Random ideas: Morphosyntax

A forum for all topics related to constructed languages
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Random ideas: Morphosyntax

Post by Creyeditor »

I had an idea for a naturalistic self-segregating morphosyntax. Instead of using segments one could use phonological features or prosodic units. An example would be a language where words can only start with a marked laryngeal feature, e.g. aspirated stops, a glottal stop, etc. Phrases always end in a nasal sound (stolen from a similar process in Konni) or if you want to be even more natural in a long vowel. Here are some strings and there automatic segregation. I guess there may be similar features in natlangs for utterances and/or morphemes, especially if we look at suprasegmental features like tone.

bʰäntʰugʰamkʰebʰenʔigãkʰodägodikʰoduqinpʰagʰogiqan = bʰän, tʰu gʰam, kʰebʰen, ʔigã, kʰodägodi kʰoduqin, pʰa gʰogiqan.
pʰãkʰuqanʔetapetĩʔimgʰotidän = pʰã kʰuqan, ʔetapetĩ, ʔim, gʰotidän.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Random ideas: Morphosyntax

Post by Micamo »

loglorn wrote:Salishan languages can be analyzed as having no nouns, only verbs. There are arguments against it, but i haven't read them.
The main argument is that, although any noun can be zero-derived into a predicate in salishan languages (and wakashan), you can't do the same thing with verbs: A noun can occur independently in a clause with no special markers, but for a verb to appear in a nominal position it must take nominalization morphology.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random ideas: Morphosyntax

Post by Frislander »

Micamo wrote:
loglorn wrote:Salishan languages can be analyzed as having no nouns, only verbs. There are arguments against it, but i haven't read them.
The main argument is that, although any noun can be zero-derived into a predicate in salishan languages (and wakashan), you can't do the same thing with verbs: A noun can occur independently in a clause with no special markers, but for a verb to appear in a nominal position it must take nominalization morphology.
However we also see that a lot of root nouns seem to occur with the s-prefix type structure anyway, and nominalised verbs can be used in predicative position as well (see Saanich).
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Random ideas: Morphosyntax

Post by Micamo »

True, but the requirement for nominalization morphology on verb roots to appear in nominal positions is enough to prove that a category distinction does exist, though the language conspires to hide it well.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Random ideas: Morphosyntax

Post by eldin raigmore »

Creyeditor wrote:I had an idea for a naturalistic self-segregating morphosyntax. Instead of using segments one could use phonological features or prosodic units. An example would be a language where words can only start with a marked laryngeal feature, e.g. aspirated stops, a glottal stop, etc. Phrases always end in a nasal sound (stolen from a similar process in Konni) or if you want to be even more natural in a long vowel. Here are some strings and theretheir automatic segregation. I guess there may be similar features in natlangs for utterances and/or morphemes, especially if we look at suprasegmental features like tone. …


In my experience so far, it is very difficult (AFAICT impossible) to use such features both to unambiguously self-segregate morphemes and to unambiguously self-segregate words; or both words and phrases; or both morphemes and phrases.
I would love to be proven wrong! You could be more skilled than I; I'd like to see this "random idea" carried out.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random ideas: Morphosyntax

Post by Frislander »

Small sketch:

/p t t͡ʃ k <p t c k>
/s ʃ h/ <s sh h>
/m n ŋ/ <m n g>
/ɾ/ <r>

/i a u/ <<i a u>>

(C)V

Vowels may occur in hiatus, however no more than two identical vowels may appear together. Vowels may also take high tone, marked by an acute accent.

The main word classes are nouns and verbs, which can be divided into stative and dynamic.

Nouns:

naka - person
úiti - cat
shíni - knife
atuahi - house

Verbs:

naua - to be big
úrí - to sit (in)
iaáka - to take
surá - to give to
maíígi - to eat
tápa - to hit

Standard word order is SV(O). The subject of a transative or intransitive verb is marked with the suffix -ku. A transitive verb takes the suffix -raí.

atuahiku naua
house-NOM be.big
the house is big

nakaku táparaí úiti
person-NOM hit-TRANS cat
The person hits the cat.

The transitivising suffix can serve to increase the transitivity of an intransitive verb:

úitiku úríraí atuahi
cat-NOM sit-TRANS house
the cat is sitting in the house

EDIT: unfinished, more to come soon.
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Random ideas: Morphosyntax

Post by Creyeditor »

This may be in the wrong thread [xD]
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random ideas: Morphosyntax

Post by Frislander »

Creyeditor wrote:This may be in the wrong thread [xD]
OK, what thread should I put it in, then?
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Random ideas: Morphosyntax

Post by Creyeditor »

I don't know. I might have misunderstood you, but what exactly are your random morphosyntax ideas?
Nouns vs. Dynamic Verbs vs. Stative Verbs as parts of speech?
SV(O) word order? Nominative accusative alignment?
Overt transitivity marking?
A combination thereof?

I guess I was just confused, because you started with the phonemic inventory, so I thought you were providing a brief sketch of your new conlang.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random ideas: Morphosyntax

Post by Frislander »

Creyeditor wrote:I don't know. I might have misunderstood you, but what exactly are your random morphosyntax ideas?
Nouns vs. Dynamic Verbs vs. Stative Verbs as parts of speech?
SV(O) word order? Nominative accusative alignment?
Overt transitivity marking?
A combination thereof?

I guess I was just confused, because you started with the phonemic inventory, so I thought you were providing a brief sketch of your new conlang.
You know, to be honest it kind-of is. I was going to save it as a draft and put up more stuff (which is where it gets interesting), but I accidentally hit the post button instead. I'll start a new thread on it, where I'll have the chance to explore it further.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Random ideas: Morphosyntax

Post by eldin raigmore »

Creyeditor wrote:I had an idea for a naturalistic self-segregating morphosyntax. Instead of using segments one could use phonological features or prosodic units. An example would be a language where words can only start with a marked laryngeal feature, e.g. aspirated stops, a glottal stop, etc. Phrases always end in a nasal sound (stolen from a similar process in Konni) or if you want to be even more natural in a long vowel. Here are some strings and their automatic segregation. I guess there may be similar features in natlangs for utterances and/or morphemes, especially if we look at suprasegmental features like tone. …
I'm interested; how did this work out?
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Random ideas: Morphosyntax

Post by Creyeditor »

eldin raigmore wrote:
Creyeditor wrote:I had an idea for a naturalistic self-segregating morphosyntax. Instead of using segments one could use phonological features or prosodic units. An example would be a language where words can only start with a marked laryngeal feature, e.g. aspirated stops, a glottal stop, etc. Phrases always end in a nasal sound (stolen from a similar process in Konni) or if you want to be even more natural in a long vowel. Here are some strings and their automatic segregation. I guess there may be similar features in natlangs for utterances and/or morphemes, especially if we look at suprasegmental features like tone. …
I'm interested; how did this work out?
Well, one problem that I encountered is the limited amount of consonants that can appear in word initial position. If we assume a simple phoneme inventory /p t c k ʔ b d g f s h m n l r j w/ (assuming only voiceless plosives can be aspirated) we get only seven possible sounds at the beginning of a word. In non-intial syllables on the other hand, we get 15 consonants + 5 vowels at the beginning of a syllable. This means that at the edge of a word, there are less possibilities than inside a word, which strikes me as unnatural and also seems to be kind of ineffecient. I have not yet found a satisfactory solution for that.
Phrases on the other hand are pretty straightforward. If we assume a (C)V(N) syllable structure words can only end in a nasal or nasalized vowel, whereas word-internal syllables can only end in an oral vowel.
I also had an idea for morpheme seperation based on stress. The stem would get the primary stress and every morpheme would begin with a secondarily stressed syllable. That would require every morpheme to have at least one vowel, but I think in a CV language that is okay. This is taken from Nimboran, which has a similar system.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random ideas: Morphosyntax

Post by Frislander »

Creyeditor wrote:
eldin raigmore wrote:
Creyeditor wrote:I had an idea for a naturalistic self-segregating morphosyntax. Instead of using segments one could use phonological features or prosodic units. An example would be a language where words can only start with a marked laryngeal feature, e.g. aspirated stops, a glottal stop, etc. Phrases always end in a nasal sound (stolen from a similar process in Konni) or if you want to be even more natural in a long vowel. Here are some strings and their automatic segregation. I guess there may be similar features in natlangs for utterances and/or morphemes, especially if we look at suprasegmental features like tone. …
I'm interested; how did this work out?
Well, one problem that I encountered is the limited amount of consonants that can appear in word initial position. If we assume a simple phoneme inventory /p t c k ʔ b d g f s h m n l r j w/ (assuming only voiceless plosives can be aspirated) we get only seven possible sounds at the beginning of a word. In non-intial syllables on the other hand, we get 15 consonants + 5 vowels at the beginning of a syllable. This means that at the edge of a word, there are less possibilities than inside a word, which strikes me as unnatural and also seems to be kind of ineffecient. I have not yet found a satisfactory solution for that.
It doesn't strike me as at all unnatural: plenty of languages have a good number of phonemes which only contrast word-internally just look at the phonotactics of Inuit or one of the Aboriginal Australian languages: in the latter case sometimes the language only permits non-apical stops, nasals or glides to appear initially.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Random ideas: Morphosyntax

Post by eldin raigmore »

Thanks, @creyeditor and @Frislander.

Another idea;
(1) Only word-initial syllables can have onset clusters.
(2) Only word-final syllables can have coda clusters.
(3) Only one-syllable words can have both an onset cluster and a coda cluster.
(4a) Any word-internal syllable's onset, if it has one, must be at most a single consonant;
and
(4b) Any word-internal syllable's coda, if it has one, must be at most a single consonant.

This is naturalistic and realistic because a natlang does it.
Maybe it doesn't automatically self-segregate absolutely every word, but it probably does in most cases, and helps a lot in most of the rest.


[hr][/hr]


Another idea:
(1) Sounds whose airstream is glottalic or velaric or ingressive can only be word-initial; any non-word-initial consonants must be pulmonic egressive.
(2) The velaric nasal can occur only word-finally.

These are naturalistic and realistic because some natlangs do each of them.
Once again, maybe they don't automatically self-segregate absolutely every word, but they probably do in most cases, and help a lot in most of the rest.


[hr][/hr]


Would ideas like those be of any help?


[hr][/hr]
Edit: Sorry, I just realized that not only do those ideas not necessarily automatically segregate every word; the also don't depend on suprasegmental features. So they're definitely not what you're looking for.

The most obvious thing to do would be for every word with two or more syllables to have the first syllable primarily stressed, and every word with three or more syllables have the last syllable secondarily stressed -- or vice-versa. An unstressed syllable between a secondarily stressed syllable and the next (or previous) primarily stressed syllable, must be a one-syllable word. A string of such unstressed syllables must be a string of one-syllable words. And if there are two or more syllables between two primarily stressed syllables that don't have a secondarily-stressed syllable between them, then that's a two-syllable word followed (or preceded) by one (or more) one-syllable word(s).

Here are some more ideas, for a tonal language with more than two level tones and some glide-tones or contour-tones.
1. The first syllable of any sentence is lengthened by inserting a medium tone just before its tone, unless its tone already begins at medium; and the last syllable of any sentence is similarly lengthened by inserting a medium tone just after its tone, unless it already ends with a medium tone.
2. Also, the first and last high tones in a sentence are raised to extra-high unless they're already extra-high; and
3. the first and last low tones of a sentence are lowered to extra-low, unless they're already extra-low.
Obviously 1 requires there to be a medium tone; 2 requires there to be an extra-high tone; and 3 requires there to be an extra-low tone. (1 also requires that lengthening a tone by inserting a medium tone just before or just after it, might sometimes result in a glide or contour that is a legal tone in your 'lang.) You could use any one of those three ideas, or any two of them, or all three of them, depending on which level tones your 'lang has.
You might use each idea on phrases or on clauses instead of on sentences. And that decision might be independently made for each idea. You might use 1 for phrases, 3 for clauses, and 2 for sentences, for instance.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Random ideas: Morphosyntax

Post by eldin raigmore »

Here's an idea. (Maybe it's a good one, or maybe not!)

The language I wish to consider here has pitch-tones attached to its syllables.
It has five level tones; medium (3), high (4), low (2), extra-high (5), and extra-low (1).
It also has several glide tones, both rises and falls.
And it has simple contour tones; peaking (rise-fall) tones and dipping (fall-rise) tones. But the occurrence of contour tones may be somewhat restricted, except on syllables that are either clause-initial or clause-final or both.
And it has complex contour tones; that is, peak-dips and dip-peaks. However, these do not occur at all on syllables that are neither sentence-initial nor sentence-final.

If a syllable's tone is 3153 or 3513, it must be a one-syllable sentence; that is, a one-word sentence consisting of a one-syllable word.
Syllables whose tone is a 353 peak or a 313 dip also stand a good chance of being one-syllable sentences.
Finally, some one-syllable sentences could have just a medium level tone (i.e., 3).

If a syllable's tone is 3151 or 3152 or 3154, or 3512 or 3514 or 3515, it must be a sentence-initial syllable.
If a syllable's tone is 1513 or 2513 or 4513, or 2153 or 4153 or 5153, it must be a sentence-final syllable.

If a syllable's tone is a 315 dip or a 351 peak, it's probably sentence-initial.

If a syllable's tone is a 153 peak or a 513 dip, it's probably sentence-final.

But, maybe, some sentence-initial syllables will have a 354 or 353 peak, or a 312 or 313 dip tone; and maybe some sentence-final syllables will have a 213 or 313 dip, or a 353 or 453 peak tone.

And some sentence-initial syllables can have a 35 rise or a 31 fall; and some sentence-final syllables can have a 53 fall or a 13 rise.

But no other tones than those mentioned above can occur on sentence-initial or sentence-final syllables.

That's thirty-seven tones so far, of which thirty-five might occur on sentence-initial or sentence-final syllables. But I'm not saying they do, in fact, all occur. However there should be some tones that occur only on "sentence-medial" syllables. They should include the level tones 1 and 2 and 4 and 5, and probably or maybe most of the rises 12, 14, 15, 23, 25, 34, and 45, and probably or maybe most of the falls 21, 32, 41, 43, 51, 52, and 54. And maybe there'd be one or a few peaking and/or dipping simple-contour tones, not mentioned above, that also can occur only on "sentence-medial" syllables.

The natlang I remember seeing reported with the most pitch-tones was reported to have fourteen of them. (And the natlang I remember seeing report with the most level pitch-tones has six of them.) This conlang I'm talking about probably has around eighteen pitch-tones (of which five are level tones), unless my first guess is wrong.
Hyolobrika
hieroglyphic
hieroglyphic
Posts: 45
Joined: 19 Sep 2016 15:45
Location: le kibro gugde
Contact:

Re: Random ideas: Morphosyntax

Post by Hyolobrika »

  1. Background/foreground inflection on nouns
    So: "I like the smell.BCK of rain" means that I like the smell being there or I like it in conjunction with other sensory experiences and "I like the smell.FGD of rain" means that I like noticing it.
    Does that make any sense to you? :?:
  2. Instead of a separate set of role markers for nouns, use agreement between different nouns
    So: to form 'computer' in the instrumental case, simply agree the word for computer with the word for tool. This system can also be used for a copula.
    I think this would be good for minimalist toki pona ish language. It would depend on context, for instance, whether "make.PST undergoer-I tool-I me-II agent-II" means that I made a tool or that I used the undergoer of a different action to make something.
User avatar
gestaltist
mayan
mayan
Posts: 1617
Joined: 11 Feb 2015 11:23

Re: Random ideas: Morphosyntax

Post by gestaltist »

Hyolobrika wrote:
  1. Background/foreground inflection on nouns
    So: "I like the smell.BCK of rain" means that I like the smell being there or I like it in conjunction with other sensory experiences and "I like the smell.FGD of rain" means that I like noticing it.
    Does that make any sense to you? :?:
I love this idea!
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Random ideas: Morphosyntax

Post by eldin raigmore »

gestaltist wrote:
Hyolobrika wrote:
  1. Background/foreground inflection on nouns
    So: "I like the smell.BCK of rain" means that I like the smell being there or I like it in conjunction with other sensory experiences and "I like the smell.FGD of rain" means that I like noticing it.
    Does that make any sense to you? :?:
I love this idea!
I do too! [<3]






(P.S. I also love not having my last previous post still be the top post on this thread. [:$] )
Hyolobrika
hieroglyphic
hieroglyphic
Posts: 45
Joined: 19 Sep 2016 15:45
Location: le kibro gugde
Contact:

Re: Random ideas: Morphosyntax

Post by Hyolobrika »

eldin raigmore wrote:
gestaltist wrote:
Hyolobrika wrote:
  1. Background/foreground inflection on nouns
    So: "I like the smell.BCK of rain" means that I like the smell being there or I like it in conjunction with other sensory experiences and "I like the smell.FGD of rain" means that I like noticing it.
    Does that make any sense to you? :?:
I love this idea!
I do too! [<3]
It came to me one night as I was walking home after reading this very thread and smelling the rain.BCK then rain.FGD!
Hyolobrika
hieroglyphic
hieroglyphic
Posts: 45
Joined: 19 Sep 2016 15:45
Location: le kibro gugde
Contact:

Re: Random ideas: Morphosyntax

Post by Hyolobrika »

  • A quick way to refer to the appearance of an object rather the actual object (in practice just the object as it appears to us with superior methods of perception because we aren't perfect)
    EDIT: So: According the the Copernican model, "the sun rises.APPEARANCE" is true because it's useful to act as if "the sun rises.ACTUAL" is true even though it's not. Perhaps I should have said "useful to act", sorry [:|]
    Makes me think of Ithkuil's Imputative sanction
Post Reply