Zompist Bboard Again

So, I'm gonna repost the stuff I've posted on the CBB on my current a priori conlang Asta, full name ‘astə‘ə ‘iya "this language". I'll be mostly reposting from that board, though with some rearrangement of posts into a more ordered arrangement.

We'll start with the phonology, which many of you will have some idea of but here's a recap.

/p t t͡ʂ c ʔ/ <p t tr ty ‘>
/s x xʷ/ <s x xw>
/m n ɲ ŋ ŋʷ/ <m n ny ŋ ŋw>
/w ɾ j/ <w r y>

/i ə u/ <i u>
/ɛ ɑ/ <e ə a>

There are distributional restrictions on /ə/. Before a palatal consonant /c ɲ j/ is merges with /i/, while it merges with /u/ before labiovelars /xʷ ŋʷ w/. Additionally schwas assimilate to the quality of a following vowel after a glottal stop.

Syllable structure is CV(C), where coda consonants are restricted to /ʔ x n/. However there are multiple phonetic realisations for the fricative and nasal. /x/ is fronted to a coronal fricative /s~sʲ~ʂ/ <s> before coronal /t t͡ʂ c s n ɲ j/, with /j/ being deleted after this /s/, and /n/ assimilates to the place of articulation of a following consonant - this is written as <m> before labial /p m/ and <ŋ> before /ŋ w/, while the cluster /nx/ is simplified to /ŋ/. Additionally /ʔ/ is dispreferred immediately following another consonant - this crops up in both the abstract prefix ‘- and general nominaliser suffix -‘ə appear after a consonant, where metathesis happens, in the former case universally, the latter only with -t (min-‘uxra > mi‘nuxra "our fire", witə w-əmitrat-‘ə > witə wəmitra‘tə "a local person").

Stress is basically word-initial. Some speakers show shift to the post-initial in cases when the initial has a schwa and the post-initial a non-schwa vowel, but this isn't universal.

A final note is that one or two morphological operations which result in palatalisation of consonants.
(there is lexical and speaker variation for the palatalisation of ŋ in diminutives; in verbal inflection is is universally ns)

As a result of the /r w/ sound changes some vowel contractions occur, which are schematised as such
EDIT: the palatalisation process has been slightly changed, and the glottal stop is now transparent for schwa assimilation.

Some notes on distribution and frequency.

In terms of distribution, there are some rather strict restrictions, plus some fairly strong tendencies in the language. The most notable ones are that, mainly as a result of the prevalences of noun and person marking prefixes, the set of permissible word-initial consonants is limited to p, t, ty, s, x, xw, m, n, ny, r, y, w. Furthermore p and xw are restricted to particles (pa‘ "negative", pu "in this manner, like so" and xwe "prohibitive"). The velar nasals (both labialised and not) and the retroflex affricate do not occur word-initially at all.

In terms of clusters that are some interesting restrictions. The set of intervocalic clusters is listed below (in a phonetic transcription):
It will be noted that the clusters involving post-consonantal /ʔ/ are not found in lexical roots, and only occur due to morphological processes, and furthermore the clusters with second member /ŋ/ or /ŋʷ/ are very rare indeed, likely being restricted to certain hapax legomena such as -uŋŋə "to throw" and derivatives. Dashes represent clusters which are prevented from occurring due to morphophonology: geminite glottal stops are simplified, /nx/ becomes a velar nasal, and the glides labialise/palatalise a preceding consonant and drop.

As noted in previous posts, schwa is heavily restricted as a vowel, being somewhat lesser in status relative to the "full vowels" /i ɛ ɑ u/. As noted above it is not found at all before palatal/labialised consonants and clusters including them, and when morphology would produce such an instance the schwa is fronted/rounded to /i/ and /u/ respectively. Furthermore, with a handful of exceptions a root can never contain all schwas: all roots contain at least one full vowel. The exceptions consist of the intensifier particle ‘əŋə, the generic noun tən "thing", and the reflexive pronouns məxən/rəxən, which are formed by adding the 1[super]st[/super] and 2[super]nd[/super] person prefixes to a base -xən (the singular personal pronouns muwə/ruwə are similarly formed as mə-wə/rə-wə, however the labial glide causes the schwa to round as described above). On the other hand most roots do contain at least one schwa: those that don't commonly are either monosyllabic, such as -un "to be similar to X", or have an environment fitting the conditions for schwa fronting or rounding, e.g. xistyi "leg". The constraint is weak though in bisyllabic roots: it is much more prevalent in trisyllabic roots. However, on the flipside the /ɛ/ is fairly strongly restricted to occurring only once in a root, the only exception being ne‘e‘ "ant".

Obviously without a fully fleshed-out language with a large corpus it's difficult to make any conclusive statements as to frequency, however the following observations can be made. tr is far in a way the most uncommon consonant (and clusters involving it are similarly rare), followed at a distance by xw, ŋw. On the flipside, the most common consonants likely include t, ‘, y, w, however I have much less of a sense of that so I can't be sure just yet. Clusters are even more difficult to guage in this respect. On the other hand when it comes to vowels /ɛ/ is the lest common vowel, which the most common vowels overall are probably /a ə/.

So due to the structure of the language I have decided not to start with purely nominal morphology, rather I'm gonna start with the most central part of the morphology, noun-classes, and move onto verbal morphology, and only then move onto what little there is in terms of nominal morphology.

Before I describe noun class, I'll just note that Asta lexical roots are all vowel-initial, but due to the phonotactics of the language no words are, and these noun class prefixes fill that onset slot for noun/verb root. So when you see a noun like, say, nuŋwə "worm", it can be segmented as n-uŋwə IIs-WORM.

Asta has 5 noun classes, which are marked by prefixes, which are found on both nouns and verbs. 4 of these classes may also show plural marking. The class of a given noun is pretty much lexicalised (though sometimes a single root may be found in multiple classes, but with different meanings depending on the marker), but plural marking is consistently inflectional in those noun classes which show plural marking.

The noun class markers are as follows:
For example witə "person" has a plural form yitə "people".

In verbs and demonstratives these same noun class markers appear before the stem, where they serve to agree with the absolutive argument of the clause (in intransitives the subject, transitives the direct object), e.g.

məsumə‘ susti
mə-s-umə‘ s-usti
1s-IIIpl-eye IIIpl-yellow
My eyes are green/yellow

məna‘piyə namə niya
mə-n-a‘pə -yə n-amə n-iya
1-IIs-catch-PRF IIs-fish IIs-PROX
I caught this fish (in my hand)

This agreement takes place even if the absolutive argument is a 1st or 2nd person, where class I is used pretty much invariably.

ruwupupuŋwə
rə-w-VC-upən-wə
2s-Is-PROG-fall-PROG
You are falling

mesəmetyə rinyiyapə‘ə
mex-y-əmet-yə rin-y-iyapə-‘ə
1>2-Ipl-leave-PRF 2pl-Ipl-together-NOM
I abandoned all of you

A final note then on semantics. There appear to be some fairly strong semantic patternings for noun-class assignment. Class I seems to most clearly function as an animate/rational class, however like such classes the world over there some seeming exceptions to this where some apparently inanimate nouns are raised to "animate" status, e.g. waxŋa "sun". Class II seems to function as a sort of "animal" class, but again there are exceptions: netyə "river, stream". Class III could be said to be the "vegetable" class, although "wastebasket" might be more appropriate - it contains names for plants, certainly, but also body parts as well as some landscape features, e.g. xutri "mountain". Class IV is a kid of "abstract/collective" class, featuring mass nouns like ‘uxra "fire", collectives like ‘ex‘ə "crowd, group, mass", and abstract deverbal nouns like ‘ə‘re‘ə "a journey, wander". Finally Class V can be described as a sort of "concrete inanimate/neuter" class, though it can sometimes also be used for countable but non-tangible referents, for example taxŋa "day" (in contrast to ‘axŋa "daytime").

Next: verb stems and transitivity.

OK, verb stems.

Asta verbs are either transitive or intransitive, and there's not a damn thing you can do about it!

Actually that's not strictly true, there are ways to change the valency of a verb - two regular valency-increasing voice infixes, and several irregular stem alternations to produce transitive and intransitive variants of the same root.

The productive voice infixes are the causative -ər- and the applicative -atr-. These are inserted after the last consonant onset (note: ignoring stem-final consonants (see below): -ətreŋ "to run away" > -ətrəreŋ "to let run away, cause to run away"). The causative can have both direct and indirect interpretations, and is also used in special constructions characteristic of the formal register. The applicative can raise either a benefactive or a location to direct object status. In both these cases when the original verb is transitive the noun class marking on the verb tends to still agree with the original direct object, unless that noun is incorporated, in which case the verb will switch to agreeing with the new direct object.

yin mə‘ansəriyintyə‘ə
yin mə-‘-ansi<ər>-intyə-‘ə
EXST 1s-IV-sing<CAUS>-INCEP-NOM
I begin to sing (very formal: the normal expression would be "muwansiyintyə"

‘inatyəruyə wessəsə ye‘i‘(yəx) nessən
‘i-n-atyu<ər>-yə wessəsə ye‘i‘(-yəx) nessən
3-IIs-eat<CAUS>-PRF fisherman young_man.pl(-ADV) fish_catch
The fisherman let the young men eat the catch of fish

(Note the optional use of the adverbial affix to mark the causee).

muwisatriyətixarə wiyan
mə-w-isə<atr>-yə-tixarə wiyan
1s-Is-strum<APPL>-PRF-lute child
I played the lute for the child

In comparison there are no regular and productive valency-decreasing morphological processes in Asta.

Verb stem alternation generally consists of the alternation of stem final consonants/clusters, in particular -t, -x, -n, -ŋ and -nt. This is highly lexicalised an idiosyncratic, however there are a few patterns that can be adduced, notably that a nasal form is generally intransitive while -t/-x are transitive. Further -nt may appear on roots which are normally vowel-final with an inchoative sense, e.g. -ənta "be big" > -əntant "to grow in size" (intr.) However these rules are not unbreakable by any means.

As for the actual formants themselves, they only occasionally appear as their underlying forms; often they are collapsed when word-final, or palatalised/labialised by a following glide. The various forms of these stem suffixes are summarised below:
Note that the -unt suffix affects the preceding vowel, with au, əu and uu collapsing to u, iu becoming yu (triggering platalistion as described above) and eu breaking into ayu.

E.g. -uxwint "to fly": nuxwi‘ "it flies", nuxwintyə "it flew", nuxwuxuyuntə "it is flying", nuxwintuxwi‘ "it flies repeatedly", nuwxintintyə "it begins to fly".

Finally a note on vowel-final stems: these behave much more simply morphophowise. With weak stems (ending in -ə) the ə is lost before another vowel and retained elsewhere, while other vowels insert an epenthetic consonant before another vowel: -r after a, -y after i, -w after u, and e once more breaks up into ay.

e.g. -atə "to swim" > muwatintyə "I begin to swim", -əne "to dream" > muwənayintyə "I begin to dream".

Now we'll move onto morphological TAM. This is more consistent, thankfully, but still relatively complex.

Mood is simple enough: there is only a basic realis-irrealis distinction, with irrealis being marked by a -ŋə- infix before the first consonant of the root/after the first vowel. This basically serves to mark any kind of mood that isn't straight up indicative or a plain interrogative.

mexwəŋənnenənnenutyi
mex-w-<ŋə>ənnen-ROOT-utyi
1>2-Is-<IRR>see-ITER-tomorrow
I might see you tomorrow

tiŋəŋiŋawə wa mintuŋwərayə
t-<ŋə>VC-iŋa-wə wa min-t-uŋwa<ər>-yə
Vs-<IRR>PROG-open-PROG CONJ 1pl-Vs-closed<CAUS>-PRF
If it (the door) is open we will close it

rəsəŋə‘runtyə ‘u pa‘ məsatyuyə
rə-s-<ŋə>ə‘runt-yə ‘u pa‘ mə-s-atyu-yə
2s-IIIpl-<IRR>mix-PRF CCONJ NEG 1s-IIIpl-eay-PRF
Even if you mixed them together I would still not eat them

(Note that while wa and ‘u are both coordinating conjunctions corresponding roughly to "and" and "but", they are used for a much wider range of constructions, including some which we wuld consider to be subordinate clauses).

Aspect is much more developed as a system. The default unmarked aspect is a kind of gnomic/habitual.

naŋwi‘ naŋwuwənta
naŋwi‘ n-aŋwə-wənta
duck IIs-quack-andative
Ducks quack

The perfective is formed from this by a -yə suffix (this causes a consonantal stem to palatalise as described in the previous post).

muwitrityəxati muwuxripu‘
mə-w-itrət-yə-xati mə-wuxrə-ipu‘
1s-Is-break.TRANS-PRF-arm 1-brother-in_law
I broke my brother-in-law's arm

The iceptive -intyə might perhaps be a combination of said perfective suffix plus a productive version of the -nt formative discussed in the post above.

məsuxatrintyə suyə‘an
mə-s-uxə<atr>-intyə suyə‘an
1s-IIIpl-dig<APPL>-INCEP bulb.PL
I began to dig for bulbs

The iterative shows simple full reduplication of the root. This aspect is ambiguous as to whether it is perfective or imperfective: all it conveys is the meaning of repeated action.

riya‘maŋa‘maxra
rə-y-a‘maŋ-ROOT-ra
2s-Ipl-count-ITER-REP
You counted them several times

mintəntayənte tuxwa‘ ‘ummiyəx
min-t-ənte-ROOT tuxwa‘ ‘ummi-yəx
1pl-Vs-cover-ITER corpse mud-ADv
We cover(ed) the bodywith several layers of mud

Finally the progressive is marked by a kind of circumfix, with V(C)C- reduplication before the root plus a -wə suffix following (again see above for how this affects consonant stems).

miyatan yityityiyuntə
mə-yatan y-VC-ityint-wə
1s-father.PL Ipl-PROG-squabble-PROG
My father and his brothers are squabbling

pa‘ mə‘əxməxmiwə mestitatrax‘ə
pa‘ mə-‘-VCC-əxmi-wə mex-t-itax<atr>-‘ə
NEG 1s-IV-PROG-be_able-PROG 1>2-V-send<APPL>-NOM
I can't send it to you right now

Now for person marking.

Person marking in Asta is at once simple and complicated, being as it is a mix of nominative and ergative alignments depending on person, and with some markers being able to appear in several different contexts.

As we have said already the noun class prefixes on verbs agree with their absolutive argument. Correspondingly there is a prefix ‘i- which marks the presence of a 3rd person transitive agent (with no noun class agreement). This never corresponds with the argument marked by the class prefix. This agent may be marked as plural by a suffix -ax, which I am going to make a definitive statement about and say that it may appear on either side of the reported suffix -ra (which otherwise is the final suffix in the verbal complex) depending on both context and the speaker.

‘iyansatriyə we‘ə yiyan
‘i-y-ansi<atr>-yə we‘ə yiyan
3ag-Ipl-sing<APPL>-PRF man child.PL
The man sang for the children

‘inatyatyuwəmi‘xaxra~‘inatyatyuwəmi‘xərax namiyə‘ə
‘i-n-VC-atyu-wə-mi‘xə-ax-ra~‘i-n-VC-atyu-wə-mi‘xə-ra-x n-amiyə-‘ə
3ag-IIs-PROG-eat-PROG-flesh-PL-REP~3ag-IIs-PROG-eat-PROG-flesh-REP-PL IIs-125-NOM
They were eating all kinds of flesh

Things become more complicated when speech-act-participants are brought in. There is only a single set of prefixes for marking SAPs:
When these appear alone before a noun class prefix, they mark the nominative subject. In intransitives there is no noun-class marking and the schwa of the prefix is deleted (note that this is a change from before, where there was still noun class marking.)

mexuwintyə
m-exu-intyə
1sg-sit-INCEP
I sit down

ri‘nəriyə ‘axxwə‘ə
rin-‘-əri-yə ‘-axxwə-‘ə
2pl-IV-ask_for-PRF IV-quite-NOM
You (pl.) asked for quiet

However, when a 3rd person agent prefix is added in transitives, this prefix will mark a transitive object/recipient (note that while in monotransitives the noun class marking is again dropped, in ditransitives the noun class marking may still mark theme).

‘iriŋwarax riyatan
‘i-r-iŋwa-ax rə-yatan
3ag-2sg-respect-PL 2sg-son.PL
Your sons respect you

(Note here that the -ax affix does not delete or merge with the preceding stem-vowel, but instead there is consonantal epenthesis since it is appended directly onto the vowel-final stem)

‘imətintyatrenyəra təŋarə‘ə tiya
‘i-mə-t-intyen<atr>-yə-ra təŋarə‘ə t-iya
3ag-1s-Vs-se_up<APPL>-PRF-REP mancala Vs-PROX
It seems he set up this game of mancala for me

Finally there are two fused affixes for entirely local actions: yur(ə)- for 2nd person acting on 1st and mex- for first person acting on second.

yurətresə tuxəntatreyənyəx
yur-ətrex-yə tuxəntatreyən-yəx
2>1-chase-PRF burial_ground-ADV
You chased me through the burial ground

pa‘ mexwiŋəmpəximpərəx muwi‘yə
pa‘ mex-w-<ŋə>impəx-ROOT<ər> mə-wi‘yə
NEG 1>2-Is-<IRR>have_sex-ITER<CAUS> 1s-sister
I will not allow you to keep shagging my sister

Next in the verbal template: noun incorporation.

Asta has at least types I-III noun incorporation according to Mithun's schema, and maybe type-IV if I can even be bothered with it (so probably not). These incorporated nouns follow the verbal stem (after it's been inflected for aspect), retaining their class prefixes, and with corresponding (singular) noun-class agreement being lost as the verb is detransitivised.

Type I noun incorporation is of course mainly lexical and just backgrounds things.

minyessessuwə nyesin
mə-ny-VCC-exsə-wə nyesin
1s-IIpl-PROG-stab-PROG puppy.PL
I'm killing puppies

messənesin
m-essə-nesin
1s-stab-puppy
I kill puppies

However, as a side note, there is one suppletive incorporated noun: -mi‘xə "flesh" (not related to the independant nantyə).

nuŋwiya nantyə niya
n-uŋwiya nantyə n-iya
IIs-umami meat IIs-PROX
This meat is delicious

miŋə‘ti‘tuwə miŋwəmi‘xə‘ə
m-<ŋə>VCC-i‘tə-wə m-iŋwə-mi‘xə-‘ə
1s-<IRR>PROG-like-PROG 1s-cook-meat-NOM
I would like to cook some meat

Type II incorporation comes from the re-addition of a new argument. With no other marking this is usually interpreted as the possessor of the incorporated noun, however the applicative can be used to raise other roles.

yiyan miyessiyənesin
yiyan mə-y-essə-yə-nesin
child.PL 1s-Ipl-stab-PRF-puppy
I killed the children's puppy/puppies

miyessatriyənesin yiyan
mə-y-essə<atr>-yə-nesin yiyan
1s-Ipl-stab<APPL>-PRF-puppy children
I killed puppies for the children

muwuŋŋiyinyen muwatan wəstriyu‘ə
mə-w-uŋŋə-yə-nyen mə-watan w-əstriyu-‘ə
1s-Is-throw-PRF-water 1s-father 1s-deep-NOM
I cried for my father buried in his grave

For type III I will just reproduce an up to date version of a short narrative I posted originally on the CBB.

maya yin natyə wa taŋatrara ‘iwə. ‘inityatrəra netyə. ‘inya‘piyəra nyemenyə nyammə nyantyə ‘isatiyəx. ‘inyuntuwəntayənaməra ‘inya‘pə‘ən taŋayəx wa natyuyənaməra.
maya yin natyə wa taŋa<atr>-ra ‘i-wə. ‘i-n-ityə<atr>-ra netyə. ‘i-ny-a‘pə-yə-ra nyemenyə ny-ammə ny-antyə ‘i-s-ati-yəx. ‘i-ny-untə-wənta-yə-namə-ra ‘i-ny-a‘pə-‘ə-n t-aŋa-yəx wa n-atyu-yə-namə-ra.
DECL EXST otter CONJ house<APPL>-REP 3-SG. 3-IIs-walk.PRF<APPL>-REP river. 3-IIpl-catch-PRF-REP minnow IIpl-five IIpl-three 3-IIIpl-hand-ADV. 3-IIpl-carry-andative-PRF-fish-REP 3-IIpl-catch-NOM-NOMP Vs-house-ADV CONJ IIs-eat-PRF-fish-REP
There was an otter that lived in a house. He went to the river. He caught 8 minnows with his hands. He carried his catch home and ate it.

Finally one could argue that the following counts as a borderline instance of type IV NI, however classificatory incorporation will probably not be common in inflectional contexts, however it'll probably be fairly common inderivation, e.g. xəxxaxi‘səsə "hawthorn", from x-əxxa-xi‘sə-sə IIIs-spiky-tree-AGT.

mənaŋətyumi‘xə‘uŋwə nesin
mə-n-<ŋə>atyu-mi‘xə-‘uŋwə nesin
1s-IIs-<IRR>eat-meat-easily puppy
I would happily eat puppy flesh

Adverbial suffixes now. These are a somewhat open-ended class, so I'll only discuss a few of them, and likely in the future I'll create even more.

Firstly there is the distributive -pan, which normally is a true distributive, but can also have some interesting semantic effects with certain verbs.

‘ina‘wayərax ne‘mu nanə
‘i-n-a‘wa-yə-ra-ax ne‘mu n-anə
3ag-IIs-hunt-PRF-REP-PLUR pig IIs-one
They killed a pig (together)

‘ina‘wayəpanrax ne‘mu nanə
‘i-n-a‘wa-yə-pan-ra-ax ne‘mu n-anə
3ag-IIs-hunt-PRF-DIST-REP-PLUR pig IIs-one
They each killed one pig

miyitasəpan yitə‘intə
mə-y-itax-yə-pan yitə‘intə
1s-Ipl-send-PRF-DIST messenger.PL
I sent messengers to every corner of the land

The excessive -tyuxə carries a general meaning of "too much", "beyond the point of reasonableness".

yurənətrarətratyuxə nantyə
yurə-n-ətra-ROOT-tyuxə nantyə
2>1-IIs-give-ITER-excess meat
You kept giving us too much meat

The frustrative -ra‘a encodes an event which has been attempted but failed to take place.

wuyuwuyura‘a
w-uyu-ROOT-ra‘a
Is-sleep-ITER-FRUS
He tried (and failed) to get to sleep several times

The frustrative can also be used as a kind of narrow-scope verbal negation (other kinds of negation being handles by the particle pa‘), like so.

pa‘ witə wintyentyə
pa‘ witə w-intyent-yə
NEG person Is-stand.up-PRF
No-one stood up

yin yitə yintyentyəra‘a‘ə
yin yitə y-intyent-yə-ra‘a-‘ə
EXST person.PL Ipl-stand.up-PRF-FRUS-NOM
Some people did not stand up

The andative -wənta encodes the notion of "going and doing X" or "going along doing X".

wansiwəntasami
w-ansi-wənta-sami
Is-sing-AND-sometimes
Sometimes he would go along singing

The suffix -nyixə is the equivalent of the modal auxilliary "to be able to".

pa‘ məxatyatyuwinyixə
pa‘ mə-x-VC-atyu-wə-nyixə
NEG 1s-IIIs-PROG-eat-PROG-can
I can't eat it

rəxatyatyuwəra‘anyixə
rə-x-VC-atyu-wə-ra‘a-nyixə
2s-IIIs-PROG-eat-PROG-FRUS-can
You don't have to eat it

There are other suffixes as well, many with temporal meanings like "sometimes" or "tomorrow".

There is one more verbal suffix I haven't properly covered, which follows the other suffixes (bar potentially the 3rd person agent plural), and that is the reported -ra. This is essentially a marker of reported evidentiality, that is, events which have been related to the speaker from other sources.

ruwityatrəra wuŋwityə
rə-w-ityə<atr>-ra wuŋwityə
2s-Is-walk<APPL>-REP moon
I heard you went to the moon

Some bits of nominal morphology. Basically the only bits of nominal morphology (outside of the noun-class prefixes outlined above) are possession and the adverbial case marker.

Possession is rather simple: possessive affixes attach to the noun, with any nominal (as opposed to pronominal) possessors following the possessed noun in addition to this. There is no appreciable alienable-inalienable distinction. For speech act participants the set of markers is identical to the verbal subject/object markers (though of course always in the full forms since the possessed nouns retain their noun class prefix), while the 3rd person forms are identical to the 3rd person agent prefix/circumfix. These are repeated below as a reminder:
muwaman
mə-waman
1sg-mother/daughter/aunt
my mother/daughter/aunt

(note that Asta kinship terms merge alternate generations, and moreover follow an Iroquoian system: I've previously only given one possible translation when using these terms, but I thought I should make this clear).

‘isuxwə xi‘sə sinyu‘ə
‘i-suxwə xi‘sə s-inyu-‘ə
3-root.PL tree IIIpl-long-NOM
The tree's long roots

The other bit of nominal morphology is the adverbial suffix -yəx, which is used to mark any kind of oblique case role other than genitive, including locatives, instrumentals and so on (though typically not benefactives or comitatives, since the former tend to be marked using the applicative while the latter use verbal constructions). It can also sometimes be used with certain kinds of agent-like inanimates. This suffix precedes the 3rd plural possessor suffix -ax.

wiŋu‘sə ‘ityisisuwə tyutri tiŋu‘siyəx
wiŋu‘sə ‘i-ty-VC-isə-wə tyutri tiŋu‘sə-yəx
painter 3ag-IVpl-PROG-brush-PROG rock.PL paintbrush-ADV
The painter is brushing/caressing the rock with their paintbrush

wupinyə wessəsə netyiyəx
w-upən-yə wessəsə netyə-yəx
Is-fall-PRF fisherman river-ADV
The fisherman fell into the river

minnyintyəmi‘xə nupə‘iyəx
m-innyənt-yə-mi‘xə nupə‘ə–yəx
1-chill-PRF-flesh wind-ADV
The wind chilled me to the bone

OK, time for pronominal forms!

There are three classes of pronouns morphologically speaking in Asta: personal/reflexive, demonstrative and interrogative.

Personal and reflexive pronouns are built on the same principle: a common element with personal/noun class prefixes being prefixes on. This common element is -wə for singular personal pronouns, -yə for plural personal pronouns and -xən for reflexives. With noun class prefixes in order to create the pronominal forms a schwa is epenthesised, with of course vowel colouring before the glides, however 3rd person pronouns other than the reflexive are of course very rare, even more so that personal pronouns generally, due to demonstratives being more commonly used. There is some variation in the verbal inflection used with reflexives: some continue to inflect the verb transitively, others intransitively, and of those that inflect transitively, there is uncertainty as to whether with 1st and 2nd the prefix should be doubled or whether a class I prefix (agreeing in number) should fill the "object slot".

pa‘ ‘imessəriyə wuwə nəmawə
pa‘ ‘i-m-essə<ər>-yə w-wə nəmawə
NEG 3.ERG-1s-hit<CAUS>-PRF Is-SG ox
He wouldn't let me kill the ox

mə‘aŋəstiyə rinyə ‘intə
mə-‘-<ŋə>astə-yə rin-yə ‘intə
1s-IV-<IRR>speak-PRF 2p-PL message
I will tell you the news

ministyatrux/minyistyatrux/mimministyatrux miŋən
min-istyuŋ<atr>/min-y-istyuŋ<atr>/min-min-istyuŋ<atr> min-xən
1p-recite<APPL>~1p-Ip-recite<APPL>~1p-1p-recite<APPL> 1p-REFL
We recite poetry to ourselves

Demonstrative pronouns encode a simple two-way distance contrast of proximal -iya vs. distal -əra and take noun class prefixes to agree with their referant. They can be used both adnominally and pronominally, with adnominal demonstratives following their heads.

mə‘iŋə‘ti‘tuwə rətiŋu‘tə tiya
mə-‘-<ŋə>VCC-i‘tə-wə rə-t-iŋut-‘ə t-iya
1s-IV-<IRR>PROG-like-PROG 2s-Vs-paint-NOM Vs-PROX
I want you to paint this

‘inya‘wa‘wawə nyutura nyəra
‘i-ny-VCC-a‘wa-wə nyutura ny-əra
3.ERG-IIp-PROG-hunt-PROG swamp_rat.PL IIp-DIST
They are hunting those swamp rats

Interrogative pronouns are much simpler, in that they do not inflect for noun class at all, aside from an animate-inanimate contrast wa‘ vs. ta‘ respectively. These appear utterance-initially, and serve to mark the sentence as specifically interrogative, such that one could almost count them as part of the list of sentence-initial particles (see below).

wa‘ ‘iyaxi‘nyəriyə yiyə?
wa‘ ‘i-y-axi‘nyə<ər>-yə y-yə
ANI.INT 3.ERG-Ip-angry<CAUS>-PRF Ip-PL
Why are they angry?/Who has made them angry?

ta‘ rə‘eŋisə?
ta‘ rə-‘-eŋix-yə?
INA.INT 2s-IV-spill-PRF
What have you spilt?

OK, time to talk about nominalisations.

Basically there are four nominalising suffixes in Asta: -‘ə, -sə, -n(ə) and -yən. I will discuss each of these in turn, explaining their semantics more than their syntactic functions (which are many and varied and would be a post unto themselves).

The -‘ə is the most general: it is used to form a kind of abstract noun, which can then be applied in many different syntactic situations. It shows some allomorphy: when it touches -t it metathesises, and is simply realised as -ə after -nt.

‘uxə‘ə
‘-uxə-‘ə
IV-dig-NOM
digging

wə‘rə‘rewə ruwuxrə wu‘ya‘ə
w-VCC-ə‘re-wə rə-wuxrə w-u‘ya-‘ə
Is-PROG-wander-PROG 2sg-grandad Is-old-NOM
Your old grandad's gone walkabout

The other three suffixes are more lexicalised. -sə tends to form agents or instruments depending on the accompanying noun class.

wa‘wasə
w-a‘wa-sə
Is-hunt-AGT
hunter

turu‘sə
t-urut-sə
Vs-cut-AGT
blade, scythe

Similarly -n(ə) forms patients or results, but occasionally also experiencers (the vowel only appears when following a consonant).

xəpiŋun
x-əpiŋu-n
IIIs-flip-PAT
pancake

tuxən
t-uxə-n
Vs-dig-PAT
hole in the ground

Finally the -yən nominaliser can mark either place or time depending on the noun class (places take either class III or class V, times typically class IV).

təmayən
t-əma-yən
Vs-make-PLACE
workshop

‘uwintyən
‘-uwənt-yən
IV-bear.fruit-PLACE
fruiting season

xi‘siyən
xi‘sə-yən
tree-PLACE
forest, wood

Any of the final three nominalisers can concatenate with -‘ə to mark the role of arguments in a relative clause when ambiguity would otherwise arise.

pa‘ ‘iya‘ma‘maŋwinyixə yitə yimpəx‘ən
pa‘ ‘i-y-VCC-a‘maŋ-wə-nyixə yitə y-impəŋ-‘ə-n
NEG 3ag-Ipl-PROG-count-PROG-be.able person.PL Ipl-have.sex-NOM-PAT
He can't count how many people he's shagged

yin witə mətux‘əsə
yin witə m-ətux-‘ə-sə
EXST person 1sg-follow-NOM-AGT
Someone is following me

OK, so in order to make a post on this language before the new year, I'm going to discuss the core points of Asta syntax.

Asta, like many polysynthetic languages, is relatively non-configurational. There is a tendency towards verb-initial word-order, however the only firm placement rule is that particles (on which more later) are fixed, mainly to clause-initial position, but in two cases instead between two constituents and in one case to clause-final position. After clause-initial particles, nominal constituents may precede the verb, generally as a function of topicality (though the standard notions of "topic" and "focus" are somewhat vague and hazy, and to describe Asta as "topic-prominent" would be to imply a similarity of this system to that of languages like Japanese, when it is significantly less fixed than those versions of it). On the other hand, particularly long or heavy phrases are nearly always placed after the main verb. Clauses with mutiple free noun phrases are uncommon, but when they do occur there is generally only one at most before the verb - SOV and OSV orderings are generally dispreferred, however there is no absolute restriction to either subject-object or object-subject order, i.e. SVO, OVS, VSO and VOS are all grammatical. Similar points can be made about adverbial arguments marked with the -yəx, though these tend to be classified as "heavy" and are thus more likely to find themselves at the end of the clause.

The above only really applies to clauses headed by a finite verb, however. In non-finite clauses (which in Asta entirely consist of nominalisations, see above), there are more strict constraints. Notably in a nominalised clause the nominalised verb always appears first, followed by its nominal arguments. Additionally, having multiple overt core nominal arguments in a nominalised clause is almost unheard of, meaning that no concrete statements can or need be made about the relative ordering of arguments in such clauses. Such nominalised clauses also tend to drift towards the end of the finite clause.

To illustrate these ideas I've given the following short dialogue, which shows these in action.

"maya muwuxrə wepepuwəxunimə."
maya mə-wuxrə w-VC-epə-wə-xunimə
DECL 1s-granddad 1s-PROG-empty-PROG-belly
"OK so my granddad is hungry"

"məxə‘ratreyə xi‘siyən."
mə-x-ə‘re<atr>-yə xi‘siyən
1s-IIIs-wander<APPL>-PRF forest
"I'll go foraging in the woods"

"nantyə ‘inəriyəri wa ra‘wayəmi‘xəŋa‘tə."
nantyə ‘i-n-əri-ROOT wa r-a‘wa-yə-mi‘xə-ŋa‘tə
meat 3.ERG-IIp-ask_for-ITER CONJ 2s-hunt-PRF-meat-ought
"He keeps asking for meat, so you ought to get some"

"ta‘ nantyiyəx ‘ini‘tə?"
ta‘ nantyə-yəx ‘i-n-i‘tə
INA.INT meat-ADV 3.ERG-IIs-like
"What kind of meat would he like?"

"nutura rənexsiyə. pa‘ runəriyə rəxən ‘a‘wa‘iyəx ne‘mu!"
nutura rə-n-exsə-yə. pa‘ r-unə<ər>-yə rə-xən ‘-a‘wa-‘ə-yəx ne‘mu
swamp_rat 2s-IIs-stab-PRF | NEG 2s-die<CAUS>-PRF 2s-REFL IV-hunt-NOM-ADV boar
"Get (him) a swamp rat. Don't get yourself killed trying to get a wild boar!"

"pu miriyə."
pu m-irə-yə
thus 1s-act-PRF
"I'll do that."

Frislander wrote: ↑Mon Dec 31, 2018 12:07 pm Asta, like many polysynthetic languages, is relatively non-configurational. There is a tendency towards verb-initial word-order, however the only firm placement rule is that particles (on which more later) are fixed, mainly to clause-initial position, but in two cases instead between two constituents and in one case to clause-final position. After clause-initial particles, nominal constituents may precede the verb, generally as a function of topicality (though the standard notions of "topic" and "focus" are somewhat vague and hazy, and to describe Asta as "topic-prominent" would be to imply a similarity of this system to that of languages like Japanese, when it is significantly less fixed than those versions of it). On the other hand, particularly long or heavy phrases are nearly always placed after the main verb. Clauses with mutiple free noun phrases are uncommon, but when they do occur there is generally only one at most before the verb - SOV and OSV orderings are generally dispreferred, however there is no absolute restriction to either subject-object or object-subject order, i.e. SVO, OVS, VSO and VOS are all grammatical. Similar points can be made about adverbial arguments marked with the -yəx, though these tend to be classified as "heavy" and are thus more likely to find themselves at the end of the clause.

One thing I realised when I was thinking about this for Sint is that a lot of verb initial languages seem to have surprisingly similar pragmatics and marking for pre-verbal fronting of argument(s). I found it interesting since I used to find it quite confusing that verb-initial languages seemed to use the same process (fronting) for both topic and focus marking, but the more I read the more I realised that while the motivation of moving the noun to a prominent location was the same, the two functions of that location tend to be differentiated by intonation at least and often syntactically or morphologically as well.

Some language were different, but I found a lot of summaries from Mayan to Polynesian languages all of which went something like this:

1. Both topic switches and contrastive focus can be marked by pre-verbal fronting

2. BUT with a difference in intonation and/or resumptive pronoun use and/or agreement and/or voice

3. Fronted topics:
(i) tend not to be under the same intonational contour as the following clause (e.g. "John, he went")
(ii) tend to have in the clause itself whatever normally happens when the topical argument is not represented by a full NP (e.g. resumptive pronouns, anaphoric agreement, ...)
(iii) The topic itself may be marked by a particle or special morphology, but often there is no special morphological marking inside the clause itself

-> The fronted topic is not a constituent of the clause, it's outside and just referenced in the clause

4. Fronted contrastive focus:
(i) Is under the same intonational contour
(ii) may trigger special extraction morphology in the clause (e.g. Mayan actor focus marking, anti-agreement effects, ...) as well as on the fronted noun or between the noun and the verb

-> The fronted focus is a constituent of the clause, and since fronting is especially favoured for very marked focus such as transitive actor focus, often other changes highlight the marked state of affairs

Just thought this might be of interest.

That's greatly of interest, thank you, and somewhat reassuring. Although now I'm curious about the special morphology (particularly on the noun itself - you mean, like, case marking that only shows up on fronted arguments?) and the "other changes"...

That's good information, and basically reflects my understanding of how the topic and focus are typically differentiated. And I will be honest that I've not actually given much thought to Asta intonation, so I should definitely cover the differences if/when I get round to it.

As for the morphology though I'm not entirely sure how I might go about working that in. As you can see, the language is really finely balanced when it comes to transitivity and the like, and its agreement system is also fairly rigid. Because Asta verbs are so heavily specified for transitivity I don't think I can necessarily afford to drop marking for all focused arguments. I guess I could perhaps make the ‘i- prefix for transitive subjects drop when the agent is focused but that's probably about as much as I feel I can allow without completely restructuring the whole thing, which I don't really want to do because that could eventually lead to me abandoning the language, because that's how it tends to work with me.

OK, before I do clause-conjunction, I guess I'll bring over the post on particles I made on the CBB, because they're kind of important.

As discussed, Asta is fairly non-configurational, but there is a preference for verb-initial order, and there are a number of particles which find themselves in clause-initial position. The first of these is pa‘ the sentential negative, which as mentioned is a broad-scope negator.

pa‘ mintənnenyə tən
pa‘ min-t-ənnen-yə tən
NEG 1p-Vs-see-PRF thing
None of us saw anything

The next is yin, the existential.

yin namə ‘axwəsiyəx
yin namə ‘-axwə-sə-yəx
EXST fish IV-flow-AGT-ADV
There is/was a fish in the water

An interesting use of this is found in formal language, where it is used with a nominalised verb in order to signify remoteness - sister languages to Asta grammaticalised this as an indirect evidential instead of -ra.

yin mi‘tə‘ə rəmen‘ə tərayəx
yin m-i‘tə-‘ə r-əmen-‘ə t-əra-yəx
EXST 1s-want-NOM 2s-stay-NOM Vs-DIST-ADV
I would like you to stay there

A fairly important particle is maya, which serves both as a call to attention and also partly as a marker of a shift in the topic of discussion.

maya weŋisə‘aŋu muwaman
maya w-eŋix-yə-‘aŋu mə-waman
DECL Is-spill-PRF-blood 1s-daughter
So my daughter has had her [first] period

Then there's the interrogatives; xə the polar interrogative and na the tag question (the only such particle which follows the utterance rather than preceding it).

xə yurastyəriyə?
xə yur-astyərə-yə
POL 2>1-protect-PRF
Did/will you protect me?

rinumpan riŋən yasnyə‘ə na?
rin-un-pan rin-xən y-asnyə-‘ə na
2pl-similar-DIST 2pl-REFL Ipl-two-NOM TAG
You two are very similar to each other aren't you?

It has also been mentioned that sentential negation is achieved using pa‘, and that narrower-scope verbal forms use -ra‘a, however there is a third type of negative in Asta, that being the prohibitive xwe (note that the verb still takes second person marking; this is true of imperatives as well).

xwe yurixmityə
xwe yur-ixmət-yə
PROH 2>1-show-PRF
Don't show it to me!

There is a counterfactual particle nyi‘, which occurs in conjunction with the irrealis infix, specifically for speculations about events the speaker knew did not take place. This can sometimes function as a kind of optative "if only...".

nyi‘ ruwiŋityatrə wa pa‘ wuniyə
nyi‘ rə-w-<ŋə>ityə<atr> wa pa‘ w-unə-yə
COUN 2s-Is-<IRR>walk.PRF<APPL> CONJ NEG
If you'd gone to hem they wouldn't have died

nyi‘ məmenyə
nyi‘ m-əmen-yə
COUN 1s-stay-PRF
If only I'd stayed

There is a particle pu, which is a general illocutionary force indicating that the event happened "like so", or that a state appeared "like that", and is often used for the last clause in a narrative to indicate a kind of "closing sentence" in the same manner as maya might indicate an "opening line".

pu minyessiyə nyiya nyənta
pu mə-ny-essə-yə ny-iya ny-ənta
THUS 1s-IIpl-spear-PRF IIpl-PROX IIpl-big
...and that's how I speared this big [catch of fish]

There is an "intensifier", corresponding to what in many languages is identical to a reflexive in form, but has a non-reflexive meaning e.g. "by oneself".

‘əŋə rətiŋutyə taŋa təra
‘əŋə rə-t-iŋut-yə taŋa t-əra
INTENS 2s-Vs-paint-PRF house V-DIST
You pointed that house yourself

Finally, and this is a new addition, there is a quotative particle xwən which marks directly quoted speech.

xwən ‘inyuŋŋəru‘ nyaŋuyi‘ nyanyi‘iyəx
xwən ‘i-ny-uŋŋə-ru‘ nyaŋuyi‘ ny-anyi-‘ə-yəx
QUOT 3erg-IIp-throw-into_fire duckling.PL IIp-alive-NOM-ADV
He said "I chuck live ducklings into fires"

I will cover wa and ‘u the coordinating and contrasting conjunctions respectively in my discussion of clause combining.

OK, let's discuss relative clauses at last.

So relative clauses are formed from nominalisations. These are basically formed from the generic nominaliser -‘ə. In the case of intransitive subjects and also transitive core arguments where the other argument is clear from context this is all that is required.

wəmux we‘xwə‘ə
wəmux w-e‘xwə-‘ə
woman Is-young-NOM
the young woman

wa‘wasə ma‘piyə‘ə
wa‘wasə m-a‘pə-yə-‘ə
hunter 1s-catch-PRF-NOM
the hunter that saved me

we‘i‘ wəra ruwimpəsə‘ə
we‘i‘ w-əra rə-w-impəx-yə-‘ə
young_man Is-DIST 2s-Is-have_sex-PRF-NOM
that young man you shagged

However in other circumstances with transitive verbs some disambiguation may be required. In these cases the agent nominaliser -sə or the patient nominaliser -n is appended after the -‘ə suffix as appropriate. Additionally, when the agent nominaliser is used the ergative prefix is dispreferred.

nuŋwe ‘inatyu‘ən nə‘pex
nuŋwe ‘i-n-atyu-‘ə-n nə‘pex
worm 3erg-IIs-eat-NOM-PAT lizard
worms that lizards eat

wixpəssə watyu‘əsə witə
wixpəssə w-atyu-‘ə-sə witə
clown Is-eat-NOM-AGT person
the man-eating clown

Benefactors and locations may be permitted to be the head of a relative clause using the applicative.

taŋa muwəmatretyə‘ə muwi‘yə
taŋa mə-w-əmet<atr>-yə-‘ə mə-wi‘yə
house 1s-Is-leave<APPL>-PRF-NOM 1s-sister
the house I left my sister in

Finally there are some syntactic differences between restrictive and non-restrictive relative clauses. Both restrictive and non-restrictive relatives may behave as described above, however non-restrictive relative clauses may additionally be expressed instead by paratactic clauses, particularly in regular speech.

ruwaman waxi‘nyəmatrə ‘u ‘iwa‘əxa‘əx muwusərə
rə-waman w-axi‘nyə-matrə ‘u ‘i-w-ROOT-a‘əx mə-wuxrə-DIM
2s-mother Is-angry-difficult CONT 3erg-I-ITER-beat 1s-brother-DIM
My mother, who is slow to anger, beat up my brother

OK, let's just go over other kinds of clausal conjunction quickly, caus there isn't really that much too it.

Asta is one of those languages that doesn't go in for complex subordination strategies or a plethora of conjunctions. In fact there are only two true conjunctions in the entire language, and they are both coordinating - the "conjunctive" wa and the "contrastive" ‘u. These share the load of clause-coordination with nominalisation. As discussed before the semantics of the two are simply that the former marks when the coordinated events are corresponding in some way and the latter when they contrast.

yuyuyuwə wa minyu‘miyə
y-VC-uyu-wə wa min-y-u‘mə-yə
Ip-PROG-sleep-PROG CONJ 1p-Ip-raid-PRF
We will attack them while they sleep

yu‘tə‘ə ‘i‘ənnənneŋwəra’ax ‘ənnun wa yu‘piyuwastrəsə ‘irinyəx
y-wu‘tə‘ə ‘i-‘-VCC-ənnen-wə-ra’a-ax ‘ənnun wa y-u‘pi-yə-wastrəsə ‘irən-yəx
PL-sage 3.ERG-IV-PROG-see-PROG-FRUS-PL land CONJ Ip-inspect-PRF-star habit-ADV
The sages could not see land so they consulted the stars as is customary

wuniyə ‘u nyi‘ ri‘iŋityatrə ‘əra
w-unə-yə ‘u nyi‘ rə-‘-<ŋə>ityə<atr> ‘-əra
Is-die-PRF CONTR COUN 2s-IV-<IRR>come.PRF<APPL> IV-DIST
He would have died even if you had been there

Causative senses can also be conveyed by this means, though nominalisation is also used for this.

minəmataŋa səturisiyəx wa pa‘ minunyə
min-əma-taŋa y-xəturisə-yəx wa pa‘ min-unyə
1p-make-house PL-ironwood-ADV CONJ NEG 1p-be_afraid
We are not worried because we build our houses out of ironwood

‘i‘axəntaxəntə ‘imuntəriyuwənta wəpu‘siyəx
‘i-‘-ROOT-axənt-‘ə ‘i-m-untə<ər>-yə-wənta wəpu‘sə-yəx
3.POSS-IV-ITER-cough-NOM 3.ERG-1s-carry<CAUS>-PRF-AND healer-ADV
Because he kept coughing I had to go fetch the doctor

Does it ever make a difference which order they go in? The aspect-marking in the first example is maybe enough to make it reversible, but the others seem like maybe they could only go in the order you give them in. (E.g., in all three that give reasons/causes, the reason or cause is given first.)

akam chinjir wrote: ↑Sat Mar 02, 2019 12:56 pm Does it ever make a difference which order they go in? The aspect-marking in the first example is maybe enough to make it reversible, but the others seem like maybe they could only go in the order you give them in. (E.g., in all three that give reasons/causes, the reason or cause is given first.)

I think the third one is the only one where ordering could have an effect, and in the case of the fifth one it's a nominalisation and thus the nominalised clause behaves like nouns do with respect to position.