Hoavi

[Avo]

I've been working on this for quite a while now, so I might as well share it. The language is part of my conworld's Muran languages and also a member of the Bay Linguistic Area, a Sprachbund I'm working on.

The Hoavi call themselves Kasivot [ˈk̠ʰɑsʲɪβoʰt] (sg. Kasivo) and their language Kasimmo mai [ˈk̠ʰɑsʲɪʔbmoˌmɑi̯]. The term Hoavut [ˈho̯ɑβɨʰt] is also used but includes some other closely related but distinct people and their languages. The word comes from a Proto-Muran word meaning "real, true" as in "true people" and is used in opposition to hukkwut [ˈhɨʰkʷɪʰt] "other (non-Muran speaking) people".

I'll continue with a brief outline of the phonology and the orthography I use.

Phoneme inventory

Hoavi has 16 consonant phonemes.

Additionally, there is a marginal phoneme /ʔ/ that only appears between vowels. Plosives are heavily aspirated word initially and preaspirated when geminated or syllable finally. /p t k kw/ are also lenited to [β ð ɣ ɣʷ] between vowels.

Hoavi's vowel inventory is a bit skewed, with a gap in the close-back region:


The vowels are written <i ö e u o a>. Long vowels as well as a bunch of diphthongs exist but are analyzed as two vowels.

Both vowels and especially consonants are subject to great allophonic variation, which means, even though Hoavi orthography is mostly phonemic, the exact pronunciation might look unexpected at first. I'll discuss allophony in a separate post though.

Stress & Syllable structure

Hoavi's basic syllable structure is CV(V)(C). Consonant clusters occur, but are limited and only found between vowels. Words may only end in a vowel or /t s n l/.
All Hoavi words must be at least two moras long. Hence, monosyllabic words undergo vowel lengthening or their syllable is closed by a glottal stop:

/ŋɨ/ > [ˈŋɨː] "blood"
/βe/ > [ˈβɛː] "no"
/pa/ > [ˈpʰɑʔ] water"
/jo/ > [ˈjoʔ] "good"

Stress is assigned from left to right. Regularly, stress is fixed on the heaviest syllable of the root. If there is no heavy syllable, the first syllable will be the stressed one:

/la/ [ˈlɑː] "palm (of hand)"
/mo.a/ [ˈmo̯ɑ] "already"
/ho.ŋo/ [ˈhõŋo] "tree"
/o.ra.βi/ [ˈʔoʐɑβɪ] "squirrel"
/i.ron.ko/ [ʔɪˈʐõn̥ko] "backwards"
/ro.koi/ [ʐoˈɣo̯i] "thing"

I'll stop here for now. I'll probably post more about the phonology next, or some grammar, when I feel like it. Any thoughts/flaws/peanuts so far?

[Creyeditor]

I really like it so far, especially the velar-labiovelar-distinction. It's very funky for a language to lack /u/, but I guess it will appear as an allophone of /o/, won't it?
However I have a question: Is this sentence

Most notably, Hoavi merged /ʔ/ with /h/ and pre-glottalized consonants with pre-aspirated ones.

referring to the preceding sentence? If so, I did not quite understand it.

/ro.koi/ [ʐoˈɣo̯i] "thing"

I really like [ʐ] as an allophone of /r/ and I forgot to mention that I like the structure of your consonant inventory
I am looking forward to seeing more of this

[MONOBA]

Please post more!

[Avo]

I really like it so far, especially the velar-labiovelar-distinction. It's very funky for a language to lack /u/, but I guess it will appear as an allophone of /o/, won't it?

Thanks! There really is no /u/ in Hoavi, not even as an allophone. /ɨ/ can have the allophone [ɯ] though.

However I have a question: Is this sentence

Most notably, Hoavi merged /ʔ/ with /h/ and pre-glottalized consonants with pre-aspirated ones.

referring to the preceding senten? If so, I did not quite understand it.

I meant that there are more conservative dialects with /ʔ/ and pre-glottalized consonants, but that Hoavi merged them with /h/ and pre-aspirated consonants.

Please post more!

I will!

A few more words on Vowels

The pronunciation of vowels in Hoavi is pretty straightforward. /ø o/ are always [ø o] as in tsöhmiet [ˈtsʲʰøʰpmɪ̯ɛʰt] "all, every", noo [ˈnoː] "name" and /e a/ are [ɛ ɑ] as in eje [ˈʔɛjɛ] "bow" and aala [ˈʔɑːlɑ] "dog".
/i/ is [ɪ] almost universally, except when preceded by another vowel, where it is [ i]: kinnu [ˈkʲʰɪʔdnɨ] "hole", hiisi [ˈhiːsʲɪ] "swan".
/ɨ/ varies between [ɨ~ə~ɯ], with [ə] being found in unstressed position and [ɯ] after labials, mostly. The most common realization in all positions, however, is [ɨ]. Note that one might also encounter [ɯ] as main allophone, especially in the speech of females.

As mentioned above, Hoavi allows a large range of consecutive vowels, sequences ending in /i/ being the most common by far. /ai/ is also the only sequence allowed starting in /a/. In /Vi/ sequences, /i/ will be the non-syllabic part, in all other sequences the first vowel is non-syllabic.
There are no sequences ending in /ɨ/ and /e/ may only be preceded by /i/. /e ø o/ also don't occur one with another in any order.

There are some notable sound alternations involving vowels. I'll describe some of the more common ones above.

/i ~ ɨ/

This alternation is so common because it involves the very productive nominative suffixes -Pi and -Ti (*-pa and *-ta in Proto-Muran). Hoavi changed *a to /ɨ/ when followed by another /a/. Thus, when the plural marker *-ta was added, the endings *-pa-ta and *-ta-ta yielded into /-Pɨt -Tɨt/ in Hoavi, creating the alternation between /i/ and /ɨ/:

kotsi [ˈkʰotsʲɪ] "man", kotsut [ˈkʰotsʲɨʰt] "men" ( < PM *kaca, *kaca-ta)

This pattern became generalized so that some nouns ending in /i/ also show it even if there is no historical reason for that:

ongi [ˈʔoŋɪ] "k.o. evil spirit" > ongut [ˈʔoŋɨʰt] ( < PM *waŋle(-ta) from *waŋ- "to harm")

/Vi ~ Vr/

This pattern arose when a historical word-final *n was lost, which was retained when suffixes where added to the word. Intervocalic /n/ later changed to /r/, thus, some final /i/ alternate with /r/.

poi [ˈpʰoi̯] "sibling of different sex (ABS)" > poron [pʰoˈʐõŋ] "idem (OBL)". ( < earlier **pon, **ponon)

There are a few others, but I will deal with them later, as I will with ablaut patterns on verb stems.

[Ceresz]

I'm afraid I don't have much to say other than that I love this phonology. Like Creyeditor, I especially enjoy the realization of /r/ as [ʐ]. I actually realize my /r/'s as [ʐ] (or something close to that, at least) from time to time .

The vowel inventory looks a bit weird, but it's the good kind of weird. I like it all the more because it's not something I would have done.

Oh, and

Please post more!

[Avo]

Posting more! I decided to skip the remaining phonology parts and start with some real stuff.

I don't really know where to begin, so here are some random facts about the language. Hoavi is a predominantly verb-initial language with VSO as unmarked word order. There is a strong animacy hierarchy and animates are treated differently from inanimates in several parts of the grammar. The language also features direct-inverse voice and split ergativity, even though the "ergative" case has a range of other functions aswell, which is why I prefer to call it oblique here instead.
Nominal morphology is relatively simple. There is some suffixation, compunding and reduplication. Apart from that, there are a number of verbalizing suffixes both for nouns and ajectives.
Verbs are quite an open class in Hoavi and there are several devices to form new ones. In fact, most verbs are derived and some verb roots never occur in their "underived" form, always requiring at least a dummy derivational morpheme. This also means, that verbs never occur uninflected. There are quite a few "complex verbs", meaning that they consist of two or rarely more combined verb stems. A complex verb can also incorporate a noun, however, noun incorporation is limited and occurs in constructions meaning "to do X by/with Y", mostly.

So that's basically it, for now. I'll post some words on verbs when I've more time tonight.

[Avo]

Introducing: Verbs

Verbs in Hoavi are either simple, consisting of one verb root or an incorporated noun and a verb root, or complex, being formed by two or more verb roots. The division between verb root and verb stem is very strict and most verb roots need derivational morphemes to form a verb stem. For example, some simple verbs with the verbalizing suffix -n:
(v)oa- "sleep" > voa-n- "to sleep" (sleep-VERB-)
ka- "do" > hopsi-ka-n- "make bread" (bread-do-VERB)

Verbs can also be formed out of nouns and, more rarely, adjectives. For example, the suffix -ng forms intransitive verbs:
olma "human" > olma-ng- "to be human"
vie "day" > vie-ng- "to stay, last"

As mentioned above, verbs can be formed by two or more consecutive verb roots:

Repputut luika
[ˈʐɛʰpɯðəʰt̪ ˈl̪ɨi̯ɣɑ]
rep-put=t luika
burst-fall=3IN rain
"Suddenly, it starts raining"

The verb in the example consists of the roots rep- "burst" and put- "fall, precipitate" to create a new verb meaning "to suddenly fall". Complex verbs with more than two verb roots occur but are rare. This is, however, the difference between complex verbs and simple verbs with incorporated nouns. While the latter can take two roots (noun + verb) at most, the former can take up to three ([verb or noun]+verb+verb).

I feel like I should continue with person marking here, but that's one of the more complicated parts of the language and should have its own post.

[Prinsessa]

Repputut luika
[ˈʐɛʰpɯðəʰt̪ ˈl̪ɨi̯ɣɑ]

I cannot even begin to explain how æsthetically appealing and sexy this is to me.

[Avo]

Hoavi is a predominantly verb-initial language. As mentioned above already, it has a strong animacy hierarchy and uses a direct-inverse voice system. For person marking, the language uses pronominal enclitics that always appear at the second position of the clause, which means they do not always attach to the verb.
In its least complex form, a Hoavi sentence consists of a simple verb followed by a pronominal enclitic indicating the subject:
Völnu
['βølʔdnɨ]
völ=nu
run=1SG
"I run"

Additional information, like aspect or evidentiality, is optional. The clitic slot on the verb can't be empty though, and if a pronominal doesn't attach to the verb, the verb final enclitic =(a)s is used instead:
Vippai immuornunas.
[ˈβɪʰpɑi̯ ɪʔˈbmɨ̯oɭʔdnɨˌnɑsʲ]
vit=pai imm-uor-nan=as
then=3PL dance-INCH-HAB=FIN
"And then they would start dancing"

First person is indicated with =nu "1SG" and =nut "1PL", second person with =n "2SG" and =met "2PL". For third person animates, =i "3SG.An" and =Pai "3PL" are distinguished. For animates, the pronominal clitic =t is used regardless of number. In general, these enclitics refer to the subject of a verb if there is no inverse marker attached to the verb or the verb is intransitive. They can also denote the object or sometimes possessors.



Unrelated, but I thought about adding /θ/ to the consonant inventory to make it more balanced. I'm not sure how it fits into the sound of the language, but I'll give it a try.

[Creyeditor]

Unrelated, but I thought about adding /θ/ to the consonant inventory to make it more balanced. I'm not sure how it fits into the sound of the language, but I'll give it a try.

I always liked the absence of englishy but otherwise uncommon /θ/
However, it's your conlang and you can do what you want

[Avo]

I always liked the absence of englishy but otherwise uncommon /θ/
However, it's your conlang and you can do what you want

It's not common but it appears in a number of Native American languages, like Mapuche, where I really like it. And if I add /θ/, every stop would have its corresponding fricative.

Anyway! Let's move on to one of the more complex parts of the language: Inversion and animacy

I mentioned already that the pronominal clitics don't always refer to the verb's subject, but can aswell refer to its object. Generally, when there is a transitive verb in a sentence, the most animate entity is implied to be the subject:
Aikoremi nuori haakwi
[ʔɑi̯ˈɣoʐɛmɪ ˈnɨ̯oʐɪ ˈhɑːɣʷɪ]
a-jokor-em=i nuori haakwi
out-pull-APPL=3SG.An old.woman salmon
"The old woman guts the salmon"

It is clear that nuori "old woman" is the subject here because it is more animate than haakwi "salmon". When the subject acts onto an equally animate object, the object is usually marked with the dative -si:
Totee kaihu kaihusi
[tʰoˈðɛː ˈk̠ɑi̯hɨ ˈk̠ɑi̯hɨsʲɪ]
tote=i kaihu kaihu-si
bite=3SG.An bear bear-DAT
"The bear bites the (other) bear"

This is not true if both are humans, in this case, subject and object are distinguished by word order alone. The canonical word order for sentences like the ones above is Verb-Subject-Object. Now if a lesser animate acts upon a higher animate, or a third person upon a second or first person, one can not rely on word order. Hoavi's animacy hierarchy forbids nouns from appearing before a higher ranking noun in a sentence. This is where inversion comes into play. Consider the following examples:
*Voallui jangi mosalo(si)
[ˈβo̯ɑʔdlɨi̯ ˈjɑŋɪ ˈmosʲɑlosʲɪ]
oall-a=i jangi mosalo(-si)
devour-PST=3SG.An whale fisherman(-DAT)
Intended meaning: "The whale devoured the fisherman"

Mo'oallui mosalo jangun
[moˈʔo̯ɑʔdlɨi̯ ˈmosʲɑlo jɑˈŋɨ̃]
mo-oall-a=i mosalo jangi-n
INV-devour-PST=3SG.An fisherman whale-OBL
"The whale devoured the fisherman" or "The man was devoured by the whale"

The first example is non-grammatical because jangi "whale" is less animate than mosalo "fisherman" and thus can't appear in front of it in the sentence. Instead, the inverse marker mo- ~ me- is prefixed to the verb and jangi takes the oblique case. This also means when lesser animate act upon higher animates, word order switches to Verb-Object-Subject.
The oblique case implies some kind of volition and consequently, it can't be used with inanimates. Instead, the instrumental case is used:
Kestiot masin hongoo!
[ˈkʰɛsʲti̯oʰt mɑˈsʲɪ̃ hoˈŋoː]
kets-t-io=t ma-sin hongo-Wo
fall-TRNS-IRR=3SG.In 2SG-DAT tree-INST
"May a tree fall on you!" ~ "May you be hit by a falling tree!"

There is a second type of inverse prefix, but I'll write on that later on.

[xijlwya]

Why are your verb clitics clitics? Could you provide examples were they act as such? Because, if I'm correct, you examples all attach the clitic to the verb stem and nowhere else.

[MrKrov]

Because, if I'm correct, your examples all attach the clitic to the verb stem and nowhere else.

vit=pai
then=3PL

[Creyeditor]

I've got a question: is your vowel system influenced by Hopi?

[Avo]

Inverse II

There is another inversion prefix pa- and I'm not sure how to gloss it. It is used only when a third person acts upon another third person, and the actor is not known or not mentioned. Sort of like a passivizer.

Sormo'i paniellas paiva
[ˈsʲoɭʔbmoʔɪ̥ pʰɑˈni̯ɛʔdlɑsʲ ˈpʰɑi̯βɑ]
sor-mo=i pa-niell=as paiva
fast-ADV=3SG.An INV2-swallow=FIN sun
"The sun will set soon"

Mun jerie pakaani?
[ˈmɨ̃ŋ ˈjɛʐi̯ɛ pʰɑˈɣɑːnɪ]
mu=n jerie pa=ka-ni
Q=2SG pain INV2-do-PFV
"Are you hurt?"

Why are your verb clitics clitics? Could you provide examples were they act as such? Because, if I'm correct, you examples all attach the clitic to the verb stem and nowhere else.

See above! And the sentence MrKrov is refering to.

I've got a question: is your vowel system influenced by Hopi?

Influenced would be kind of an understatement. I started with /i e a ɨ o/ and then decided to add /ø/ because it's a beautiful vowel and it adds some Finnish flavour to the language.

I'm not having /θ/ in Hoavi by the way. It doesn't fit.

[eldin raigmore]

Anyway! Let's move on to one of the more complex parts of the language: Inversion and animacy

And the rest of that post: Very interesting!

---

There is some suffixation, compunding and reduplication.

Inverse II
...
Sort of like a passivizer.

If I were to say "my infant son was crying until I gave him his passivizer", would that be an example of compunding?

Spoiler:

[Avo]

"my infant son was crying until I gave him his passivizer"

I'll take that as a translation challenge.

That sentence isn't easily translatable into Hoavi. First, there is no past progressive because past and durative can not occur on a verb at the same time. So, one possible way to translate that sentence would be renarrating what happened in chronological order, leaving the verbs unmarked:

Vaavee'ei en nuktuni ii tatnu ot aaturroi iike aare'i.
[βɑːˈβɛːʔɛɪ̯ ʔɛ̃ˈnɨʰktɨnɪ ʔiːˈtʰɑʰtnɨ ʔɑːðɨˈʐoɪ̯ ˈʔiːɣɛ ˈʔɑʐɛʔɪ̥]
vaavee-e=i e-n nuktuni ii tat=nu ot aatur-roi ii=ke aare=i
cry.DUR=3SG.An 1SG-OBL baby then give=1SG that.An calm.down-thing then=and stop.INTR=3SG.An
Lit. "My baby cried then I gave it (the) calming thing and then it stopped (crying)"

Marking the verbs for past tense here would not be appropiate, because past tense suffixes are mostly used to emphasise that the event happened in the past, which is unnecessary here.

The above sentence implies that it's a woman speaking because the word nuktuni "baby" is used in woman's speech only. It is derived from the body part prefix nuk-, a shortened form of nukoi [nɨˈɣoɪ̯] "arm", the verb root tu- "hold, carry, bear" and a nominalising suffix. So the exact translation of nuktuni "that which is carried/hold on the arm".

[eldin raigmore]

Could you point out the compunds, for those of us who have a little trouble seeing them?

[Avo]

If you mean the compuns, pointing them out would ruin the whole joke!

[Click]

I've just skimmed over this thread and my first impression is that Hoavi is a great conlang.

I like the animacy-inversion system very much, especially the second inverse prefix pa-.