Vjatjackwa, or whatever we're calling it this week

This language is under construction. The first burst of invention on the language took place in November of 2003, but progress halted after the sixth email to the Conlang list on January 9th 2004. On January 13th of 2006, the gauntlet was taken back up after I saw the beauty of Jan van Steenbergen's Vozgian and was reminded of this project, which I'd always meant to have a Slavic-sounding phonology.

To quote from the December 12th 2003 email to the Conlang list,

The whole thing started because of a bug I got in the brain months ago
when reading on the list about where Slavic languages' phonotactics come
from (so it's come full circle!).  That conversation got me started thinking
about consonant clusters formed by vowel deletion, and I thought of a
word "Apaterakasuto -> Apatrakasto -> Aptraksto", which stuck in my brain.

The actual end result, as you will see, sounds quite different from this initial inspiration.

Mli Vjacgu, circa 2006:


The phonetic inventory of Mli Vjacgu is as follows:

WrittenCXS*English approximation
r4flapped r
LL\dark l
yyFrench eu
ee, Eay, eh
oo, Ooh, o
aa, Aa, ah

*CXS stands for Conlang X-Sampa.


Noun forms

The definite form evolved from an infixed _ek_, originally prefixed _ke_, which used to be a word meaning "this". The indefinite form evolved from an infixed _at_, originally a prefixed _ta_, which once meant "some". It was not necessary to mark definiteness when an object was possessed; having another noun in the genitive associated with it made the possessed noun definite by default. When the old genitive suffix _ha_ withered away and died, the possessed object was left in an unprefixed state, which became the construct form.

As sound changes made the definite and indefinite forms less regular, the resulting confusion of forms came to be analyzed as a number of noun classes, membership in which governed a noun's indefinite/definite/ construct declension.


IndefiniteDefiniteConstruct (possessed)

In the example above, "language" would be in the vat/vek/v noun class; "speaker" and "word" in the voct/vixt/viD noun class; and "giver" in the ped/pig/bo noun class.

Demonstrative pronouns

The demonstrative pronouns evolved from a very regular, if convoluted, system of prefixes and infixes where the wh-form began with _hi_, the "this" form contained the infix _ek_, "that" contained the infix _uw_, the "some" form contained the infix _ta_, etc. The oldest demonstrative pronouns were all accented on the first syllable, but as the system was expanded other accentual patterns were included, further complicating the matter.

questionrelativethis, herethat, thereyon, yondersome, anyeverynowhatevera different
how muchidytamyckydvyjacLuvytadyrdyzdyftyjacpady
using whatymumumumigmumwamumilwamumedmurwamusumupymuftumu
along which wayhjalrimelrilixrilwafrililwafrilocrirwalrisulripjalriftelri
like whatjaDihwaotmihwaickihwawadvihwatjaLvihwaoctihwawartihwauztihwajaftihwatjaptihwa

Personal pronouns

Where sound changes did not wipe them away, personal pronouns retained the old genitive and accusative cases. Due to inherent definiteness of personal prounouns, they do not have a construct form; if a pronoun is possessed by a noun, neither is marked, and their juxtaposition alone shows the relationship. (If a noun is possessed by a pronoun which has a genitive case form, both are marked.)

1st personvwalmlivwalmimlvwalamli
2nd persontufdvitufmidvtufadvi
3rd person femaleemjamiemyhaemja
3rd person male/mixedruromjarumiromyruaromja
3rd person inanimatesirserwasirmiserwasiraserwa

Verb forms

Originally, verb forms were related to noun forms by differences in the stressed syllable: verbs were stressed on the second syllable, and the resultative nominalization was stressed on the first syllable. Other nominalizations, as well as participles, come from the form stressed on the third syllable, if applicable.

NounVerbParticipleDerived nominal
[sad/Sk/s]udthingzdubeudzodubeing[sez/sig/zo]dorjalway of being
[saD/Sk/s]ertagreetingssarjathail, callosridoknown[sot/sik/sa]riDedmimeeting

Verbs themselves come in four basic stems plus two imperative forms: the non-past, past, habitual, and hypothetical stems, and the imperative and negative imperative.

Non-pastPastHabitualHypotheticalImperativeNegative imperative

The imperative is simply the most basic verb form; the negative imperative evolved from a suffixed "su", which may be realized, depending on the verb, as "swa", "zu", or just "z". Verbs whose negative imperative end in "z" only also prefix "waz" for emphasis.

The regular negative particle for other verb forms is a standalone "swaz" which precedes the verb.


"Our language is a way of being."
"Mli vjacgu zdi sezdorjal."
"A language is a tool!"
"Vatjacgu zdi lotsikwar!"
"Don't speak so. The language is a gift."
"Mwaf vitjagzu. Vekjacgu zdi padZi."

Overall evolution of the language

This is how I envisioned the language evolving:

Stage 1: Begin with a simple Polynesian-esque phonology entirely of CV syllables.

Phonology: ptkmlrswh/aeiuo

Stage 2: Apply derivational use of fixed accents. For example, create 3-syllable roots giving an abstract noun concept when accented on the first syllable, a verb root when accented on the second syllable, and various other nominals based on the root created by accenting the third syllable.

Phonology: ptkmlrswh/aeiuo

Stage 3: Apply selected prefixes and affixes which conform to the existing stress pattern - things that were once transparent but will be fossilized by the time the language is done evolving. For example, prefixes to nouns will denote definite, indefinite, or (with no prefix) possessed, with the end results that nouns have definite, indefinite, and possessed root forms, which differ from one noun class to another due to the effects of sound change. Similar use of affixes could create past and non-past, or perfective and imperfective, verb root pairs.

Phonology: still ptkmlrswh/aeiuo

Stage 4: Voicing and lenition: voicing in front of back/rounded vowels, lenition in front of front vowels, but the two have different scope as far as whether accented or ultra-short vowels affect them. (See 4.sc in sound changes below)

Phonology: ptksfcxSbdgzmlrLwh/aeiuo

Stage 5: Drop shortest vowels, or H. (See 5.sc in sound changes below)

Phonology: still ptkmlrswhfcxSbdgzL/aeiuo

Stage 6: Combine double vowels. Maybe prevocalic i->l, u->r. Minimal umlauting occurs. (See 6.sc in sound changes below)

Phonology: ptkmlrswhfcxSbdgzLjv/aeiuoy

Stage 7: Rotate just the stressed vowels, lose stress. Partial application of sound changes in process at the time Mli Vjacgu is being described: deletion of first-syllable unaccented vowels directly preceding an accented syllable, voicing of intervocalic consonants, h-deletion. (See 7.sc and 7prime.sc in sound changes below)

Phonology: ptkmlrswhfcxSbdgzLjv/aeiuoy

Stage 8: Regularize some paradigms by analogy. Add new affixes and clitics using the new phonology. Apply considerable semantic drift to obscure origins of word pairs/triads created in Stage 2. Provide suppletive roots for extremely common words. Create a grammar.

Phonology: still ptkmlrswhfcxSbdgzLjv/aeiuoy

Sound changes

The original sound change concepts are documented below in the Vjatjackwa section, in the emails and the sound change files provided. I have changed the sound change rules for more variety. The vowel changes (e -> i, i -> ja, o -> u, u -> wa, a -> e or o) only happen on stressed syllables now.

For example, several forms related to "speak", one of which will be the name of the language, appear as follows in the old and new versions (alternative forms from the metathesizing dialect are shown where applicable):

Proto-witcku versionVjatjackwa versionMli Vjacgu version
wititkuvjatjatigviDiDig (or jaftiDig)
witikvjatjakwaviDikwa (or jaftikwa)

The sound change rules of Mli Vjacgu, as of February 6, 2006, are captured in the following files (designed for a modified version of Mark Rosenfelder's Sound Change Applier):

Vjatjackwa 2003/2004:

Archival information about the earlier version of the language.

Emails to the Conlang list in 2003/2004:

The sound change rules files that I was using when I stopped work on this in 2004:

Last updated: 2/6/2006