Proto-Occidental

[Ralph]

I've been taking some time of working on Kantaranyan to make a Proto-Language as a parent for it so that I can make related languages too. I've called it Proto-Occidental because it’s the ancestor of the most widespread language family in the western hemisphere of my conworld.

I’ll start things off with a brief description of the phonology:

The phoneme inventory is as follows:

Plosives: /p b t d c ɟ k g kʷ gʷ q/
Nasals: /m n ɲ ŋ/
Fricatives: /s/
Liquids: /r l/
Semivowels: /j w/
Long vowels: /eː aː iː uː/ (I’ll be writing long vowels with a macron elsewhere in this thread)
Short vowels: /e a ə i u/

Phonotactics:

Any consonant may occur in a syllable onset, but no onset clusters are permitted. It is not obligatory for a syllable to have an onset, although onsetless syllables are rare.

A syllable may have a coda, although this may consist of a single consonant only, and may not contain a voiced plosive or /s/. There are, however, no further restrictions on medial clusters. Final /j/ does not occur after /i(ː)/, and similarly /w/ does not occur after /u(ː)/. Syllables closed by a semivowel were typically treated as diphthongs in the daughter languages.

It is rare for vowels to appear in hiatus within a root, and where they do both are often long (for example in the root *sāī (to see). Otherwise, it is rare for two long vowels to occur in the same root. Long vowels can, however, occur in both open and closed syllables.

Dialects:

Proto-Occidental is divided into two early dialects.

The first, which I’m calling Proto-Ngamakan, is characterised by the use of the root *ŋāmak to mean ‘person’ (hence the name), as well as the development of *ɲ to *j in a syllable onset, and *n in a coda, and the loss of syllable final *j and *w after long vowels. Kantaranyan is a Ngamakan language.

The second is Proto-Kattaic, which is characterised by the use of the root *katta to mean ‘person’, and the metathesis of medial clusters whose second element is more sonorous that the first (compare, for example, Ngamakan *taksā with Kattaic *taskā, both meaning ‘tail’).

Finally, here are a few Proto-Occidental roots (with their Kantaranyan reflex given in brackets [note that some of the Kantaranyan forms have been extended with fossilised derivational suffixes]).

*qatū ‘make, build’ (Kant. qatu ‘make’)
*wicu ‘sky’ (Kant. esome ‘cloud’)
*ɟacəw ‘15’ (Kant. xaxu ‘15’)
*katə ‘take’ (Kant. kat ‘take’)
*tēdə ‘give’ (Kant. tiat ‘give’)
*taksā ‘tail’ (Kant. tatxa ‘tail’)
*tegʷe ‘red’ (Kant. teve ‘be red’)
*darēt ‘bird’ (Kant. tarieh ‘bird’)
*ŋakaŋ ‘heart’ (Kant. akan ‘heart’)
*gʷage ‘head’ (Kant. vahe ‘head’)
*juqsā ‘foot’ (Kant. yotxa ‘foot’)
*kaŋə ‘tree’ (Kant. kan ‘tree’)
*raŋja ‘area of land’ (Kant. ranya ‘island’)
*gasī ‘sea’ (Kant. haxi ‘sea’)
*awmam ‘back’ (Kant. umon ‘back’)
*ŋəru ‘fish’ (Kant. oroneh ‘fish’)
*piqta ‘high place’ (Kant. petta ‘hill; mountain’)
*qimā ‘night’ (Kant. qema ‘night’)
*cagīn ‘day’ (Kant. sahin ‘day’)
*caŋkī ‘stand’ (Kant. sanki ‘stand’)
*rēsat ‘water’ (Kant. reaxeh ‘water’)
*gʷīja ‘fire’ (Kant. viyaxi ‘fire’)
*ɲegī ‘man’ (Kant. yehi ‘man’)
*agʷēt ‘woman’ (Kant. avieh ‘woman’)
*ŋāmak ‘person’ (Kant. amah ‘person’)
*kele ‘boy’ (Kant. kele ‘boy’)
*ŋəjrit ‘girl’ (Kant. ireh ‘girl’)
*nənī ‘child’ (Kant. neni ‘child’)
*taŋra ‘tongue’ (Kant. tanra ‘tongue’)
*wəɟa ‘eat’ (Kant. axa ‘eat’)
*gʷālan ‘drink’ (Kant. valan ‘drink’)
*batum ‘hear’ (Kant. patom ‘hear’)
*sāī ‘see’ (Kant. xai ‘see’)
*piwla ‘fruit’ (Kant. pula ‘fruit’)
*gʷaciw ‘hand’ (Kant. vasu ‘hand’)
*ɟarma ‘sun’ (Kant. xavma ‘sun’)
*līnə ‘moon’ (Kant. lin ‘moon’)
*katu ‘star’ (Kant. kato ‘star’)

Hopefully I'll have some morphology for you this weekend.

[Chagen]

This is very nice looking.

I always have trouble with this kind of historical linguistics stuff, so I despect those who can do it.

[Ralph]

This is very nice looking.

I always have trouble with this kind of historical linguistics stuff, so I despect those who can do it.

Thanks! Glad you like it

While I'm here, one thing I forgot to mention in the first post was that the contrast between plain and labialised velars is neutralised before (long and short) /u/.

[CMunk]

despect

I guess you meant to hit the r-key

[Ralph]

Morphology:

Nouns:

Nouns inflect to denote plurality and possession. The plural is indicated by the suffix *-r. When the plural suffix is added to a noun ending in a consonant, said consonant is elided, and the preceding vowel lengthened in compensation. There is also a dual form using the suffix *-n (which has the same effect on a consonant final root), which is most commonly used for objects which typically appear in pairs (e.g. hands).

Examples of plurals:

*katu ‘star’ > *katur ‘stars’
*ŋāmak ‘person’ > *ŋāmār ‘people’
*gʷaciw ‘hand’ > *gʷacīn ‘hands’
*ɲūsa ‘eye’ > *ɲūsan ‘eyes’

Possession is marked by a personal prefix, along with a suffix if the possessor is plural (or dual).

The possessive prefixes are the following:

1st person singular: ni-
1st person plural inclusive: ŋu-
1st person plural exclusive: si-
2nd person: lu-
3rd person animate: mi-
3rd person inanimate: a-

If the possessor is plural (including in the case of the 1st person where there are separate plural prefixes), the suffix –mī must be added to the noun.

Examples (using *gʷaciw ‘hand’):

nigʷaciw ‘my hand’
ŋugʷacīrmī/sigʷacīrmī ‘our hands’
migʷaciw ‘his/her hand’
migʷacīrmī ‘their hands’

Several daughter branches subsequently developed further affixes by fusing clitics with nominal stems. In Kantaranyan, for example, several specifiers and postpositions eventually developed into further nominal affixes.

Verbal:

Verbs inflect for aspect and voice, using the following affixes:



The perfective and anterior affixes are suffixed to consonant final stems, and infixed before the final vowel of vowel final stems (compare, for example *imik agi wəɟija ‘he ate it’ with *imik agi gʷālanī ‘he drank it’). The imperfective inverse, passive and antipassive suffixes follow the same morphonological rules as the nominal plural and dual suffixes (compare, for example *niq wəɟat ‘I am eating’ with *niq gʷālāt ‘I am drinking’).

The direct form is the basic form of the verb, and the only form available for intransitive verbs. The other voices are used as follows:
-the inverse form is used to signal that the subject is lower in the animacy hierarchy than the object.
-the passive is used in order to allow the subject to be omitted (e.g. *agi cakʷīra ‘it broke’ vs. *niq agi cakʷija ‘I broke it’)
-the antipassive is used to allow the object to be omitted (e.g. *mīr wəɟat ‘they are eating’ vs. *mīr piwla wəɟa ‘they are eating fruit’).

(Hopefully that's all clear enough)

[Ralph]

I've been working on coming up with some Proto-Occidental vocabulary over the past week. When I've got a bit more (and have some more sound changes worked out) I might put up some cognate sets to illustrate the development of some of the proto-phonemes in the different branches.

For now, I thought I'd put up the number system I came up with a couple of days ago (just the cardinal numbers for now, the ordinals need more work). The Proto-Occidental number system is base twenty, but the numbers from 6 to 20 are built in groups of five, which each number other than the multiples of five being a blend of the multiple of five below and one of the numbers 1-4. This is probably best illustrated by the numbers themselves, so here they are:

1: *sāri
2: *wakū
3: *māni
4: *kurda
5: *mardi
6: *sāmar
7: *wakmar
8: *māmar
9: *kurmar
10: *tītu
11: *sāti
12: *waktī
13: *māti
14: *kurtī
15: *ɟacəw
16: *sāɟa
17: *wakɟa
18: *māɟa
19: *kurɟa
20: *mantē

There's also a word for 400, *māman, but no higher numerals can be reconstructed. Proto-Ngamakan has a word for 8000, *gaɟan, although in Proto-Kattaic that word simply means 'a large amount'.

[Ralph]

I've come up with the following kinship terms for Proto-Occidental. As this is the first attempt I've made at seriously working out a set of kinship terms for a conlang, I'd very much appreciate it if people would give me feedback on it.

I envision the Proto-Occidental speakers to have had some sort of matrilineal and uxorilocal kinship system, and consequently the following works on the basis that they would be more interested in distinguishing between relatives on the mother's or wife's side of the family.

Morphologically, the kinship terms are built from roots with signify the relation in question, plus a suffix indicating their sex: *-gʷē for female relatives, and *-gī for males. A wife's relatives can be signified by prefixing *gʷat- (related to *gʷadəw 'second') to the term.

The roots are the following:

Mother: *ma-
Relative in father’s lineage: *tak-
Sibling: *nup-
Mother’s sibling: *kun-
Own child: *kel-
Child of a male member of own lineage: *tir-
Sister’s child: *baj-
Mother’s sister’s child: *baw-
Grandmother: *tam-
Spouse: *təj-
Ancestor (general term for a relative within one’s own lineage in an older generation that one’s grandmother: *ɟūti (or, if sex-specific *ɟutgʷē/*ɟutgī).
Member of one’s own clan: *kūmi (general); *kumgʷē/*kumgī (sex specific)

Other relations are denoted by possessive constructions: for example, 'husband's mother' would be *təjgī mimagʷē (spouse-male 3.animate.possessive-mother-female)

[Ralph]

Apologies for disappearing for the past few weeks - got distracted by Easter holidays and coursework.

Thought I ought to have something to post for coming back, and, given I've come up with at least a bunch of sound changes for other Occidental languages that I don't plan on developing in as much detail as Kantaranyan and Eijian, I thought I'd post some brief descriptions of them in this thread. Here's a language I'm calling Kammar, which is closely related to Kantaranyan.

Sound laws:
As Kammar is closely related to Kantaranyan, many of the phonological developments are the same. The main differences are noted below:
• PBN *ɛ and *ɔ diphthongise as /i̯a/ and /u̯a/ respectively, except after uvular consonants, when they become /a/ (e.g. PBN *awɛt (woman) becomes /aβi̯aʔ/, but *ʁɛsat (water) becomes /ʁaʃaʔ/)
• PBN *b and *d become /β/ and /l/ in all positions.
• Syllable final consonants are debuccalised or lost, but without colouring the preceding vowel. The loss of a liquid results in lengthening of the vowel and the loss of a nasal in nasalisation, while a syllable-final plosive develops into a glottal stop.
• *ŋ is not lost in initial position, and instead becomes /m/ before a rounded vowel, and /n/ otherwise.

Phonology:

Phoneme inventory:

/p t k q ʔ/ <p t k q ʔ>
/m n/ <m n>
/s ʃ h/ <s x h>
/β ʝ ʁ/ <v y r>
/l/ <l>
/i e a o u/ <i e a o u>
/iː eː aː oː uː/ <ii ee aa oo uu>
/ĩ ẽ ã õ ũ/ <in en an on un>
/i̯a u̯a/ <ia ua>
/i̯aː u̯aː/ <iaa uaa>
/i̯ã u̯ã/ <ian uan>

Phonotactics:

All syllables are either open, or, if they have a short, oral vowel or diphthong in the nucleus, may be closed with a glottal stop.

Diphthongs and high vowels may not occur after /q/ or /ʁ/.

All syllables must start with a consonant. A syllable initial glottal stop only appears at the beginning of a word, or after an open syllable with a short, oral vowel or diphthong.

Words are stressed on the second-to-last syllable.

Grammar:
Grammatically, Kammar is generally similar to Kantaranyan.

Nouns form plurals by converting the final vowel into the equivalent long vowel (and deleting a final glottal stop): e.g. pula ‘fruit’ > pulaa ‘fruits’, sahin ‘day’ > sahii ‘days’, ʔaviaʔ ‘woman’ > ʔaviaa ‘women’ etc.

Kammar has rather more irregular plurals, often retaining reflexes of the short vowel + consonant vs. long vowel + r alternations inherited from Proto-Occidental (these have been eliminated by analogy in Kantaranyan). Kammar also has some nasal vowel plurals corresponding to Kantaranyan’s irregular plurals in –n (e.g. vasu ‘hand’ > vasin ‘hands’).

Like in Kantaranyan, verbs take possessive affixes, which are much the same: ni- ‘my’, lu- ‘your’, mi- ‘his/her/their’ ʔa- ‘its/their’, mu- ‘our (inclusive), xi- ‘our (exclusive)’, -mi (plural possessor).

Kammar verbs show the same inflectional categories as in Kantaranyan, but the endings themselves have become suffixes. Verbal roots are extended by k before adding suffixes (probably originally an auxiliary verb, perhaps from POC *kā ‘do’), except for stems ending in a nasal vowel, which add n instead (compare for example neʔ ʔahi ʔaxaki ‘I ate it’ with neʔ ʔahi valanni ‘I drank it’, from the verbs axa (eat) and valan (drink) respectively).

Not quite sure on how to romanise the glottal stop really, so I've just stuck to the IPA symbol. I've anyone's got any better suggestions, go ahead and let me know.