Vreleksá Forum Index Vreleksá
The Alurhsa Word for Constructed: Creativity in both scripts and languages
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

A Poly-Synthetic Conlang for Langover.

 
Post new topic   Reply to topic    Vreleksá Forum Index -> Conlangs
View previous topic :: View next topic  
Author Message
langover94



Joined: 21 Aug 2007
Posts: 509
Location: USA

PostPosted: Sat Jun 14, 2008 5:02 pm    Post subject: A Poly-Synthetic Conlang for Langover. Reply with quote

Ok guys, I want to make a poly-synthetic lang that is Slavic in nature. So far, I have worked out a phonology, and that nouns can be declined to two genders (masculine/feminine), three numbers (singular/dual/plural), and five cases (nominative/accusative/genetive/prepositional/vocative).

My question is, what other things are needed to make a language especially poly-synthetic?
_________________
Join us at: The Renewed Spirits Forum!
Please join for good discussion. (We need members!)
Back to top
View user's profile Send private message
Tolkien_Freak



Joined: 26 Jul 2007
Posts: 1231
Location: in front of my computer. always.

PostPosted: Sat Jun 14, 2008 9:48 pm    Post subject: Reply with quote

Combining the subject and object INTO the verb itself, not even as separate words.
Back to top
View user's profile Send private message
langover94



Joined: 21 Aug 2007
Posts: 509
Location: USA

PostPosted: Sun Jun 15, 2008 11:22 pm    Post subject: Reply with quote

Surely there must be more?
_________________
Join us at: The Renewed Spirits Forum!
Please join for good discussion. (We need members!)
Back to top
View user's profile Send private message
Tolkien_Freak



Joined: 26 Jul 2007
Posts: 1231
Location: in front of my computer. always.

PostPosted: Mon Jun 16, 2008 12:10 am    Post subject: Reply with quote

That's just all I could think of.

Essentially, what you want to do is combine as many concepts as possible into as few words as possible.
Back to top
View user's profile Send private message
langover94



Joined: 21 Aug 2007
Posts: 509
Location: USA

PostPosted: Wed Jun 18, 2008 9:12 pm    Post subject: Reply with quote

Not even Aeetlrcreejl wants to give some adivice?
_________________
Join us at: The Renewed Spirits Forum!
Please join for good discussion. (We need members!)
Back to top
View user's profile Send private message
Aeetlrcreejl



Joined: 08 Jun 2007
Posts: 839
Location: Over yonder

PostPosted: Thu Jun 19, 2008 11:09 am    Post subject: Reply with quote

Make your morphemes short. That's all I can offer at this early hour. I'll come back later.

I think the L&L Museum on the other board has a section on polysynthetic conlangs.
_________________
Iwocwá ĵọṭãsák.
/iwotSwa_H d`Z`Ot`~asa_Hk/
[iocwa_H d`Z`Ot`_h~a_Hk]
Back to top
View user's profile Send private message Visit poster's website
Hemicomputer



Joined: 04 Feb 2008
Posts: 610
Location: Calgary, Alberta

PostPosted: Wed Jun 25, 2008 6:41 pm    Post subject: Reply with quote

Any suggestions as to what a good, average number of morphemes is? I've also been thinking of making a polysynthetic lanuage.

EDIT:I accidentaly confused the words "phoneme" and "morpheme". D'oh!
_________________
Bakram uso, mi abila, / del us bakrat, dahud bakrita!
Back to top
View user's profile Send private message
mrtoast2



Joined: 19 Feb 2008
Posts: 123
Location: Goromonzi

PostPosted: Thu Jun 26, 2008 9:08 pm    Post subject: Reply with quote

I would love to make one, but I think it would be too much to wrap my brain around. Gwaken is good enough for now.
_________________
Tôśt drônén kókślán! Vón kríngénã Tôśt! Gâgén šníkél dér îwâ!
Back to top
View user's profile Send private message Send e-mail Visit poster's website
eldin raigmore
Admin


Joined: 03 May 2007
Posts: 1621
Location: SouthEast Michigan

PostPosted: Thu Jun 26, 2008 10:19 pm    Post subject: Reply with quote

Hemicomputer wrote:
Any suggestions as to what a good, average number of morphemes is? I've also been thinking of making a polysynthetic lanuage.

EDIT:I accidentaly confused the words "phoneme" and "morpheme". D'oh!


IMO your lexicon should start out with about 800 to about 1400 morphemes, (or 800 to 1400 roots?), to get good and started. Eventually you'll want 3,000 to 5,000 to cover everyday speech (conlangers who get to 5000 are justifiably proud!), and to cover expert vocabularies as well you might want 30,000 to 50,000.

If your average number of morphemes per word is more than four, then your average morpheme is neither the first nor the last morpheme in its word. Then your language will be towards the "synthetic" end of that "synthetic vs analytic" or "synthetic vs isolating" dimension.

"Polysynthetic" is a term not consistently applied. You should research it (Google, Wiki, search for it on the ZBB and the CBB, etc.) to see what it means; it could mean several different things, and sometimes means different things to different linguists.

In general, though, it means that morphology trumps syntax; that whole sentences can be single words; that there's a lot of word-coining going on; that objects or patients tend to get incorporated into the verb; especially, that they get incorporated even when they are specific or referential; that there is poly-personal agreement, perhaps to the extent that the verb agrees with every participant; especially if it does so to the extent that, if every participant is pronominal (rather than a full-noun-phrase), the verb is the only word that needs to be in the sentence.

-------------------------------------------------------------------------------------

A good polysynthetic language's agreement system might have the following characteristics.

Nouns might have many nominal categories; each might have many values. So for instance there might be lots of cases, definiteness-values, genders, locations (a la demonstratives), numbers, person, pragmatic status, referentiality/specificity, etc. Pronouns could agree with all of these.

The verb might actually "incorporate" reduced forms of the pronouns for its participants; this kind of agreement is called "cross-reference".

But in some real-life polysynthetic languages, there is some "fusion" going on. A single affix might tell the gender and number of the agent as well as the gender and number of the patient. There may be no way to predict, for example, from knowing what value of that affix to use for each of
  • feminine singular agent and feminine plural patient;
  • feminine singular agent and masculine singular patient;
  • feminine pural agent and feminine singular patient;
  • masculine singular agent and feminine singular patient;
to know what to use for a feminine singular agent and a feminine singular patient.

The "agglutinative vs fusing" dimension is separate from the "isolating vs synthetic" dimension. If more than half of your morphemes combine two or more "meanings" (e.g. aspect and mood, or aspect and polarity, or aspect and tense, or aspect and voice, or mood and polarity, or mood and tense, or mood and voice, or polarity and tense, or polarity and voice, or tense and voice; or, case and definiteness, or case and gender, or case and number, or case and person, or definiteness and gender, or definiteness and number, or definiteness and person, or gender and number, or gender and person, or number and person) then your language will be, to that degree, "fusional" as opposed to "agglutinating".

An isolating, fusional language would have one-morpheme-per-word; each morpheme being an entire word; but the word would express two categories of meaning simultaneously.

An isolating, non-fusional language would have one morpheme per word and one meaning per morpheme.

A synthetic, fusional language would have more than one (say, two) morphemes per word on average; not most morphemes would be the first in their word, and not most would be last, (though if it isn't polysynthetic maybe most would be either first or last). And also each morpheme would usually have two meanings; say, case and number and gender and person, and if you changed case but not gender or number or person you wouldn't know how to change the morpheme unless you'd just memorized it.

A synthetic, agglutinating language would have more-than-one morpheme per word (say, two per word on average) and would also have only one meaning per morpheme (so a pronoun would have one part for case, one for gender, one for number, and one for person; and a verb would have one morpheme to agree with its agent, and another to agree with its patient, etc.)

You want your language to be very highly synthetic; but perhaps you'd like it to be very fusional instead of agglutinating. Or perhaps you'd like it to be very agglutinating instead of fusional. You haven't said. But Slavic nouns are quite fusional, not agglutinating; is the same true of their verbs? I don't know.

-------------------------------------------------------------------------------------

You're probably going to want three thousand to five thousand morphemes to be able to keep translating or composing in your language with adequate freedom; and thirty thousand to fifty thousand to make it complete enough for experts to use.

You probably can start with, say, eight hundred to twelve hundred morphemes.

As for how many phonemes you want; that depends on how many syllables you want.

Your morphemes needn't contain entire syllables, especially not the morphemes which never stand alone as entire words; indeed, maybe none of your non-root morphemes need be (nor contain) an entire syllable. The "-s" plural ending of English is an example of a morpheme shorter than a syllable.

You should lay out the morphological pattern, blueprint, formula, or whatever-you-want-to-call-it, of nouns; and do the same for verbs. Where does the root go? Where do the morphemes (probably affixes) for case, number, and pragmatic status go? The verb will be even more complicated, but just because it has so many more accidents (including agreement with the nouns!), not because the basic idea is any harder to understand.

You should also lay out the phonological pattern of the root morphemes of each major part-of-speech; e.g. nouns, verbs, pronouns, adjectives, or adverbs. (Nouns and verbs are the biggies; polysynthetic languages usually don't have adverbs (they have various kinds of affixes instead, or some of them have special kinds of adjectives instead), and frequently don't have adjectives (they have special kinds of nouns instead, or have special kinds of verbs instead). But you do need to know about your pronouns; and pronouns may have morphemes for gender and person, while nouns probably have neither.)

Your noun-roots' phonological pattern needn't be quite the same as your verb-roots' phonological pattern.

Once you decide on the phonological patterns of each of your various big kinds of morphemes, you can decide how many different syllables you want.

I recommend that you have at least 791 different syllables if your roots are going to be at most two syllables long, and at least 86 different syllables if your roots are going to be at most three syllables long.

To get 791 syllables you probably need to allow two consonants in a syllable. This means probably that you either allow CV(C) syllables, or that you allow C(C)V or (C)CV syllables.

To get 86 syllables you can probably get away with allowing only CV syllables.

10 consonants and 8 vowels,
11 consonants and 7 vowels,
12 consonants and 6 vowels,
13 consonants and 5 vowels,
15 consonants and 4 vowels, or
17 consonants and 3 vowels,

will give you 791 or more syllables of any of the patterns
(C)V(C) or (C)(C)V or simlar.

So if your roots are bisyllabic -- (C)V(C)((C)V(C)) or (C)(C)V((C)(C)V) -- you could make 625000 or more of them. And you're probably not going to make more than about 50,000, though of course you're welcome to.

10 consonants and 9 vowels,
11 consonants and 8 vowels,
13 consonants and 7 vowels,
15 consonants and 6 vowels,
18 consonants and 5 vowels,
22 consonants and 4 vowels, or
29 consonants and 3 vowels,

will give you 86 or more syllables of the pattern (C)V.

So of your roots are trisyllabic -- (C)V((C)V((C)V)) -- you could make 625000 or more of them.
_________________
"We're the healthiest horse in the glue factory" - Erskine Bowles, Co-Chairman of the deficit reduction commission
Back to top
View user's profile Send private message
Hemicomputer



Joined: 04 Feb 2008
Posts: 610
Location: Calgary, Alberta

PostPosted: Fri Jun 27, 2008 1:37 pm    Post subject: Reply with quote

eldin raigmore wrote:
IMO your lexicon should start out with about 800 to about 1400 morphemes, (or 800 to 1400 roots?), to get good and started. Eventually you'll want 3,000 to 5,000 to cover everyday speech (conlangers who get to 5000 are justifiably proud!), and to cover expert vocabularies as well you might want 30,000 to 50,000.

Thanks!
eldin raigmore wrote:
If your average number of morphemes per word is more than four, then your average morpheme is neither the first nor the last morpheme in its word. Then your language will be towards the "synthetic" end of that "synthetic vs analytic" or "synthetic vs isolating" dimension.

"Polysynthetic" is a term not consistently applied. You should research it (Google, Wiki, search for it on the ZBB and the CBB, etc.) to see what it means; it could mean several different things, and sometimes means different things to different linguists.

Hmm, if it weren't for my isolating the tense, Holxûs could almost be polysynthetic!
eldin raigmore wrote:
In general, though, it means that morphology trumps syntax; that whole sentences can be single words; that there's a lot of word-coining going on; that objects or patients tend to get incorporated into the verb...
Okay, that makes sense.

eldin raigmore wrote:
A good polysynthetic language's agreement system might have the following characteristics....

Thanks, that will be interesting to look over.

eldin raigmore wrote:
But in some real-life polysynthetic languages, there is some "fusion" going on. A single affix might tell the gender and number of the agent as well as the gender and number of the patient. There may be no way to predict, for example, from knowing what value of that affix to use for each of
  • feminine singular agent and feminine plural patient;
  • feminine singular agent and masculine singular patient;
  • feminine pural agent and feminine singular patient;
  • masculine singular agent and feminine singular patient;
to know what to use for a feminine singular agent and a feminine singular patient.
That is a neat concept. I'll be considering that or a similar operating dynamic.
eldin raigmore wrote:
The "agglutinative vs fusing" dimension is separate from the "isolating vs synthetic" dimension. If more than half of your morphemes combine two or more "meanings" /CLIP/ then your language will be, to that degree, "fusional" as opposed to "agglutinating".

/CLIP various definitions/

You want your language to be very highly synthetic; but perhaps you'd like it to be very fusional instead of agglutinating. Or perhaps you'd like it to be very agglutinating instead of fusional. You haven't said. But Slavic nouns are quite fusional, not agglutinating; is the same true of their verbs? I don't know.
I haven't really considered that aspect, but I like the idea of a fusional language. This is both because it's more interesting and because it would require that I create fewer words.

eldin raigmore wrote:
As for how many phonemes you want; that depends on how many syllables you want.

Your morphemes needn't contain entire syllables, especially not the morphemes which never stand alone as entire words; indeed, maybe none of your non-root morphemes need be (nor contain) an entire syllable. The "-s" plural ending of English is an example of a morpheme shorter than a syllable.
Useful stuff, thanks.

eldin raigmore wrote:
You should lay out the morphological pattern, blueprint, formula, or whatever-you-want-to-call-it, of nouns; and do the same for verbs. Where does the root go? Where do the morphemes (probably affixes) for case, number, and pragmatic status go?...
You should also lay out the phonological pattern of the root morphemes of each major part-of-speech; e.g. nouns, verbs, pronouns, adjectives, or adverbs....
Okay, that's a good first step.

eldin raigmore wrote:
Your noun-roots' phonological pattern needn't be quite the same as your verb-roots' phonological pattern.
This would help distinguish them.

eldin raigmore wrote:

I recommend that you have at least 791 different syllables if your roots are going to be at most two syllables long, and at least 86 different syllables if your roots are going to be at most three syllables long.

To get 791 syllables you probably need to allow two consonants in a syllable. This means probably that you either allow CV(C) syllables, or that you allow C(C)V or (C)CV syllables.
/CLIP the extremely useful info/


Great!
_________________
Bakram uso, mi abila, / del us bakrat, dahud bakrita!
Back to top
View user's profile Send private message
jimhenry



Joined: 21 Aug 2008
Posts: 4

PostPosted: Fri Aug 22, 2008 12:03 am    Post subject: Reply with quote

I think J.C. Wells, quoting Joseph Greenberg, describes languages as polysynthetic which have an average number of morphemes per word of >3. There was a thread a while ago on the CONLANG list about the index of synthesis (average # morphemes per word), with people calculating the index for their conlangs; Henrik Theiling's polysynthetic conlang Qthyn|gai had an index around 8, if I remember right; most others were 4 or less.

http://archives.conlang.info/jae/zeinpei/bianzhorsuen.html

Something I don't think previous posters have mentioned is that you could have verbal affixes for various concepts that most languages express with adverbs. E.g. affixes for fast, slow, deliberately, accidentally, often, too much, carelessly, carefully... And similarly express adjectival concepts with nominal affixes instead of stand-alone words.

Wells found a lot more variation in index of synthesis for Esperanto, calculating over different texts from different authors, than Greenberg did comparing various English texts. There was a thread on the CONLANG list a while ago about polysynthetic wordplay (in the sense of "unusually high number of morphemes per word", not "incorporating subject and object into the verb") in Esperanto:

http://archives.conlang.info/qae/whelgen/zuarduerkuan.html
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    Vreleksá Forum Index -> Conlangs
All times are GMT
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2002 phpBB Group
Theme ACID © 2003 par HEDONISM Web Hosting Directory


Start Your Own Video Sharing Site

Free Web Hosting | Free Forum Hosting | FlashWebHost.com | Image Hosting | Photo Gallery | FreeMarriage.com

Powered by PhpBBweb.com, setup your forum now!
For Support, visit Forums.BizHat.com