My Proto-Language and Where to Go from Here

A forum for all topics related to constructed languages
Post Reply
teotlxixtli
cuneiform
cuneiform
Posts: 85
Joined: 05 Jan 2021 04:37

My Proto-Language and Where to Go from Here

Post by teotlxixtli »

Below is my naturalistic proto-language, as of yet unnamed, from which I want to derive three daughter languages. The phonology was inspired by Swahili, but the grammar is a hodgepodge of features I happen to like. First I'd like to know from the community what I've done well, what I can improve on, and what kind of sound changes I can apply and what kind of new grammar I can introduce to develop the daughter languages.

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

PHONOLOGY

Consonants

m, n
p, b, t, d, k, g, ᵐb, ⁿd, ᵑɡ
tʃ, dʒ, ⁿdʒ
s, z, ⁿz
f, v, h, ᵐv
j, w
r
l

Vowels

i, u
e, o
a

(C)V syllable structure
Vowels must be separated by a consonant
[h] is inserted between vowels in compounds
Consonant clusters are forbidden
Penultimate stress 

GRAMMAR

Agglutinative language
Adjectives are actually types of verbs

Verb-Subject-Object word order
Noun phrases are consistently head-initial
Adjectives follow nouns
Possessors follow possessees
Auxiliaries precede lexical verbs
Prepositions

Personal pronouns are distinguished by person (1st, 2nd, and 3rd Person), number (Singular, Dual, Paucal, and Plural), and gender (Masculine, Feminine, Neuter) in the 3rd Person

Personal Pronouns
Singular Dual Paucal Plural
1st Person u udʒani uniⁿdi utʃupa
2nd Person ka kadʒani kaniⁿdi katʃupa
3rd Person Masculine wi widʒani winiⁿdi witʃupa
Feminine da dadʒani daniⁿdi datʃupa
Neuter po podʒani poniⁿdi potʃupa

Nouns are marked for plurality and possession
There are four grammatical numbers: Singular (one thing), which is unmarked; Dual (two things), which is marked with –dʒani; Paucal (between three to eight things), which is marked with -niⁿdi; and Plural (greater than eight things), which is marked with –tʃupa
Mass nouns are those to which adding more of the substance does not result in countably more of the substance, like liquids, gases, and abstractions, among other things
Possession is indicated by suffixing the personal pronoun of the possessor onto the noun being possessed; abi is “chest”; abihu is “my chest”, abika is “your chest”, abiwi is “his chest”, etc.
Case relationships are indicated with word order (in the case of subject and object) and prepositions (in the case of all other arguments); aᵑɡo bawu is “off the person”
Augmentatives are formed by suffixing –(h)esi onto the noun (the [h] being necessary to break up vowel sequences); dohu is “fire”, dohuhesi is “conflagration”. Diminutives are similarly formed by suffixing -puⁿda; dohupuⁿda is “spark, cinder”

Subject and object are marked on the verb in the order ROOT-SUBJECT PERSONAL PRONOUN – OBJECT PERSONAL PRONOUN
- dʒaᵐboᵐvi + (h)u = dʒaᵐboᵐvihu, meaning “I insist”
- emi + (h)u + ka = emihuka, meaning “I bother you”
- There are 400 possible combinations of subject and object on a verb
There are two tenses; Future (marked with the prefix ⁿdoᵐvu-) and Non-future (unmarked)
- fawohu means “I work”; ⁿdoᵐvufawohu means “I will work”
There is also the Infinitive, formed with the prefix ᵐba-. Note that subject and object marking does not appear on infinitives
- ᵐbafawo means “to work”
Further aspectual and modal distinctions are formed with auxiliary verbs. Auxiliary verbs receive all verb marking and the lexical verb follows that complex
- The Perfective is unmarked
- The Imperfective is marked by reduplicating the first syllable of the verb; gedo is “scream”, gegedo is “screaming”
- The Perfect is marked with the auxilliary verb ⁿdoⁿzu; ⁿdoⁿzuhu jaⁿdu means “I have listened”; when combined with the Future tense marker the Future Perfect is formed; ⁿdoᵐvuⁿdoⁿzuhu jaⁿdu means “I will have listened”
Verbs are negated by following the verb with howi; hohotʃehu is “I worship”; hohotʃehu howi is “I do not worship”
Reflexives are formed by possessing the word “self” ⁿdo with the pronoun doing the reflexive action. This is combined with the same pronoun appearing in the subject and object verb slots; idohuhu ⁿdohu means “I excite myself”. Saying idohuhu would be like saying “I excite me”
This is not the case for the third person, where the absence of the reflexive pronoun complex is interpreted as the presence of separate arguments; idowiwi by itself would mean “he sees him”, wherein the second third person personal pronoun is a different person than the first refers to
Adjectives act like verbs and are marked in the subject slot on the verb template for the personal pronoun of the noun they modify, though they do not take tense marking; dʒahaⁿdʒe ajipo means “tiny brain”
Nouns are distinguished by gender when combined with adjectives. Thankfully gender is totally semantic. When agreeing with the adjective in person and number, the subject marking slot in the adjective template must agree in gender with the noun it modifies
- Masculine nouns refer exclusively to nouns refering to male humans and their respective titles, occupations, etc as well as male animals. iti, meaning “man”; iti bawutafiwi meaning “free man”
- Feminine nouns refer exclusively to nouns refering to female humans and their respective titles, occupations, etc as well as female animals. esa meaning “woman”; esa bawutafida meaning “free woman”
- Neuter nouns refer to anything not included in the previous categories, including non-gendered human terms; bawu, meaning “person”; bawu bawutafipo meaning “free person”

Numbers are base-eight. Numbers greater than eight are formed with the sequence NUMBER + lo + uᵐbu up to the number fifteen; tʃedi lo uᵐbu means “fourteen”. Numbers do not agree in gender with numbers, and merely follow the noun they modify; eⁿdo aro means “one cat”

Comparatives are formed with fixed expressions
- ᵐboᵐvipahu iⁿdʒa means “I am tall”; ᵐboᵐvipahu iⁿdʒa he means “I am taller”
Articles follow the nouns they appear with; fitʃo ⁿdʒi is “the fish” (as well as “that fish”); fitʃo aro is “a fish” (as well as “one fish”)
Yes-no questions are formed by ending the question with either “yes” ewo or “no” howi, depending on what the expected answer is
- ⁿdoᵐvuhusokapo upipo, ewo? means “You will see the house, yes?”
All other questions are formed by plugging in the interrogative pronoun in the sentence wherever appropriate; usokapo bu? means “What do you see?” (literally “You see what?”)
User avatar
qwed117
mongolian
mongolian
Posts: 4094
Joined: 20 Nov 2014 02:27

Re: My Proto-Language and Where to Go from Here

Post by qwed117 »

Well, the language looks fine, although it would be nice to have a greater view of its lexicon. From my understanding, in most languages the "biggest change" between languages is typically lexical replacement- that is that the definitions of words change, or that words are replaced by substrate or adstrate words.

If you have an idea for the kind of aesthetic you would like for the daughter language (perhaps what kind of phonology or phonotactics you would prefer), you can start by listing the phonology you want and the phonotactics. Let's say I want to go from this language to a phonology like this:

/m n ŋ/
/b t t͡ʃ d d͡ʒ k g/
/f v θ s/
/w j l r/
/i ɪ ʊ u/
/e ɛ ɔ o/
/a/
with (C)V(C) phonotactics.

Starting with /ˈiti bawutaˈfiwi/ "free man" and /hohoˈtʃehu/ "I pray"
1) I might start with saying /h/ causes preceding vowels to lax
2) vowels immediately before the penult become /ə/
3) then perhaps have a chain shift that ᴺD > D > T > Þ
4) convert /w/ in stressed syllables to /v/, and collapse syllables of /wB/ to /u/
5) {D Þ} -> Ð behind a /ə/
6) All vowels behind a voiceless consonant are laxed, already lax vowels become ə, except in stressed syllables,
7) the last /ə/ is deleted in a word, and all other /ə/ become /o/
already we have
8) backwards voicing assimilation and deletion of postvocalic glottal consonants, and initial glottal consonants
9) CVC antepenults get preferential stress
10) p > f universally
11) unstressed initials are fronted and laxed
12) finals of an antepenult-stressed word are deleted
12) and let's say the definitions change a bit

/iθɪˈfauθfiv/ "person" and /eˈʃɛhu/ "I say"

This looks a lot different from the origin, and fulfils all the goals we set up earlier! Obviously, there's more to go, though to get further from the source. You can look at things like the Index Diachronica or PBase to get ideas for more changes (personally my preference is for the former). Another option is to use scattered changes with no particular "goal" aesthetic. This can work decently too! That being said, it sometimes is messy, so you might need to look at the output after some amount of changes, with the intent to look for and create and aesthetic.
Spoiler:
My minicity is [http://zyphrazia.myminicity.com/xml]Zyphrazia and [http://novland.myminicity.com/xml]Novland.

Minicity has fallen :(
The SqwedgePad
teotlxixtli
cuneiform
cuneiform
Posts: 85
Joined: 05 Jan 2021 04:37

Re: My Proto-Language and Where to Go from Here

Post by teotlxixtli »

Thanks qwed, some of the sound changes you suggested were knocking around my head before I even posted. Some Bantu languages have that wonderful advanced tongue root harmony thing going on with /i ɪ u ʊ e ɛ o ɔ a/ and I can see a number of different ways to derive them (although several involve losing the prenasalized consonants, which I'm hesitant to do since they're the aesthetic backbone of the whole language family as I'm imagining it). I want at least one of the three daughter languages to sound a little more like Arabic, and losing [p] was also an idea I'd had.

As for the vocabulary, I have about 900 words, both root words and compound/derived ones, about most general areas of topic. With a bit of finagling you can say pretty much anything you need to say. I can post it but it would be quite a list to read.
DV82LECM
sinic
sinic
Posts: 280
Joined: 16 Dec 2016 03:31

Re: My Proto-Language and Where to Go from Here

Post by DV82LECM »

Gotta say, this is actually cool to see what a Bantu language could look like, you know, given time, especially, to my knowledge, none of those languages that have inspired this HAVE, yet.
𖥑𖧨𖣫𖦺𖣦𖢋𖤼𖥃𖣔𖣋𖢅𖡹𖡨𖡶𖡦𖡧𖡚𖠨
teotlxixtli
cuneiform
cuneiform
Posts: 85
Joined: 05 Jan 2021 04:37

Re: My Proto-Language and Where to Go from Here

Post by teotlxixtli »

An update on my first daughter language: I've made the following sound changes so far, which have led to all kinds of weird grammatical mix-ups (like f turning to b when certain verbs are conjugated, or phantom vowels appearing and disappearing with declensions). I've been trying to derive inspiration from other desert-dwelling African languages (because that's the area of the world the speakers of this language will inhabit) but haven't come up with enough sound changes yet. Losing p and tʃ was derived straight from Arabic's sound inventory, but I can't find any way to make it more Arabic (like adding x and ɣ or even q). Any suggestions?

1. p is realized as b word-medially and f word-initially
2. tʃ is realized as dʒ word-medially and ʃ word-initially
3. s is realized as ʃ before i
4. Unstressed vowels are lost between obstruents and w
5. h is lost word-initially
6. Unstressed word-initial vowels are lost
7. Word-final vowels are lost following m n t k r l
8. j is realized as ʒ
dva_arla
cuneiform
cuneiform
Posts: 184
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Re: My Proto-Language and Where to Go from Here

Post by dva_arla »

Some comments:
2. tʃ is realized as dʒ word-medially and ʃ word-initially
tʃ > ʃ is, typologically and phonetically speaking, more likely to happen coda-finally than coda-initially. (edit: I stand corrected. You have perhaps read of Tamil.)

But if you're going to voice all initial /tʃ/'s (presumably in order to eliminate the phoneme) you should affect the same change for all initial /s/'s, for reasons of symmetry. Typologically the latter is more likely (and therefore the earlier in the scheme of things) to happen.

And if you want more quirks I suggest something the likes of Verner's law:

https://en.wikipedia.org/wiki/Verner%27s_law

whose outcome (in the case of Pr.-Germanic, voicing) depends on whether the preceding vowel is stressed (in your conlang, the vowel could as well be one following).
3. s is realized as ʃ before i
And then neutralisation of vowel qualities in unstressed syllables (whether pre-stressed or post-stressed ones or both), thereby bringing /ʃ/ closer to phonemicisation? In Irish almost all unstressed vowels are reduced to schwa (exceptions are matters of detail and therefore not taken the trouble to be elaborated here).

Perhaps also vowel breaking in the stressed ones: Pr.-Mongolic *sïra > *sjara > Khalkha шар?

As for the uvular sounds, you might want to look at how Kurdish pops up /q/ (and maybe /ħ/ and /ʕ/) out of the blue in native words. There may be papers out there discussing the phenomenon (for lack of a better term in my mental lexicon-- I fear that my vocabulary has recently begun to dry out of my mind, therefore reducing the quality of my writing).
all kinds of weird grammatical mix-ups (like f turning to b when certain verbs are conjugated, or phantom vowels appearing and disappearing with declensions)
a.k.a Grammatische Wechsel. Noice! Hence Verner, and you might want to extend like changes to the other stops.
Conlangs in progress:
Modern Khotanese
Modern Gandhari
?? - Japonic language in the Mekong Delta
Locna - Indo-European language in N. Syria
Wexford Norse
A British romlang, &c.
dva_arla
cuneiform
cuneiform
Posts: 184
Joined: 25 Oct 2019 21:03
Location: Realm of Ideas

Re: My Proto-Language and Where to Go from Here

Post by dva_arla »

The following chain-shift popped up all-of-a-sudden in my mind:

/s/ > /z/ (enviromentally conditioned, either word-mid [as in English] or word-mid + initially [as in German]).
Which pushes /z/ to > /ʒ/, then
/ⁿz/ > /dz/ > /dʒ/, (in turn) > /ɣ/

For your inspiration. You might find it too drastic or 'mild' enough for your tastes; I'm not too much acquainted with them. But again, it's your language, so.. you might only take in a part or a stage of this chain, or all, or none... your call.
Conlangs in progress:
Modern Khotanese
Modern Gandhari
?? - Japonic language in the Mekong Delta
Locna - Indo-European language in N. Syria
Wexford Norse
A British romlang, &c.
teotlxixtli
cuneiform
cuneiform
Posts: 85
Joined: 05 Jan 2021 04:37

Re: My Proto-Language and Where to Go from Here

Post by teotlxixtli »

The full list of sound changes I ended up going with resulted in the Ujo thread I posted a little while ago:
viewtopic.php?f=6&t=7440

But to summarize the full change list, it is:

1. p is realized as b word-medially and f word-initially
2. tʃ is realized as dʒ word-medially and ʃ word-initially
3. s is realized as ʃ before i
4. Unstressed vowels are lost between obstruents and w
5. h is lost word-initially
6. Unstressed word-initial vowels are lost
7. Word-final vowels are lost following m n t k r l
8. j is realized as ʒ
9. h is lost between like vowels
10. Double vowels are realized as long vowels
11. Word-final vowels are lost following s ʃ f
12. a: e: i: o: u: are realized as a e i o u in unstressed position
13. Vowels prior to prenasalized consonants become long if not already
14. Prenasalized consonants merge with modal counterparts

For whatever reason I really like the fact that Ujo has no /j/, it really gives it a distinct flavor.
Post Reply