Polysynthetic Conlang

The best topics from Languages & Linguistics, kept on a permanent basis.
Post Reply
Aurora Rossa

Polysynthetic Conlang

Post by Aurora Rossa »

http://www.angelfire.com/mo3/terrapvlch ... g/cl1.html

What do you thing of the conlang I made? It is polysynthetic and is the only conlang I have ever seen to hav clicks in it. It also has rarely seen implosive consonants.

Mne//amaa ga@'o'amaifsqu s@a!omie.
Mne//a-ma-a ga-@'o'a-ma-i-fsqu s@a!o-mi-e.
write-1st.-past need-1st.-gen.-hand sleep-3rd.-fut.
I wrote so much that my hand needs to rest.

1st.=1st. person
3rd.=3rd. person
gen.=genitive case

I don't think anyone has made a conlang this cool.

User avatar
Jaaaaaa
Lebom
Lebom
Posts: 165
Joined: Fri Sep 13, 2002 6:01 pm
Location: Illinois, USA
Contact:

Post by Jaaaaaa »

It looks interesting. To be honest, I think it's too simple, but it's still cool.

User avatar
vohpenonomae
N'guny
N'guny
Posts: 91
Joined: Sat Nov 02, 2002 4:23 am

Re: Polysynthetic Conlang

Post by vohpenonomae »

I'll have to take a look. Something that seems odd on a cursory glance is that the language seems not to use object agreement markers (though that may have an explanation deeper in the grammar).
"On that island lies the flesh and bone of the Great Charging Bear, for as long as the grass grows and water runs," he said. "Where his spirit dwells, no one can say."

Eddy the Great

Post by Eddy the Great »

I'll have to take a look. Something that seems odd on a cursory glance is that the language seems not to use object agreement markers (though that may have an explanation deeper in the grammar).
It didn't seem to need them.

jburke

Post by jburke »

Eddy the Great wrote:
I'll have to take a look. Something that seems odd on a cursory glance is that the language seems not to use object agreement markers (though that may have an explanation deeper in the grammar).
It didn't seem to need them.
One of the hallmarks of poly languages is the marking of subject and object on the verb. While not absolutely necessary for comprehension, it greatly helps. E.g., in Mohawk, a word stem is made of a verb root and a noun root; the noun root is not inflected, so you need some indication of the person and number of the object--and this comes from the pronominal prefix (which for transitive expressions provides information on both subject and object). Another instance when object agreement markers are useful is when using free-standing objects. Object incorporation is always optional in Mohawk, e.g.; and if you have a free-standing object and subject both, you often need that object agreement marker to help you discern which free-standing noun is the subject and which the object.

Eddy the Great

Post by Eddy the Great »

So I'd add a suffix to the verb to indicate the case, person, and number of the object? I had to figure out how to make a polysynthetic language without any help because there little information on them on the internet and almost no polysynthetic conlangs exist. All I could find was basically: polysynthetic languages are highly synthetic.

jburke

Post by jburke »

Eddy the Great wrote:So I'd add a suffix to the verb to indicate the case, person, and number of the object? I had to figure out how to make a polysynthetic language without any help because there little information on them on the internet and almost no polysynthetic conlangs exist. All I could find was basically: polysynthetic languages are highly synthetic.
I'm an Indian enthusiast, and most of my own conlangs are polysynthetic, and grow from the poly languages I know. And you're right, there's damn little information on them out there; you really have to dig.

As for object agreement markers: yeah, create a table of affixes that signify the person, number, gender, etc. of objects, and give these affixes a place in the morphology. There's a lot of freedom here. Mohawk, an Iroquois language, uses fusional markers that encode details of subject and object (e.g., the marker yuk- indicates 3rd. pers. singular feminine subject and 1st pers. singular object). Cheyenne, an Algonquian language, (usually) uses prefixes to mark subject and suffixes to mark objects.

Eddy the Great

Post by Eddy the Great »

As for object agreement markers: yeah, create a table of affixes that signify the person, number, gender, etc. of objects, and give these affixes a place in the morphology. There's a lot of freedom here. Mohawk, an Iroquois language, uses fusional markers that encode details of subject and object (e.g., the marker yuk- indicates 3rd. pers. singular feminine subject and 1st pers. singular object). Cheyenne, an Algonquian language, (usually) uses prefixes to mark subject and suffixes to mark objects.
I don't have gender in my conlang and number is optional. Do you ahve any good sites to help me improve my conlang. I'm going to have to change the very basis of this language to fix this.

jburke

Post by jburke »

I don't have gender in my conlang and number is optional. Do you ahve any good sites to help me improve my conlang. I'm going to have to change the very basis of this language to fix this.
I don't know of any sites that describe poly grammars on this level of detail. My best advice is to get a grammar of poly language. Or, as I've told others, you can wait until I conquer the morphology of N?yat?k?h, my own polysynthetic opus; I speak two poly languages myself, and know them inside and out, so I'll be able to provide some insights that a typical grammar doesn't.

But, for now, think of object agreement markers as case inflections, only affixed to the verb instead of to the noun. (And you definitely will want to reconsider the optionality of number; one of big reasons that languages grammaticalize stuff like number and gender is to help tie things together--such as I described in the earlier post. E.g., if you have a two free-standing nouns, and you have three dimensions of distinction--person, number, gender--odds are that you'll always be able to always discern subject from object via those distinctions. Person, number, gender and other distinctions help us parse expressions.)

Eddy the Great

Post by Eddy the Great »

But, for now, think of object agreement markers as case inflections, only affixed to the verb instead of to the noun. (And you definitely will want to reconsider the optionality of number; one of big reasons that languages grammaticalize stuff like number and gender is to help tie things together--such as I described in the earlier post. E.g., if you have a two free-standing nouns, and you have three dimensions of distinction--person, number, gender--odds are that you'll always be able to always discern subject from object via those distinctions. Person, number, gender and other distinctions help us parse expressions).
What would be a good language grammer to look at?

jburke

Post by jburke »

Eddy the Great wrote:
But, for now, think of object agreement markers as case inflections, only affixed to the verb instead of to the noun. (And you definitely will want to reconsider the optionality of number; one of big reasons that languages grammaticalize stuff like number and gender is to help tie things together--such as I described in the earlier post. E.g., if you have a two free-standing nouns, and you have three dimensions of distinction--person, number, gender--odds are that you'll always be able to always discern subject from object via those distinctions. Person, number, gender and other distinctions help us parse expressions).
What would be a good language grammer to look at?
The Iroquois languages (Mohawk, Seneca, Oneida, Cherokee) are classic paradigms of polysyntheism; and also the Algonquian languages (Cheyenne, Ojibwa, Miami, Shawnee, Blackfoot, lots of others). But these languages are rather obscure and materials on them hard to find; you really have to search. My Mohawk grammar is by Nancy Bonvillain and my Cheyenne grammar by Wayne Leman. Grammars tend to be of a reference nature, and hence won't often include explicit discussion of topics like this; but you can try.

Eddy the Great

Post by Eddy the Great »

K'ulatqsaksa.
K'ula-tqsa-ksa.
building-random_location_case-is.
He is somewhere around the building.
Kak'ula'ikemaa.
Ka-k'ula-'ike-ma-a
Imperf-building-make-I
I have been making a building.
So show me an exaple of what I need on these sentences.

jburke

Post by jburke »

K'ulatqsaksa.
K'ula-tqsa-ksa.
building-random_location_case-is.
He is somewhere around the building.
No object marker would be needed here, as the sentence is not transitive.
Kak'ula'ikemaa.
Ka-k'ula-'ike-ma-a
Imperf-building-make-I
I have been making a building.
Let's say you add -sa to the end of this word (making it Kak'ula'ikemaasa), indicating that the object (building) is singular third person. Or, let's say that 'building' is masculaine, -sa could indicate that the object is singular, third person and masuline all three.

Eddy the Great

Post by Eddy the Great »

How would inflected nouns work? Should they still be incorporated? Won't this drive this conlang's degree of synthesis, around 3.95, to 4.95, which is impossibly high? How would this be handled?
Mne//amaa ga@'o'amaifsque s@a!omie.
Mne//a-ma-a ga-@'o'a-ma-i-fsqu-e s@a!o-mi-e.
write-I-past so_much_that-need-I-genitve-hand-future sleep(rest)-it-future.
I wrote so much that my hand will need to rest.

jburke

Post by jburke »

How would inflected nouns work?
Invent a nominal morphology.
Should they still be incorporated?
No reason why they can't be. Noyatukah, e.g., allows incorporation not just of objects, but of nominals in locatives ("upon the hill," e.g.) and the subjects of certain kinds of intransitive expressions. This is inspired by Mohawk morphology, which allows much of the same.
Won't this drive this conlang's degree of synthesis, around 3.95, to 4.95, which is impossibly high? How would this be handled?
There's no theoretical upper limit on the degree of synthesis allowed; Mohawk is incredibly synthetic, e.g., and exhibits a large degree of fusion in its morphemes.

BTW, how are calculating "degree of synthesis" exactlty?

Eddy the Great

Post by Eddy the Great »

Invent a nominal morphology.
This will sound dumb, but what is a nomonal morphology.

Eddy the Great

Post by Eddy the Great »

BTW, how are calculating "degree of synthesis" exactlty?
You take a sample of text, count the words and morphemes, and devide the number of morphemes by the number of words.

jburke

Post by jburke »

Eddy the Great wrote:
Invent a nominal morphology.
This will sound dumb, but what is a nomonal morphology.
In this case, a system of indicating person, number and gender on nouns. E.g., let's say the noun stem ohwair means 'tree'; add -on to it and you indicate that the noun is plural and neuter.

Eddy the Great

Post by Eddy the Great »

In this case, a system of indicating person, number and gender on nouns. E.g., let's say the noun stem ohwair means 'tree'; add -on to it and you indicate that the noun is plural and neuter.
But there are 25 cases in my conlang. If there are 2 numnbers and 2 genders, that means there will be 100 case endings.

jburke

Post by jburke »

Eddy the Great wrote:
BTW, how are calculating "degree of synthesis" exactlty?
You take a sample of text, count the words and morphemes, and devide the number of morphemes by the number of words.
Never had a reason to do this; but where did you get the notion that 4.95 would be an impossible degree of synthesis? That's only c. 5 morphemes per word; Mohawk words regularly have 7+ morphemes per word.

(But this formula will be foiled by fusionality of morphemes: some morphemes encode multiple meanings; so the degree of synethesis that you calculate in such a case will be lower than the actual degree.)

jburke

Post by jburke »

Eddy the Great wrote:
In this case, a system of indicating person, number and gender on nouns. E.g., let's say the noun stem ohwair means 'tree'; add -on to it and you indicate that the noun is plural and neuter.
But there are 25 cases in my conlang. If there are 2 numnbers and 2 genders, that means there will be 100 case endings.
Keep the case markers separate from the gender/number markers.

Eddy the Great

Post by Eddy the Great »

So if -me is 2nd person sing. subject and -ni is 3rd person sing. object:
!akfikimeaq?
!a-kfi-ki-me-a-q
Hypo-airplane-use-you-past-question?
Would you have used the airplane?
Becomes:

!akfikinimeaq?
!a-kfi-ki-ni-me-a-q
Hypo-airplane-use-3rd.(obj)-2nd.(sub)-past-question?
Would you have used the airplane?

jburke

Post by jburke »

Eddy the Great wrote:So if -me is 2nd person sing. subject and -ni is 3rd person sing. object:
!akfikimeaq?
!a-kfi-ki-me-a-q
Hypo-airplane-use-you-past-question?
Would you have used the airplane?
Becomes:

!akfikinimeaq?
!a-kfi-ki-ni-me-a-q
Hypo-airplane-use-3rd.(obj)-2nd.(sub)-past-question?
Would you have used the airplane?
That works.

Eddy the Great

Post by Eddy the Great »

So can the subject and object now be incorporated in any order?

jburke

Post by jburke »

Eddy the Great wrote:So can the subject and object now be incorporated in any order?
Let's distinguish between subject and object markers and subject and object morphemes. Usually, there are specific places within a morphology for subject and object markers--e.g., Cheyenne (usually) prefixes subject markers and suffixes object markers. This is to help avoid confusion. OTOH, incorporating languages generally don't allow incorporation of both subject and object morphemes simultaneously.

Post Reply