(L&N) Q&A Thread - Quick questions go here

A forum for discussing linguistics or just languages in general.
User avatar
CarsonDaConlanger
sinic
sinic
Posts: 250
Joined: 02 Nov 2017 20:55

Re: (L&N) Q&A Thread - Quick questions go here

Post by CarsonDaConlanger » 11 Apr 2019 00:08

For heavily head marking agglutinative languages, what is the general order of affixes? If a verb could take affixes for:
tense
aspect
mood
subject
object
voice affixes (applicative/passive/etc)
incorporated instrumental nouns
participle

Is there a trend in which affixes are closer to the stem and which ones are more distal? Which ones are more likely to be prefixes?

náʼoolkiłí
rupestrian
rupestrian
Posts: 15
Joined: 04 Apr 2019 20:06

Re: (L&N) Q&A Thread - Quick questions go here

Post by náʼoolkiłí » 11 Apr 2019 01:21

There are indeed pretty good crosslinguistic generalizations about morpheme ordering. For a suffixing language (it'd just be mirrored for a prefixing language) you'll typically have:

Verb Root – Incorporated Root – Voice – Aspect – Tense – Mood

The position of agreement is very flexible, but object agreement is usually inside (i.e., closer to the root than) subject agreement. If your agreement is ergative-aligned, ergative agreement is usually inside absolutive agreement. Object/Ergative agreement often gravitates to positions just outside of Voice (or is fused with it), and Subject/Absolutive agreement to positions outside of Tense (or is fused with it).

Person agreement is usually inside number agreement, and if person and number agreement are on either side of the root, then person agreement is likely the prefix and number agreement the suffix.

I'd expect participle morphology to be outermost, and I'd also expect participles to lack Tense, Mood, Subj/Abs.Agr, and maybe even more morphemes.

These are all just generalizations, though. Experiment with other orders or with blending and otherwise complexifying morphological template slots.

Nachtuil
greek
greek
Posts: 522
Joined: 21 Jul 2016 00:16

Re: (L&N) Q&A Thread - Quick questions go here

Post by Nachtuil » 11 Apr 2019 04:39

Creyeditor wrote:
08 Apr 2019 17:42
Assuming no noun class system means no gender/noun class system at all, this page gives you the statistics.
There is 1 languages with gender distinctions in the 3rd person singular and plural pronouns without a gender system (Carib). Also another 4 languages with gender distinction only in third person singular and no gender system (Burmese, Mandarin, Persian, Kilivila).
English is classified as having a sex based gender system though, so I don't really know how accurate the data is.

Does this answer your question?
I've looked over it a few times but it indeed isn't really what I'm looking for. It may be the closest that can be had without doing some in depth looking. I definitely appreciate the effort though. Thanks.

User avatar
Dormouse559
moderator
moderator
Posts: 2818
Joined: 10 Nov 2012 20:52
Location: California

Re: (L&N) Q&A Thread - Quick questions go here

Post by Dormouse559 » 11 Apr 2019 07:43

Creyeditor wrote:
08 Apr 2019 17:42
Assuming no noun class system means no gender/noun class system at all, this page gives you the statistics.
There is 1 languages with gender distinctions in the 3rd person singular and plural pronouns without a gender system (Carib). Also another 4 languages with gender distinction only in third person singular and no gender system (Burmese, Mandarin, Persian, Kilivila).
The inclusion of Mandarin is intriguing. To my knowledge, the third-person singular pronoun only distinguishes gender in writing, and the written distinction was itself introduced only recently.

User avatar
Zekoslav
sinic
sinic
Posts: 257
Joined: 07 Oct 2017 16:54

Re: (L&N) Q&A Thread - Quick questions go here

Post by Zekoslav » 11 Apr 2019 10:58

náʼoolkiłí wrote:
11 Apr 2019 01:21
There are indeed pretty good crosslinguistic generalizations about morpheme ordering. For a suffixing language (it'd just be mirrored for a prefixing language) you'll typically have:

Verb Root – Incorporated Root – Voice – Aspect – Tense – Mood

The position of agreement is very flexible, but object agreement is usually inside (i.e., closer to the root than) subject agreement. If your agreement is ergative-aligned, ergative agreement is usually inside absolutive agreement. Object/Ergative agreement often gravitates to positions just outside of Voice (or is fused with it), and Subject/Absolutive agreement to positions outside of Tense (or is fused with it).

Person agreement is usually inside number agreement, and if person and number agreement are on either side of the root, then person agreement is likely the prefix and number agreement the suffix.

I'd expect participle morphology to be outermost, and I'd also expect participles to lack Tense, Mood, Subj/Abs.Agr, and maybe even more morphemes.

These are all just generalizations, though. Experiment with other orders or with blending and otherwise complexifying morphological template slots.
This information about the order of subject/absolutive and object/ergative affixes really surprised me! I've always had the impression that these suffixes would follow the dominant order of subject and object, i.e. subject > object if the language is SO and object > subject if the language is OS. In hindsight, I was probably thinking about the agreement markers too much like clitics and too little like affixes.

Could you tell me which languages have object affixes near or fused with voice affixes, so that I can see what it looks like (especially the ergative ones!)?

In the context of an IE. conlang, considering that subject agreement is inherited and object agreement is a later development (with object pronouns fusing with an already conjugated verb), would it be okay for object agreement to come after subject agreement, and both to come after all other suffixes? Would there be any pressure to "normalize" this by analogy?
Languages:
:hrv: [:D], :bih: :srb: [;)], :eng: [:D], :fra: [:|], :lat: [:(], :deu: [:'(]

A linguistics enthusiast who would like to make a conlang, but can't decide what to call what.

- Tewanian languages
- Guide to Slavic accentuation

náʼoolkiłí
rupestrian
rupestrian
Posts: 15
Joined: 04 Apr 2019 20:06

Re: (L&N) Q&A Thread - Quick questions go here

Post by náʼoolkiłí » 12 Apr 2019 00:42

Zekoslav wrote:
11 Apr 2019 10:58
Could you tell me which languages have object affixes near or fused with voice affixes, so that I can see what it looks like (especially the ergative ones!)?
Well in the South Caucasian languages object agreement prefixes appear right outside of voice prefixes (though this characterization simplifies things a bit), and in Algonquian the "theme" signs (whose behavior is fairly complicated) have been very successfully analyzed as object agreement on Voice (e.g., by Will Oxford). Hmm... the ergative languages with polypersonal agreement I'm at least somewhat familiar with don't really have easily segmentable voice morphology (NW Caucasian, Basque), or it's on the wrong side of the verbal complex (Mayan). But maybe someone can correct me.
Zekoslav wrote:
11 Apr 2019 10:58
In the context of an IE. conlang, considering that subject agreement is inherited and object agreement is a later development (with object pronouns fusing with an already conjugated verb), would it be okay for object agreement to come after subject agreement, and both to come after all other suffixes? Would there be any pressure to "normalize" this by analogy?
Sounds totally reasonable to me. There are of course plenty of exceptions to the generalizations I mentioned. Navajo, for instance, is a notable case where subject agreement comes inside object agreement.

You could poke around this WALS chapter for some more inspiration. Section 3 talks about the ordering generalization explicitly.

Nachtuil
greek
greek
Posts: 522
Joined: 21 Jul 2016 00:16

Re: (L&N) Q&A Thread - Quick questions go here

Post by Nachtuil » 13 Apr 2019 21:38

Dormouse559 wrote:
11 Apr 2019 07:43
Creyeditor wrote:
08 Apr 2019 17:42
Assuming no noun class system means no gender/noun class system at all, this page gives you the statistics.
There is 1 languages with gender distinctions in the 3rd person singular and plural pronouns without a gender system (Carib). Also another 4 languages with gender distinction only in third person singular and no gender system (Burmese, Mandarin, Persian, Kilivila).
The inclusion of Mandarin is intriguing. To my knowledge, the third-person singular pronoun only distinguishes gender in writing, and the written distinction was itself introduced only recently.
Indeed. The inclusion of Persian too actually.

User avatar
Creyeditor
mongolian
mongolian
Posts: 4503
Joined: 14 Aug 2012 19:32

Re: (L&N) Q&A Thread - Quick questions go here

Post by Creyeditor » 13 Apr 2019 23:22

Nachtuil wrote:
13 Apr 2019 21:38
Dormouse559 wrote:
11 Apr 2019 07:43
Creyeditor wrote:
08 Apr 2019 17:42
Assuming no noun class system means no gender/noun class system at all, this page gives you the statistics.
There is 1 languages with gender distinctions in the 3rd person singular and plural pronouns without a gender system (Carib). Also another 4 languages with gender distinction only in third person singular and no gender system (Burmese, Mandarin, Persian, Kilivila).
The inclusion of Mandarin is intriguing. To my knowledge, the third-person singular pronoun only distinguishes gender in writing, and the written distinction was itself introduced only recently.
Indeed. The inclusion of Persian too actually.
You can leave a comment and correct/ask about data points in WALS. Would be good if you used sources though. They regularily update it.
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
eldin raigmore
korean
korean
Posts: 6387
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: (L&N) Q&A Thread - Quick questions go here

Post by eldin raigmore » 22 Apr 2019 22:55

Concerning gerunds and their participants, and possessive classifiers.
Suppose a language can make gerunds, like “killing”, out of monotransitive verbs, like “kill”.
Suppose it also has two genitives, like English’s <‘s> “Saxon genitive” ending and English’s <of> “Norman genitive” preposition.
In lots of languages that have gerunds, the participants (or at least the agents and patients) of the gerund are spoken in the genitive.
I know “genitive” isn’t universally synonymous with “possessive”.
But, would it be reasonable for a language with two ways to mark genitive, and some tendency toward possessive classifierness (or classifierity), to use one genitive to mark the agent of a transitive gerund, and use the other genitive to mark the patient?

For instance, maybe if I say “the gangster’s killing”, we know I mean the gangster in question killed (or will kill or is killing or might kill) someone,
But if I say “the killing of the gangster”, we know I mean someone killed (or will kill or is killing or might kill) the gangster?

I have the feeling that in English this sort of thing isn’t really reliable or mandatory unless both participants are spoken of.

If I say “Frank’s killing of Al”, I think L1 English speakers, at least of my ‘lect, would know I meant Frank killed Al.
But if I say “Frank’s killing” they might think I could mean some person or persons unknown killed Frank.

OTOH if I say “the killing of Al”, they’d expect I meant somebody killed Al. If they found out I meant a murder Al committed, they’d be surprised and think that was odd; though they might accept it as possibly correct.

It may depend in part on which verb is being gerunded.
Suppose it’s “paint”.
“Frank’s painting of Al” could mean Frank is the artist and Al is the model of the portrait.
(Or that Frank owns the portrait now; no idea who painted it.)
But I think “the painting of Frank” could mean I was talking about Frank’s entire body of work in this particular medium and artform.
Or, of course, Frank sat as the model for someone to paint this particular portrait.

—————

ObNatLang: Does anyone know of any language with two classes of possessive-case markers, in which participants of gerunds are in the genitive, where one type of possession-marker is the only kind used to show the agent of a transitive gerund, and the other type is the only kind used to show the patient of such a gerund?

—————

ObConLang:
Would this be a good idea in a conlang?
Has anyone already done it?

User avatar
Reyzadren
greek
greek
Posts: 466
Joined: 14 May 2017 10:39
Contact:

Re: (L&N) Q&A Thread - Quick questions go here

Post by Reyzadren » 23 Apr 2019 00:38

^This is simple in my griuskant conlang. However, it has no cases, and re-uses its trigger voice suffixes as process nouns instead.

:con: griuskant (without the conscript)

gaewyzsaerp
/'gewYzˤserp/
demon-kill
Demon killings. (Demons killed something. Noun in active state)

gaewyzsaerpanor
/'gewYzˤserpanɔr/
demon-kill-V-PASS-EB-IMP
Demon killings. (Something killed demons. Noun in passive state)

Discussing the other sentences that involve possessive affixes (no genitive case here though), gerunds, anti-escapes, statefunctions and "priority rules" etc are off-topic/irrelevant for what you are asking now, heh.
Last edited by Reyzadren on 23 Apr 2019 03:38, edited 1 time in total.
Image Soundcloud Profile | Image griuskant conlang

Salmoneus
MVP
MVP
Posts: 1665
Joined: 19 Sep 2011 19:37

Re: (L&N) Q&A Thread - Quick questions go here

Post by Salmoneus » 23 Apr 2019 02:22

This definitely happens in some languages with A/O possession - agents of verbal nouns take A-possession, patients take O-possession. I wouldn't be suprised if it happened in some alienability systems too (patients as inalienable, agents as alienable).

[well, I'm not certain the verbal nouns in question are definitely "gerunds", but...]

User avatar
eldin raigmore
korean
korean
Posts: 6387
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: (L&N) Q&A Thread - Quick questions go here

Post by eldin raigmore » 23 Apr 2019 23:38

Reyzadren wrote:
23 Apr 2019 00:38
^This is simple in my griuskant conlang. However, it has no cases, and re-uses its trigger voice suffixes as process nouns instead.
....
Discussing the other sentences that involve possessive affixes (no genitive case here though), gerunds, anti-escapes, statefunctions and "priority rules" etc are off-topic/irrelevant for what you are asking now, heh.
Well, yeah, maybe not relevant to what I was asking about; but deuced intriguing naetheless! Where can I see more?

.....
Salmoneus wrote:
23 Apr 2019 02:22
This definitely happens in some languages with A/O possession - agents of verbal nouns take A-possession, patients take O-possession. I wouldn't be suprised if it happened in some alienability systems too (patients as inalienable, agents as alienable).
[well, I'm not certain the verbal nouns in question are definitely "gerunds", but...]
This is certainly a pertinent and relevant and informative answer. I thank you for it.

It also whets my appetite for sidequests. Where/how do I find out about A/O possession?

I’m not sure alienable-vs-inalienable counts as “possessive classifiers”. Obviously it’s a related idea; and I suspect that if any natlangs have both inalienable-vs-alieanable possession, and possessive classifiers, they’d be all up in each other’s business, so to speak.

I also worried about whether everything I wanted to ask about strictly fit the formal or technical definition of “gerund”. I didn’t mention it in my earlier post out of not wanting to put off any readers with “too many weasel-words” or whatever.

But I was interested in any nounlike nonfinite verbform which cannot take a nominative or ergative agent or an absolutive or accusative patient; but can take an agent as long as it’s genitive, and/or can take a patient as long as it’s genitive.

Such a nominalized verb, or verbal noun, or deverbal noun, might be a masdar or verbnoun or something instead of a gerund. I’d be interested anyway.
So I’m glad you just went ahead and told me the interesting stuff anyway!

(I don’t think infinitives are the kind of thing I was thinking of. as i inderstand it they can’t take an agent even if it’s genitive; and can take an absolutive or accusative object even if it’s not genitive. But I could be wrong, and if I am, I want to hear about those, too!)

—————

Thanks again to both reyzadren and salmoneus!

User avatar
Reyzadren
greek
greek
Posts: 466
Joined: 14 May 2017 10:39
Contact:

Re: (L&N) Q&A Thread - Quick questions go here

Post by Reyzadren » 24 Apr 2019 01:10

eldin raigmore wrote:
23 Apr 2019 23:38
Well, yeah, maybe not relevant to what I was asking about; but deuced intriguing naetheless! Where can I see more?
As I don't know which facet of distinguishing subjects/objects is relevant to your interests, you can copy all previous related sentences "Frank's painting of Al" etc, and just ask how they would be translated in the griuskant conlang thread, or just a general question.

Else, one can see more examples in my conlang textbook, but these grammatical units are not in 1 specific chapter; rather they are scattered across multiple chapters though.
Image Soundcloud Profile | Image griuskant conlang

Salmoneus
MVP
MVP
Posts: 1665
Joined: 19 Sep 2011 19:37

Re: (L&N) Q&A Thread - Quick questions go here

Post by Salmoneus » 25 Apr 2019 01:23

A/O is a property of Polynesian languages - so called because in some languages, like Hawai'ian, there are two possessive prepositions, a and o. Conveniently enough, "a" possession seems associated with agents and "o" with objects. That is, "a" marks possessors who have high levels of control, dominance, or initiation, while "o" marks possessors who have low levels of control, who are subordinate, and who are passive. So "post a eldin" would be eldin's post - because you chose to make the post - whereas "voice o eldin", "finger o eldin" or "confusion o eldin" would be used because those things, while 'belonging' to you, are not yours by choice.

Similar things happen elsewhere in Oceanic. Blust gives an example from a New Britain language where the two forms take different classifiers, distinguishing "my story" (the story I tell) and "my story" (the story told about me). Interestingly, he also mentions a language in which the morphemes are the same, but the same distinction is made by morpheme order (the possessive affix is a suffix for one meaning and a prefix for the other).

User avatar
Omzinesý
runic
runic
Posts: 2673
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (L&N) Q&A Thread - Quick questions go here

Post by Omzinesý » 25 Apr 2019 16:36

What languages other than Celtic have word-initial consonant mutations expressing grammatical meanings? Celtic langs cannot be that unique it only appears in them.

Salmoneus
MVP
MVP
Posts: 1665
Joined: 19 Sep 2011 19:37

Re: (L&N) Q&A Thread - Quick questions go here

Post by Salmoneus » 25 Apr 2019 17:31

Omzinesý wrote:
25 Apr 2019 16:36
What languages other than Celtic have word-initial consonant mutations expressing grammatical meanings? Celtic langs cannot be that unique it only appears in them.
The classic answer here would be "Atlantic, Mande and Senufo" (three nearby and possibly related familes of northwest africa); the overlooked answer would be "Romance languages of Italy". Other well-known answers include "Nuer" (and related nilotic languages) and "various languages of Central Vanuatu". Google tells me that some Uto-Aztecan languages also have mutation, and that a large number Bantu languages do in practice have mutation though Bantuists don't think of it that way*. And there's a bunch of languages that have more borderline cases, like mutation to mark causatives (which might not be quite as productive, so is it inflectional mutation, or is it just derivation?).

*Bantu languages have lots of prefixes, but in practice sometimes the prefixes aren't. So there's a 'nasal prefix', which in most languages is exactly that, but actualy in a bunch of language what it is in practice is nasal mutation. [that is, m+b should be mb, but in practice it may actually just be 'm', so you've got a b->m mutation]

User avatar
Creyeditor
mongolian
mongolian
Posts: 4503
Joined: 14 Aug 2012 19:32

Re: (L&N) Q&A Thread - Quick questions go here

Post by Creyeditor » 25 Apr 2019 21:07

Omzinesý wrote:
25 Apr 2019 16:36
What languages other than Celtic have word-initial consonant mutations expressing grammatical meanings? Celtic langs cannot be that unique it only appears in them.
At our university we are working on a mutation database. If you just want to know about initial consonant mutation, I can PM you a list of language (with sources) that do that. It's a lot, especially in the Austronesian family, but also in all other places of the world, except maybe Australia and South America.
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]

Nachtuil
greek
greek
Posts: 522
Joined: 21 Jul 2016 00:16

Re: (L&N) Q&A Thread - Quick questions go here

Post by Nachtuil » 26 Apr 2019 03:00

Creyeditor wrote:
13 Apr 2019 23:22
You can leave a comment and correct/ask about data points in WALS. Would be good if you used sources though. They regularily update it.
That's good to know. I'll try to when I get some more time.

User avatar
eldin raigmore
korean
korean
Posts: 6387
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: (L&N) Q&A Thread - Quick questions go here

Post by eldin raigmore » 28 Apr 2019 00:49

1. What fraction of the world’s languages have several syllable-onset consonant-clusters containing 5 or more consonants?
1a. What are a few such languages?
1b. What are a few such long onset-clusters?

2. what fraction of the world’s languages have several syllable-coda consonant clusters containing 5 or more consonants?
2a. What are a few such languages?
2b. What are a few such long coda-clusters?

3. What fraction of the world’s languages have both several syllable-onset consonant clusters with at least 4 consonants, and also have several syllable-coda consonant clusters with at least 4 consonants?
3a. What are a few such languages?
3b. What are a few syllables in each of those languages, illustrating several different combinations of a 4(-or-more)-consonant onset-cluster, and a 4(-or-more)-consonant coda-cluster, on the same syllable? I’d prefer two or more distinct long onsets as well as two or more distinct long codas —— if both exist, of course.

4a does any language have any consonantal syllables with consonantal nuclei for which said nucleus is a cluster of 2 (or more) consonants?
4b does any language have any consonantal syllables whose nucleus is consonant, for which the onset is a cluster of 2 or more consonants?
4c does any language have any consonantal syllables whose nucleus is a consonant, for which the coda is a cluster of 2 or more consonants?
4d does any language have any consonantal syllables whose nucleus is a consonant, and whose onset is a(t least one) consonant, and whose coda is a(t least one) consonant?
Spoiler:
5. Some grammarians or linguists or phonologists, describe some northwestern Native North American languages, as having syllables that have no nuclei at all, but do have both an onset and a coda; in some cases either the onset or the coda is a cluster. Others say these phoneme-groups aren’t syllables. How the hell do they get away with such blatant violations of Occam’s razor and of the definition of “syllable” as “that which is taken together”? They must have some reasoning by which the data, in their opinions, justify such departures from the norm; but I’ve never seen anything but “I’m a full professor so whatever I say must be true”, and its junior offspring, “I’m trying to earn a PhD and my professor thinks this so I better not contradict him/her in this dissertation”.
Thanks, to anyone who contributes anything!

User avatar
Pabappa
sinic
sinic
Posts: 269
Joined: 18 Nov 2017 02:41
Contact:

Re: (L&N) Q&A Thread - Quick questions go here

Post by Pabappa » 28 Apr 2019 05:08

The English word splurged can be analyzed as CCCCCCC if you call the vowel a syllabic consonant and analyze the g as /dž/.

Quoting the whole post below because I bumped:
eldin raigmore wrote:
28 Apr 2019 00:49
1. What fraction of the world’s languages have several syllable-onset consonant-clusters containing 5 or more consonants?
1a. What are a few such languages?
1b. What are a few such long onset-clusters?

2. what fraction of the world’s languages have several syllable-coda consonant clusters containing 5 or more consonants?
2a. What are a few such languages?
2b. What are a few such long coda-clusters?

3. What fraction of the world’s languages have both several syllable-onset consonant clusters with at least 4 consonants, and also have several syllable-coda consonant clusters with at least 4 consonants?
3a. What are a few such languages?
3b. What are a few syllables in each of those languages, illustrating several different combinations of a 4(-or-more)-consonant onset-cluster, and a 4(-or-more)-consonant coda-cluster, on the same syllable? I’d prefer two or more distinct long onsets as well as two or more distinct long codas —— if both exist, of course.

4a does any language have any consonantal syllables with consonantal nuclei for which said nucleus is a cluster of 2 (or more) consonants?
4b does any language have any consonantal syllables whose nucleus is consonant, for which the onset is a cluster of 2 or more consonants?
4c does any language have any consonantal syllables whose nucleus is a consonant, for which the coda is a cluster of 2 or more consonants?
4d does any language have any consonantal syllables whose nucleus is a consonant, and whose onset is a(t least one) consonant, and whose coda is a(t least one) consonant?
Spoiler:
5. Some grammarians or linguists or phonologists, describe some northwestern Native North American languages, as having syllables that have no nuclei at all, but do have both an onset and a coda; in some cases either the onset or the coda is a cluster. Others say these phoneme-groups aren’t syllables. How the hell do they get away with such blatant violations of Occam’s razor and of the definition of “syllable” as “that which is taken together”? They must have some reasoning by which the data, in their opinions, justify such departures from the norm; but I’ve never seen anything but “I’m a full professor so whatever I say must be true”, and its junior offspring, “I’m trying to earn a PhD and my professor thinks this so I better not contradict him/her in this dissertation”.
Thanks, to anyone who contributes anything!
Sorry guys, this one has the worst sting.

Post Reply