e i u/

At this stage, /i u/ eject a following homorganic semivowel when in open syllables:
i u > ij uw / _CV _#

Then, in Insular Kett:
- geminate stops fricate (cf. Iaai) [in word-initial position]
pp tt ttʷ kk kkʷ > f ts2 xʷ x xʷ
- the rounding system is lost
tʷ kʷ > tp kp
i e > y ø / _Cʷ
y > iɥ / _CV _#
w > ɥ / _i _j
Cʷ > C / #_
Cʷ > wC / a_ e_
Cʷ > C
ø > we / {V #}C_ ! C[+labial]_
ø > e
s ts ts2 > x s ts
- the mid/high contrast is mostly lost, under Arve influence
ɕ > s / {# C}_
ɕ > js
ej eɥ ew > ij iɥ iw
i > 0 / C_RC (producing syllabic resonants)
u > iw / when stressed
- the doubly articulated stops are lost
tp kp > tʙ̥ ɓ

/p ɓ t ts tʙ̥ k/
/f s x h/
/m n ŋ/
/l r j ɥ w/
/a e (i y u)/

In Continental Kett, on the other hand:
e > a / _h
h > 0 / _C
a > a: / _CV _#
ts > 0 / _#; otherwise ts > s
æ > aj / _CV _#
æ > a
hʷ > fʷ
n l r > nʲ lʲ rʲ / _i _j
j w > 0 / C_
e i u > ə ɨ ɨ
e > a / _CC#
V[-long] > 0 / VC_#; a > ə before a dropped high vowel; ɨ > ə before a dropped low vowel
j w > jɨj wɨw / #_
a: > aɰ
h > ɰ
semivowel 'flattening' as in Marshallese
C[+labial] > C[+labial]ʷ
C with no secondary articulation > Cˠ (i.e. conditions central vowels)

/pʷ tʲ tˠ tʷ kˠ kʷ/
/fʷ sʲ/
/mʷ nʲ nˠ nʷ ŋˠ ŋʷ/
/lʲ lˠ lʷ rʲ rˠ rʷ/
/j ɰ w/
/a ə ɨ/

This is based on Marshallese, but the end results are completely different - labials always condition back rounded vowels, there's a full series of rounded coronals, and there are fricatives, although they could just as well be lost along with /h/.

[Whimemsz]

Nice.

e i u > ə ɨ ɨ
e > a / _CC#

Presumably these are supposed to be in the opposite order though?

[Nortaneous]

OK, the topic thing is stupid. Let's do it this way.

Amqoli, like Tangut, has only one verbal concord slot:

Code: Select all

   1     2/3
MT d-    t-
FT d-    u-/b-
IT       che-
MS       m-
FS       mu-
IS       me-
ML       tu-
FL       chu-
IL       chu-

There are then two classes of verbs, subject-concord and object-concord. Modal verbs, stative verbs, and verbs of motion are in the subject-concord class; derived causatives and ditransitive verbs are in the object-concord class. So:

zha phrus dqare
1SG.M jerky 1MT-eat
I eat jerky

tse zha phruse deba (or maybe rba?)
2SG.M 1SG.M jerky-POS 1MT-give
you give me jerky

It's possible to derive a subject-concord verb from an object-concord verb. Some verbs have stem alternations:

chalyar zha detsor
1SG.M horse 1T.kill
the horse killed me

chalyar zha detsru
1SG.M horse 1T.kill\SBJ
I killed the horse

obzhdami es mezh cheborgu
swamp_dweller-PL 3SG.M FUT 2/3MT-destroy
the swamp-dwellers will destroy him

obzhdami es mezh cheborgub
swamp_dweller-PL 3SG.M FUT 2/3MT-destroy\SBJ
he will destroy the swamp-dwellers

In other cases, the affix -z is used.

bgul ndrya mchxe
man dog 2/3ML-bite
the dog bites the man

bgul ndrya techxez
man dog 2/3MT-bite-SBJ
the man bites the dog

For class B verbs, there's an optional subject marker dze:

bgul dze ndrya techxez
man SUBJ dog 2/3MT-bite-SBJ

Then, for subject-concord verbs, there's a division between transitive and intransitive. Transitive verbs don't need object marking (but take a different optional subject marker, aq); intransitive verbs that still have an object require the object to be in either the instrumental or the postpositional, depending on semantics.

So, for a transitive subject-concord verb:

Rabi Muk (aq) enim damgu thagul
Robby Mook (SUBJ) his_own penis 2/3MT-search
Robby Mook is searching for his own penis

Rabi Muk (aq) thagul
Robby Mook (SUBJ) 2/3MT-search
Robby Mook is searching for something

And for an intransitive subject-concord verb:

Hileri Klintam Rabi Mukime damgeb bqare
Hillary Clinton Robby Mook-GEN-POS penis-POS 2/3FT-eat
Hillary Clinton ate Robby Mook's penis

Also note that there is Suffixaufnahme - why not.

[Nortaneous]

If you've been paying attention, you'll notice that there is such a thing as the Vtsznxmqye Pvpchqpye Zzxzzyx. If you haven't:

More: show

Here's the prelude to the monologue of Shquqou in a moderately archaic poetic Hlu retelling of the Vtsznxmqye Pvpchqpye Zzxzzyx (or, if you prefer, the Úsênman-ye Pùtywà-ye Nênî), an epic relating the Hathic invasion of Narng, their defeat at Zot, and the founding of the Empire of Zzxzzyx.

Poetic Hlu admits of a lot of flexibility in word order and a lot more compactness and ambiguity than you'd get elsewhere. Standard Hlu is verb-final and suffixing, so the syntax is customarily tortured to admit decent rhymes - and since there's a great deal of cultural influence from Zzxzzyx, the empire that contains their territory, Hlu poetry (and high-register prose) can just take its word order from Zzyxwqnp. The Vtsznxmq was originally written in Zzyxwqnp, of course. Definite articles are uncommon - since the standard meter for war poetry is the pretty compact Kalevala meter, so there isn't much room for them, and they're a recent innovation that's absent in Zzyxwqnp.

Ykoeku yaehlae choe thuem jjangnihl | yoeqnga hliddaesae i thwiyihl
[kjɤku jæɬæ cʰɤ tʰɯm ʄaŋniɬ | jɤʔŋɔ ɬiɗæsæ i tʰɥijiɬ]
Ykoeku yaehlae y-khoe thuem jjangnihl | yoeqnga hli-ddae-sae i=thwiyihl
Shquqou demon of=DEF ocean north | 3SG.M all HABIT-destroy-NONF of=everything

Ykoeku thuesaeqyi i ddoeny kueny | yKatnahl myahltuen yemak yhloeythueny
[kjɤku tʰɯsæʔji i ɗɤɲ kɯɲ | kjɔtnɔɬ mjɔɬtɯn jemɔk ɬjɤjtʰɯɲ]
Ykoeku thue-saeqyi i=ddoeny kueny | ykatnahl myahltuen yemak y=hloeythueny
Shquqou kill-NMLZ.A of=man myriad | of=Katnahl evil most of=red-spear

Koentue gahle wong hlituesae | yo gas yDue hnami oebansae
[kɤntɯ ⁿgɔɬe wɔŋ ɬitɯsæ | jo ⁿgɔs ⁿdjɯ n̥ɔmi ɤⁿbɔnsæ]
Koentue gahle wong hli-tue-sae | yo gas yDue hnami oe-ban-sae
Komthag great against HABIT-offend-NONF | city and of=Narng rubble PST-COP.CAUS-NONF

ggwak gas ddoeny gas bodicila | caeddue thueny aybohli lihla
[ɠwɔk ⁿgɔs ɗɤɲ ⁿgɔs ⁿboⁿdicilɔ | cæɗɯ tʰɯɲ ɔjⁿboɬi liɬɔ]
ggwak gas ddoeny gas bo-ddici-la | caeddue thueny aebohli lihla
barbarian and man and CAUS-gather-3 | peasant spear PASS-CAUS-come 3PL

nwaya lihla kokha ykoet hmi | hlyomhla yngwe xi yac qwiyoyi
[nwɔjɔ liɬɔ kokʰɔ kjɤt m̥i | ɬjomɬɔ ŋwje çi jac ʔwijoji]
nway-a lihla kokha ykoet hmi | hlyom-hla y=ngwe xi yac qwiyo-yi
order-3 3PL sword arrow with | fear-3PL.POSS of=death for no_one scatter-REFL

hoeny i Katnahl ggwakhmun woenya | hoeny i Cihlae ddoenyhmun woenya
[hɤɲ i kɔtnɔɬ ɠwɔkm̥un wɤɲɔ | hɤɲ i ciɬæ ɗɤɲm̥un wɤɲɔ]
hoeny i=Katnahl ggwak-hmun woeny-la | hoeny i=Cihlae ddoeny-hmun
language of=Katnahl barbarian-DAT speak-3 | language of=Cihlae man-DAT speak-3

So - Shquqou, scourge of the northern seas, destroyer of all, killer of ten thousand men, vilest among the berserkers, habitual offender against the divine order, leveler of the cities of Narng, and so on, gathers an audience of barbarians and men, of peasants and warriors, threatens them with swords and arrows so none of them run for fear of death, and speaks to the barbarians in the language of Katnahl and to the men in the language of Cihlae, and... delivers a long sermon on obscure points of theology.

It is worth mentioning that Hlu is not known for the quality of its poetry.

There should probably be something about the history of Zzxzzyx. Which requires elaborating on the trade routes of the Swamp Continent.

So! There is Tsalaysia, which contains the Tsi, a Canoe Viking empire. Occasionally it also contains the Kztddajjq - or, if you prefer, Kénanywa - the remnants of a much earlier Canoe Austronesian expansion which spread pretty much everywhere. Tsalaysia lies between the Swamp Continent, which is an exception to that "pretty much everywhere", and the Dog Continent, which is not.

On the other side of the Yichvnptzp mountains - let's actually go with the Amqoli form here and call them the Zhjumna, although that isn't much better - lies Enze, which is a little like the Holy Roman Empire on horseback and in the Midwest. Trade between Tsalaysia and Enze usually lands at the Big Yeet, which is, of course, Narng for "great port". (/piəħ jeet/ < *piki ɣitaxe > Hlu bbuey (dae-)lue, if you were wondering)

The reason the Kénanywa are called that, you see, is that they're the people of the qutna, loaned into Zzyxwqnp as kztdda (if you're wondering about Katnahl up there, it's the same basic formation) - the sacred plant of Qapi Mongkoush. One day, as the Vtsznxmq has it, a Kztddajjq chief by the name of Beshiyash happened to be in the Big Yeet, and noticed that qutna had caught on as a popular plant in the merchants' gardens, and figured that this was a sign that it was time to introduce those heathens to the true religion and also take all their money and capture their trade routes.

So he raised an army, and then the army's religious fervor got the better of their materialism, another priest by the name of Shquqou killed Beshiyash, and they figured they ought to introduce all of heathendom to the true religion and most to Qapi Mongkoush, personally. As it happened, points west didn't think indiscriminate slaughter was a good sales pitch for a new religion, so they formed a confederation which proceeded to repel the Kénanywa invasion and capture the Big Yeet and the Tsalaysia-Enze silk road, and then realized they were responsible for a lot of trade routes and sprouted a regulatory bureaucracy.

That last part, of course, isn't in the Vtsznxmq.

[Nortaneous]

Man I keep forgetting what's in contact with what.

There is the Great Swamp, which is sort of like the US East Coast, but not as big. The Great Swamp is (probably) roughly triangular, pointing west into the ocean, but with a few peninsulae here and there. From east to west along the coast, there are Vengic, Zotic, Hluic, and Narngic, all of which are part of the Ziwanic family. The divergent Vynyi and the Whatic family are spoken somewhere to the south, but are not in contact. Zotic was probably pushed south by Narngic expansion; the isoglosses run Vengic-Hluic-Zotic-Narngic.

The major Vengic languages are Zzyxwqnp, which is Vengic, and Hlu, which is Hluic. There is also Gyitha, spoken on Ketas Island and in contact with the Hathic language Ketas.

The Great Swamp is bounded on the southeast by the Siamalaya Mountains, which contain Amqolic and Qoic; Amqolic, however, contains the divergent Rengni, which fucked off north to the Big Yeet, situated on the coast north of the Siamalayas. Pirka, a para-Qoic language, is spoken somewhere.

North of the Great Swamp, possibly continuing from the Siamalayas, is Tsalaysia, which contains one empire, that of the Tsi Canoe Vikings, and three language families: Tsiic, Hathic, and Kaam-Yerte. Of these, Tsiic contains the language of the empire; Hathic represents the language of a much earlier Austronesoid expansion that ended up pretty much everywhere except the mainland of the Great Swamp; and the Kaam-Yerte speakers are oppressed by the Tsi empire and driven into shitty interior highlands.

The major Tsiic languages are Tsi and Cuhbi, which occupies at least two islands somewhere far from the imperial throne. The major Kaam-Yerte language is Ubghuu, which is spoken in the shitty interior highlands of one of the Cuhbi islands. There are no major Hathic languages yet, but ones worth mentioning are Rau, the liturgical language of the followers of Qapi Mongkoush (possibly the Islam to the Christianity of Qoic Institutional Tengrism, unless there's a Latin and Greek deal?); Ketas, which is spoken on the same island as the Hathic language Gyitha; and Hathe, which is spoken very far away. Tsiic and Hathic have clicks.

East of the Siamalayas there is the Nta river delta, which contains more Qoic languages. Past the Nta delta, there is a Vast Expanse of Nothingness, which eventually the Enze will conquer; and past that, there is Enze, another Civilized Country which speaks Kannow and will eventually do a Russia and conquer most of the nothingness. Off the coast of Enze is the island of Harue, on which Insular Kett and Arve are spoken; there are various other languages around here as well, all of which are (so far) isolates.

North of Tsalaysia, near the equator, is the Dog Continent, a mountainous Tsi Empire frontier which contains Kllk.

So, as for contact, at least the parts that will be relevant if I get around to making things that aren't phonology happen:
- Hathic and Tsiic probably form an Altaioid sprachbund.
- So do Amqolic and Qoic; however, Amqolic isn't as big on tones as Qoic is.
- There are some Narngic-Amqolic and Hluic-Hathic interactions, but Ziwanic probably came from Swamp Cornwall so none of these are ancient.

[Nortaneous]

Hathic hasn't been developed much yet because I am terrible at syntax but it is probably worth mentioning to describe at least the phoneme inventory of Rau. So!

/p b t d c ɟ k g q ʁ ʔ/ <p b t d ch j k g q r h>
/f s ɬ ɕ x ꭓ/ <f s z l sh x xh>
/m n ŋ/ <m n ng>
/z j w/ <z j w>
/a e ə o i ɯ/ <a e u o i ou>

Possibly one of /l r/ exist in loans or something. It is also possible that there are tones.