Zompist Bboard Again

It seems like there's an open niche for a thread like this at the moment, so here it is: post any bits and pieces, however small, or conlanging work that you've accomplished today (or yesterday, or this week, or whatever).

As for myself, I've been trying to work out the noun paradigms for what I'm tentatively calling "Proto-West Island", since it's the proto-language of the family spoken on... the westernmost island in my conworld. Here's what I've got at the moment:
And here they are after a couple of sound changes for an as-yet unnamed branch:
I kinda don't like the result, mostly related to how the umlaut patterns have worked out. I'd like the root vowel to consistently be the one that umlauts, which I think is gonna involve restructuring the initial paradigm. Expect major revisions!

What are the inflectional categories? I notice that nouns have person markers - does that reflect possession?

So Haleza Grise wrote: ↑Mon Dec 07, 2020 4:06 pm What are the inflectional categories? I notice that nouns have person markers - does that reflect possession?

I should have specified that --yes, it's possession.

What are the sound changes?

bradrn wrote: ↑Mon Dec 07, 2020 5:57 pm What are the sound changes?

Highly unfinished, such that I ended up filling in a lot of the details of the latter paradigm with "yeah, this seems about right". Which is why I didn't post them. But:

Oh, nice idea!

I wrote a bit on noun incorporation in Simbri today:

Noun incorporation


Noun incorporation consists of joining a noun and a verb to form a new compound verb, the noun being incorporated into the verb.

There are two forms of incorporation in Simbri:

Modifying incorporation
The noun stem is added immediately before the verb stem. The noun stem is incorporated alone, without the nominal ending.

Watlekajam yitlome → Wayitlomkajam.

wa-tle-kajam yi-tlom-e → wa-yi-tlom-kajam
1-ANIM-wash 2.OBL-foot-DUAL → 1-2.OBL-foot-wash
'I wash your feet'

The relation in meaning between the nominal element and the verbal element may fall under any of the following categories:

Circumstancial relationship

The incorporated noun is a complement of manner or even a locative:

Ojôntleyin tlamjô.
o-jôn-tleyin tlamjô
UCT-smoke-dry.PL rodent.leg
'They're smoke-drying rodent legs.'

Sotlom ya tloyewa.
so-tlom ya tloy-ew
1.EXCL-son anyway town-IPRF\fight.
'My son is in town, looking for a fight'

Comparison


Either the subject or the object is compared to the incorporated noun:

Min wayimayewor!
min wa-yi-may-awor
and 1-2.OBL-lord-PRF\speak
'But I spoke to you as a lord!'

Otlekâdewar tlemâ seqoji.
o-tle-kâd-ewar tle-mâ se-qoji
4-ANIM-Hippidion-IPRF\call ANIM-mother that-baby
'That baby is calling for a mother (braying) like an Hippidion. '

Relationship of the part to the whole


This type of incorporation is the one used in our example earlier:
Watletlomkajam.

Watletlomkajam.
wa-yi-tlom-kajam
1-2.OBL-foot-wash
'I wash your feet'

The possessor 'you' as in 'your feet' is promoted to direct object, while the original object is incorporated. Some subjects may be promoted this way, promoting the possessor to subject:

Wonyan sotope. → watopwonyan
wonyan so-tope → wa-top-wonyan
suffer 1.EXCL-head → 1-head-suffer
'My head hurts.' → 'I have a headache.'

As we see in the example above, this is typically used with parts of the body, but this form of relationship isn't restricted to those:

Yindi sorontli waamnyaya. → Soron wadoyinnyaya.
yin-di soron-tli wa-am-nyaya → soron wa-do-yin-nyaya
Stone-PL garden-IN 1-MOV-PRF\seize → garden 1-BUILD-stone-take
'I removed the stones in my garden', 'I cleaned my garden'

Incorporation of the agent.


Generally the verb only occurs in the perfect, and it is often best to translate as a passive:

Wakekkakla
wa-kekk-akla
1-snake-PRF\bite
'I was bitten by a snake.'

Although the subject can still be mentioned:

Yiwka wiistanqan tletletobmer an qerwi.
Yiwka wi-is-tanq-an tle-tle-to-bmer an qerw-i
Yiwka 1.OBL-divine-teach-MED ANIM-ANIM-APPL-propitiate that rapist-PL
'The godess Yiwka taught us to sacrifice the rapist.'

TBC...

Ares Land wrote: ↑Tue Dec 08, 2020 10:15 am

I like the look of this phonology a lot, impressionistically it strikes me as somewhere between Nahuatl and an Australian language.

dɮ the phoneme wrote: ↑Tue Dec 08, 2020 10:44 am

Ares Land wrote: ↑Tue Dec 08, 2020 10:15 am

I like the look of this phonology a lot, impressionistically it strikes me as somewhere between Nahuatl and an Australian language.

Thanks a lot!

Hmm, not quite sure how realistic some of these are… for whatever reason, NI seems to be highly constrained in natlangs. (Of course, it may not matter to you depending on how Earth-like you want Simbri to be.) In particular, I know of no language which allows agent incorporation (this seems to be a universal); I also don’t know of any languages which use NI for comparison, though I can’t rule it out completely. The rest of the post looks fine though.

Ares Land wrote: ↑Tue Dec 08, 2020 10:15 am
I wrote a bit on noun incorporation in Simbri today:

[...]

Sotlom ya tloyewa.
so-tlom ya tloy-ew
1.EXCL-son anyway town-IPRF\fight.
'My son is in town, looking for a fight'

Those examples were interesting to read. I'm not sure I understand the one I quoted, though (or any other locative function that results in a meaning like that). Was this inspired by any examples from a natlang which uses noun incorporation in such a way, and if so, where could I find more examples?

Thanks!

bradrn wrote: ↑Tue Dec 08, 2020 5:06 pm Hmm, not quite sure how realistic some of these are… for whatever reason, NI seems to be highly constrained in natlangs. (Of course, it may not matter to you depending on how Earth-like you want Simbri to be.) In particular, I know of no language which allows agent incorporation (this seems to be a universal); I also don’t know of any languages which use NI for comparison, though I can’t rule it out completely. The rest of the post looks fine though.

NI for comparison is fairly common in Nahuatl. A famous example is the name Moteczuma : Motēuczōmah 'He gets angry like a lord.'
Nahuatl also allows agent incorporation in a few restricted cases: ōnicōhuacualōc 'I was bitten by a snake.'
The interesting bit is that it sort of parallels the English 'doctor-recommended' or 'flea-bitten', which are often quoted when discussing incorporation. That said, that particular construction is very uncommon in Nahuatl, and I think it's not very common in Simbri either.

In Simbri you can add an explicit subject (under certain conditions) and I don't know if that's attested anywhere. But the language needs a few weird bits

din wrote: ↑Tue Dec 08, 2020 5:11 pm Those examples were interesting to read. I'm not sure I understand the one I quoted, though (or any other locative function that results in a meaning like that). Was this inspired by any examples from a natlang which uses noun incorporation in such a way, and if so, where could I find more examples?

Thanks!
The example is perhaps a bit too difficult. The verb is in the imperfect, without object marking, which makes it atelic; the general idea is one of fighting with no specific purpose, hence 'pick a fight'. I'm not aware of any natlang that does exactly that.

This section is heavily inspired by Launey and Mackay's Introduction to Classical Nahuatl, which I heartily recommend.
In general, for anyone interested in polysynthetic languages, I recommend taking a look at Nahuatl. It's a very approachable language, compared to Iroquoian or Algonquian languages.

Ares Land wrote: ↑Wed Dec 09, 2020 2:41 am

bradrn wrote: ↑Tue Dec 08, 2020 5:06 pm Hmm, not quite sure how realistic some of these are… for whatever reason, NI seems to be highly constrained in natlangs. (Of course, it may not matter to you depending on how Earth-like you want Simbri to be.) In particular, I know of no language which allows agent incorporation (this seems to be a universal); I also don’t know of any languages which use NI for comparison, though I can’t rule it out completely. The rest of the post looks fine though.

NI for comparison is fairly common in Nahuatl. A famous example is the name Moteczuma : Motēuczōmah 'He gets angry like a lord.'
Nahuatl also allows agent incorporation in a few restricted cases: ōnicōhuacualōc 'I was bitten by a snake.'
The interesting bit is that it sort of parallels the English 'doctor-recommended' or 'flea-bitten', which are often quoted when discussing incorporation. That said, that particular construction is very uncommon in Nahuatl, and I think it's not very common in Simbri either.

Interesting, thanks for the info! Do you have any more details on which ‘restricted cases’ allow agent incorporation in Nahuatl?

In Simbri you can add an explicit subject (under certain conditions) and I don't know if that's attested anywhere. But the language needs a few weird bits

Are you still talking about agent incorporation here? If so, I can’t see how that would work; could you give an example please?

This section is heavily inspired by Launey and Mackay's Introduction to Classical Nahuatl, which I heartily recommend.
In general, for anyone interested in polysynthetic languages, I recommend taking a look at Nahuatl. It's a very approachable language, compared to Iroquoian or Algonquian languages.

As it happens, I was reading an interesting article just today about polysynthesis, which classifies Nahuatl (and Chukchi and Yimas and Bininj Gun-wok) on the one hand and Algonquian (and Greenlandic and Navajo and Abkhaz) on the other as having different types of polysynthesis: Nahuatl is ‘compositional’, relying mostly on verb serialisation and noun incorporation, while Algonquian languages are ‘affixal’, relying mostly on lexical affixes. In general, I’d say that the former type of polysynthesis is perhaps more approachable than the latter; I know that Foley’s Yimas grammar is pretty understandable, for instance (not that I’ve read much of it). Then again, it’s just a hypothesis — as you say, Iroquoian languages aren’t terribly approachable, but get classified as ‘compositional’.

bradrn wrote: ↑Wed Dec 09, 2020 3:42 am

Ares Land wrote: ↑Wed Dec 09, 2020 2:41 am Interesting, thanks for the info! Do you have any more details on which ‘restricted cases’ allow agent incorporation in Nahuatl?

Not really, except that it only occurs in the passive voice. Launey mentions it only ever occurs for a few select verb roots, but does not mention which ones. I suspect there are just a handful of fixed phrases.

In Simbri you can add an explicit subject (under certain conditions) and I don't know if that's attested anywhere. But the language needs a few weird bits

Are you still talking about agent incorporation here? If so, I can’t see how that would work; could you give an example please?

Yes, I am. Continuing from the section above:

Yiwka wiistanqan tletletobmer an qerwi.
Yiwka wi-is-tanq-an tle-tle-to-bmer an qerw-i
Yiwka 1.OBL-divine-teach-MED ANIM-ANIM-APPL-propitiate that rapist-PL
'The godess Yiwka taught us to sacrifice the rapist.'

There are several possible interpretations: this sentence could be interpreted either as a comparison: 'As a goddess, Yiwka taught us...' or as circumstancial 'Yiwka taught divinely...'
Instances of an explicit subject added to a verb with an incorporated agent appear to be a focalization strategy:
'It is Yiwka (and no other godess) that taught us to sacrifice the rapist).

In general, I’d say that the former type of polysynthesis is perhaps more approachable than the latter; I know that Foley’s Yimas grammar is pretty understandable, for instance (not that I’ve read much of it). Then again, it’s just a hypothesis — as you say, Iroquoian languages aren’t terribly approachable, but get classified as ‘compositional’.

I couldn't say... but Algonquian languages do scare me more. But I'm not really that familiar with them.

Iroquoian languages are more difficult, in that you need to take in the active/stative alignment, a solid degree of fusion and allomorphy. (The phonology doesn't help, either. Nothing difficult really, but it's so restricted that everything looks the same), and a fairly weird take on tense and aspect.

By contrast, Nahuatl is nicely agglutinating, the irregularities are there but nothing too difficult (preterits are a bit harder but that's about it). We're even familiar with some of the morphemes without knowing it. (Quetzalcoatl, Acapulco, Mexico, coyote, Popocatepetl).
Grammars sometimes treat it as if it were Spanish, which is bad linguistics, of course, and sweeps some of the complexity under the rug, but I have to admit it helps.

My own theory is that Nahuatl used to be a vehicular language which tends to smooth off the rough edges of language a bit. (I'm told Koine quietly jettisoned the weird bits of Attic, for instance.)

Another factor is that Iroquoian languages are dying off, so basically researchers are trying to cram in any little bit of linguistic knowledge into the reference grammars as fast as possible. They're not really written with the aim of helping people learn the language.
By contrast, many grammars of Nahuatl are designed with the goal of helping historians make sense of the texts.

What have I accomplished in conlanging? Nada. Squat. Haven't been able to do anything since October 17th. Work, COVID, the election, a grad class, & stressing about what my MA thesis or project will be has killed my creative energy for the moment. Looking forward to getting it back.

bradrn wrote: ↑Wed Dec 09, 2020 3:42 am As it happens, I was reading an interesting article just today about polysynthesis,

This is a really, really good paper for conlangers interested in polysynthesis, along with the one by Mithun. It definitely helped me figure out what I wanted to do with Jin.

Ares Land wrote: ↑Wed Dec 09, 2020 9:22 am Yes, I am. Continuing from the section above:

Yiwka wiistanqan tletletobmer an qerwi.
Yiwka wi-is-tanq-an tle-tle-to-bmer an qerw-i
Yiwka 1.OBL-divine-teach-MED ANIM-ANIM-APPL-propitiate that rapist-PL
'The godess Yiwka taught us to sacrifice the rapist.'

There are several possible interpretations: this sentence could be interpreted either as a comparison: 'As a goddess, Yiwka taught us...' or as circumstancial 'Yiwka taught divinely...'
Instances of an explicit subject added to a verb with an incorporated agent appear to be a focalization strategy:
'It is Yiwka (and no other godess) that taught us to sacrifice the rapist).

From what I can tell, this is generally called ‘classificatory NI’ — Mithun’s Type IV. Which, again, seems to be one of those constructions which only works with patients. (Well, Mithun doesn’t say it directly, but all her examples are of patient incorporation.) But given that Simbri already allows agent incorporation, this doesn’t seem such a big problem for it. Also, if I understand the implicational hierarchy right, the presence of Type IV incorporation will also imply the presence of Type III incorporation, which is incorporation for discourse tracking.

Vardelm wrote: ↑Wed Dec 09, 2020 6:00 pm

bradrn wrote: ↑Wed Dec 09, 2020 3:42 am As it happens, I was reading an interesting article just today about polysynthesis,

This is a really, really good paper for conlangers interested in polysynthesis, along with the one by Mithun. It definitely helped me figure out what I wanted to do with Jin.

Yep, I got it from that polysynthesis thread you started a while ago.

Not much, but I have been reconsidering the connection between my conlang Nyango, Old Japanese and Middle Chinese. Also, greetings and time words but I always seem to get stuck on those.

glad to see the return of this thread; always a fine read.


as for me...
Almost finished the Kings List for my t'Bak conscript....a king list with calques - I may or may not come to regret that decision. (particularly since one name means "downy"...and the closest I have to that in the vocab thus far, is "up")

Anybody else doing Lexember? I'm trying to get the lexicon of Keshevan up to 2500 entries. I've been doing a few every day since mid November and I'm getting close.

For Lexember this year, I took an underdeveloped language of mine I wanted to see grow... except I didn't have any derivational schemes for it.
So what I'm doing is creating a new scheme a day and applying it to a maximum number of roots. It's a constraint: there can be no more roots created this month outside the original 55 (among which only the 34 nouns, verbs, and adjectives can be subjected to derivational schemes). So far I have 107 words total.
(It's documented on my website, but in French)

I pretty much finished describing Simbri verbal morphology today.
(It's never finished, really, but at ~10,000 words it's complete enough).

A quick & dirty schematic of verbal structure I did mostly for my own reference: