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

A forum for discussing linguistics or just languages in general.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

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

Post by eldin raigmore »

All4Ɇn wrote: 08 Jun 2020 02:55I could be wrong here, but it totally seems to fit in line with the general pattern in Hungarian, which would geographically explain why you thought it was Polish [:D]
Thank you!
I’ll look into it. I hope you are right!
Even if you’re just sort of close-ish, you might put me on the right track!
Edit: https://core.ac.uk/download/pdf/35134811.pdf says Hungarian does have some word-initial three-consonant clusters.
It also says it has very few word-internal three-consonant clusters, but they are “very special”, and they all span syllable-boundaries.
I may be reading it wrong; but if not, it seems to me to be saying that word-final consonant clusters aren’t longer than two consonants.
So it looks like Hungarian is, indeed, close to what I was asking for!
User avatar
All4Ɇn
mayan
mayan
Posts: 1765
Joined: 01 Mar 2014 07:19

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

Post by All4Ɇn »

eldin raigmore wrote: 08 Jun 2020 05:17It also says it has very few word-internal three-consonant clusters, but they are “very special”, and they all span syllable-boundaries.
The only exceptions that fit under this that I can think of right now are the name István (which is a borrowing anyhow) and some verbal and nominal endings. What's interesting is some endings insert vowels to keep the pattern while others don't, so for instance: festhet (can paint) does irregularly have 3 internal consonants but festeni (to paint) adds an extra -e- in order to avoid having the 3 consonants next to each other. Virtually all of the examples I saw in the pdf seemed to be loanwords that haven't undergone much nativization. The more I read about this, the more István comes out as a particularly peculiar word. Not only is it far more nativized than the other borrowed terms, it was originally Istefán which fits perfectly in line with Hungarian phonology but for whatever reason is now István!
User avatar
Xonen
moderator
moderator
Posts: 1080
Joined: 16 May 2010 00:25

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

Post by Xonen »

eldin raigmore wrote: 08 Jun 2020 05:17
Edit: https://core.ac.uk/download/pdf/35134811.pdf says Hungarian does have some word-initial three-consonant clusters.
I'm quite sure most European languages have those by now... Kind of inevitable when people are highly literate, orthographies are mostly phonemic and new scientific and technological concepts keep getting names derived from Latin and Greek (and pop culture is mostly in English); it's not like "szklerózis" is a particularly typical word under Hungarian phonotactics, but most speakers these days will nonetheless be able to pronounce it exactly as written. Then again, if we exclude recent(-ish) loanwords, Hungarian doesn't really have word-initial clusters at all, AFAIU.
User avatar
All4Ɇn
mayan
mayan
Posts: 1765
Joined: 01 Mar 2014 07:19

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

Post by All4Ɇn »

Xonen wrote: 08 Jun 2020 18:01Then again, if we exclude recent(-ish) loanwords, Hungarian doesn't really have word-initial clusters at all, AFAIU.
This does in fact seem to be the case too. It's particularly noticeable in the word Görög (Greek). The number of common loanwords with CC initials is however still much higher than those with CCC anywhere in the word.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

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

Post by eldin raigmore »

Xonen wrote: 08 Jun 2020 18:01
eldin raigmore wrote: 08 Jun 2020 05:17
Edit: https://core.ac.uk/download/pdf/35134811.pdf says Hungarian does have some word-initial three-consonant clusters.
I'm quite sure most European languages have those by now... Kind of inevitable when people are highly literate, orthographies are mostly phonemic and new scientific and technological concepts keep getting names derived from Latin and Greek (and pop culture is mostly in English); it's not like "szklerózis" is a particularly typical word under Hungarian phonotactics, but most speakers these days will nonetheless be able to pronounce it exactly as written. Then again, if we exclude recent(-ish) loanwords, Hungarian doesn't really have word-initial clusters at all, AFAIU.
The paper was concentrating on native (i.e. non-borrowed) non-compound words.
User avatar
Xonen
moderator
moderator
Posts: 1080
Joined: 16 May 2010 00:25

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

Post by Xonen »

All4Ɇn wrote: 08 Jun 2020 20:11
Xonen wrote: 08 Jun 2020 18:01Then again, if we exclude recent(-ish) loanwords, Hungarian doesn't really have word-initial clusters at all, AFAIU.
This does in fact seem to be the case too. It's particularly noticeable in the word Görög (Greek).
There's also words like iskola 'school' - or indeed the aforementioned István – which've kept the cluster but made it non-initial by inserting a prothetic vowel in front.


eldin raigmore wrote: 08 Jun 2020 21:27
Xonen wrote: 08 Jun 2020 18:01
eldin raigmore wrote: 08 Jun 2020 05:17
Edit: https://core.ac.uk/download/pdf/35134811.pdf says Hungarian does have some word-initial three-consonant clusters.
I'm quite sure most European languages have those by now... Kind of inevitable when people are highly literate, orthographies are mostly phonemic and new scientific and technological concepts keep getting names derived from Latin and Greek (and pop culture is mostly in English); it's not like "szklerózis" is a particularly typical word under Hungarian phonotactics, but most speakers these days will nonetheless be able to pronounce it exactly as written. Then again, if we exclude recent(-ish) loanwords, Hungarian doesn't really have word-initial clusters at all, AFAIU.
The paper was concentrating on native (i.e. non-borrowed) non-compound words.
Um, which paper? At least the one you're linking to here explicitly says this (page 42):
All the words which begin with consonant clusters are loan words
User avatar
Parlox
greek
greek
Posts: 495
Joined: 10 Feb 2017 20:28
Location: Ehh

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

Post by Parlox »

So I've been working on a little conlang alongside my big one, named Celos. It mandatorily marks nouns for person. My question is, do any natlangs do this? When I searched up "person marking on nouns" the closest I got was a WALS chapter on person-marking adpositions.
:con: Gândölansch (Gondolan)Feongkrwe (Feongrkean)Tamhanddön (Tamanthon)Θανηλοξαμαψⱶ (Thanelotic)Yônjcerth (Yaponese)Ba̧supan (Basupan)Mùthoķán (Mothaucian) :con:
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus »

Person marking on nouns is definitely a thing. The famous example is Sumerian.

But I don't know whether obligatory person marking on nouns is a thing. How obligatory are you talking? If it's just "1st and 2nd are obligatorily marked when the deixis is overt [i.e. not when 3rd peson periphrastic address is being used]", I could go with that... but if there's no zero marking for 3rd person, that seems like a LOT (95% of nouns being obligatorily marked with the same non-zero suffix?).
User avatar
Parlox
greek
greek
Posts: 495
Joined: 10 Feb 2017 20:28
Location: Ehh

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

Post by Parlox »

Salmoneus wrote: 09 Jun 2020 01:48 Person marking on nouns is definitely a thing. The famous example is Sumerian.

But I don't know whether obligatory person marking on nouns is a thing. How obligatory are you talking? If it's just "1st and 2nd are obligatorily marked when the deixis is overt [i.e. not when 3rd peson periphrastic address is being used]", I could go with that... but if there's no zero marking for 3rd person, that seems like a LOT (95% of nouns being obligatorily marked with the same non-zero suffix?).
In Celos nouns always mark for the 1st and 2nd person, while the 3rd person is assumed. It isn't obligatory per say, but if a speaker were to say "The man is me, I killed him" without a person marker it'd literally be "The man killed him". Of course this could be fixed simply with pronouns, but whats the point when you can indicate the person of a noun with one simple marker.

Basically, it's not mandatory but generally expected and not doing it can lead to confusion among speakers.
:con: Gândölansch (Gondolan)Feongkrwe (Feongrkean)Tamhanddön (Tamanthon)Θανηλοξαμαψⱶ (Thanelotic)Yônjcerth (Yaponese)Ba̧supan (Basupan)Mùthoķán (Mothaucian) :con:
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus »

Important edit: huh. Apparently... now... Sumerian didn't have person marking on nouns? Just plain old possessives? I could have sworn I'd read repeatedly that it had person marking...
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus »

Parlox wrote: 09 Jun 2020 01:53
Salmoneus wrote: 09 Jun 2020 01:48 Person marking on nouns is definitely a thing. The famous example is Sumerian.

But I don't know whether obligatory person marking on nouns is a thing. How obligatory are you talking? If it's just "1st and 2nd are obligatorily marked when the deixis is overt [i.e. not when 3rd peson periphrastic address is being used]", I could go with that... but if there's no zero marking for 3rd person, that seems like a LOT (95% of nouns being obligatorily marked with the same non-zero suffix?).
In Celos nouns always mark for the 1st and 2nd person, while the 3rd person is assumed. It isn't obligatory per say, but if a speaker were to say "The man is me, I killed him" without a person marker it'd literally be "The man killed him". Of course this could be fixed simply with pronouns, but whats the point when you can indicate the person of a noun with one simple marker.

Basically, it's not mandatory but generally expected and not doing it can lead to confusion among speakers.
This seems like something that could absolutely exist... but, per my last post, I can no longer confirm it. Anyone else have an example of a language where 'nominal person marking' doesn't just mean a possessive?
User avatar
Pabappa
greek
greek
Posts: 577
Joined: 18 Nov 2017 02:41

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

Post by Pabappa »

Sal, you may be thinking of Elamite, not Sumerian. Elamite has person marking on animate nouns. It also has pronouns .... I believe you have to still use a pronoun even when youre marking the noun ... e.g. I dont think you can say "doctor-1p pill-ACC prescribe-PAST" for "I, the doctor, prescribed you a pill".
Parlox wrote: 09 Jun 2020 01:06 So I've been working on a little conlang alongside my big one, named Celos. It mandatorily marks nouns for person. My question is, do any natlangs do this? When I searched up "person marking on nouns" the closest I got was a WALS chapter on person-marking adpositions.
My main conlang, Poswa, also marks person on nouns and I can assure you it works just fine. What works for me may not work for you though, because the grammar of Poswa is different in many other ways as well. For one thing, the person-marked nouns are mostly used as standalone sentences, e.g. from the stem /sittut-/ "doctor" I can say Sittuto! "I'm a doctor".

A less common use of person-marked nouns in Poswa is to use them as the head of a clause, e.g. sittuto pappabosebebi "I, a doctor, prescribed you a pill". Poswa needs to be able to this because it entirely lacks pronouns and also entirely lacks relativizing conjunctions such as English "that/which/what" etc. So Poswa is not like Elamite at all, and following my lead may not make sense if youre not planning to reject or at least minimize the use of pronouns.

If you decide to do this, you can still *also* have a second set of person markers on nouns to indicate possession. e.g. to use Poswa again, because its all I know, sittufo means "my doctor" because the stem change of /t/ > /f/ indicates that this 2nd set of person markers is being used. For some nouns, the two sets of person markers are the same, e.g. žopo can be parsed as meaning either "I am a limestone" or "my limestone", but since the first sentence will only ever be used metaphorically, this is no problem. It happens that most of the nouns where the two sets of person markers collide are inanimates, and therefore there is little confusion to be had.
Kavunupupis, šiŋuputata.
When I see you pointing at me, I know I'm in trouble. (Play)
User avatar
Sequor
sinic
sinic
Posts: 438
Joined: 30 Jun 2012 06:13

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

Post by Sequor »

Parlox wrote: 09 Jun 2020 01:53In Celos nouns always mark for the 1st and 2nd person, while the 3rd person is assumed. It isn't obligatory per say, but if a speaker were to say "The man is me, I killed him" without a person marker it'd literally be "The man killed him". Of course this could be fixed simply with pronouns, but whats the point when you can indicate the person of a noun with one simple marker.
So basically, something along the lines of 'this man killed him' to say 'I killed him', where ""this"" is a 1st person affix marker... I don't know about natlang attestations, but I like it.
Pabappa wrote: 09 Jun 2020 02:54 Sal, you may be thinking of Elamite, not Sumerian. Elamite has person marking on animate nouns. It also has pronouns .... I believe you have to still use a pronoun even when youre marking the noun ... e.g. I dont think you can say "doctor-1p pill-ACC prescribe-PAST" for "I, the doctor, prescribed you a pill".
Do you know details on how that person marking is used in Elamite?
hīc sunt linguificēs. hēr bēoþ tungemakeras.
User avatar
Parlox
greek
greek
Posts: 495
Joined: 10 Feb 2017 20:28
Location: Ehh

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

Post by Parlox »

Ser wrote: 09 Jun 2020 03:21 So basically, something along the lines of 'this man killed him' to say 'I killed him', where ""this"" is a 1st person affix marker... I don't know about natlang attestations, but I like it.
Exactly this! One could say something like...

Lhu dyrosofon swí mhwno.
Lhu (D/t)yros-of-on sw-í mhwn-o.
DEF man-ERG-1ST 3RD.MASC-ACC kill-PAST
The man (me) killed him.

On a related note, does anyone know of any natlangs with a "split-nominative" alignment. Essentially a clause where the subject is active receives the "ergative" case if transitive, and the "intransitive" case if intransitive. But if either of these clause-types is passive it takes a "nominative" case.
:con: Gândölansch (Gondolan)Feongkrwe (Feongrkean)Tamhanddön (Tamanthon)Θανηλοξαμαψⱶ (Thanelotic)Yônjcerth (Yaponese)Ba̧supan (Basupan)Mùthoķán (Mothaucian) :con:
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

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

Post by eldin raigmore »

Xonen wrote: 09 Jun 2020 00:12
eldin raigmore wrote: 08 Jun 2020 21:27 The paper was concentrating on native (i.e. non-borrowed) non-compound words.
Um, which paper? At least the one you're linking to here explicitly says this (page 42):
All the words which begin with consonant clusters are loan words
That’s the right paper! I must have missed the line you quoted. [:S] [:$] [:3]
User avatar
cedh
MVP
MVP
Posts: 386
Joined: 07 Sep 2011 22:25
Location: Tübingen, Germany
Contact:

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

Post by cedh »

Parlox wrote: 09 Jun 2020 03:34
Ser wrote: 09 Jun 2020 03:21 So basically, something along the lines of 'this man killed him' to say 'I killed him', where ""this"" is a 1st person affix marker... I don't know about natlang attestations, but I like it.
Exactly this! One could say something like...

Lhu dyrosofon swí mhwno.
Lhu (D/t)yros-of-on sw-í mhwn-o.
DEF man-ERG-1ST 3RD.MASC-ACC kill-PAST
The man (me) killed him.
A language which has person marking on nouns is Nahuatl. The wikipedia article states:
[...] any noun can function as a standalone predicate. For example, calli is commonly translated "house" but could also be translated "(it) is a house".

As predicates, nouns can take the verbal subject prefixes (but not tense inflection). Thus, nitēuctli means "I am a lord" with the regular first person singular subject ni- attached to the noun tēuctli "lord". Similarly tinocihuāuh means "you are my wife", with the possessive noun nocihuāuh "my wife" [which is formed from the noun cihuāuh plus the first person possessive prefix no- "my"] attached to the subject prefix ti- "you" (singular).
What I don't know though is whether nouns with overt 1st or 2nd person marking can be used as ordinary noun phrases in a sentence (as in your example), or whether they are only used predicatively.
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

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

Post by Salmoneus »

Pabappa wrote: 09 Jun 2020 02:54 Sal, you may be thinking of Elamite, not Sumerian. Elamite has person marking on animate nouns.
Oh, right, yes, Elamite. Damnit. So close!

Looking into it, it seems that nobody really knows what the hell is going on with Elamite. It has this person marking, both on heads and on modifiers (agreement) - indeed, it may primarily be there to parse phrases. But sometimes not on heads and only on modifiers. This may be syntactic, or lexical (indeclinable nouns?). And sometimes there's double agreement, for no apparent reason, and sometimes plural person marking on an inanimate. And the same markers are on some verbs, if indeed they are verbs. Is there suffixaufnahme? Who knows? Is it related to definiteness? And it's used as a derivational process - is this the same as person marking, or different? It seems like everyone has guessed a slightly different explanation for what is going on - but we don't have enough data to work it out.


Now that cedh's mentioned Nahuatl: oh, yeah, there's a bunch of NA languages in which most if not all 'nouns' look a lot like verbs, including potential person marking.



From reading a little about Elamite, here's some possible routes that come to mind for how to get "person marking" on nouns...


1. They're verbs

Verbs often have person marking. So if there's widespread use of verbs as arguments, a lot of your arguments will have person marking - but not necessarily other verbal marking (because they're not the main verb of a clause). So "I, the king" is 'actually' "I am kinging". Perhaps a few 'true' nouns lack person marking?


2. They're deictics

Imagine that the deictics 'this' and 'that' get stuck onto their nouns. Then we get "This king" (I, the king) and "that king" (you, the king). But perhaps the deictic can also have a more literal sense (if, for instance, you're actually holding a king, maybe 'this king' refers to them, not you)

3. They're inalienable possessives

If we imagine that most nouns are a bit more abstract in core meaning, there's not that much difference between "I, the king, command it" and "My kingly status commands it". The complication is that these possessives agree in person with the possessum, not the possessor - but that is attested.

4. They're copulas

Or rather, the copula is zero, but is marked for person, and can attach as a clitic to a noun. So "I, the king" really translates to "I am the king".



Just a few ideas, anyway.
User avatar
Pabappa
greek
greek
Posts: 577
Joined: 18 Nov 2017 02:41

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

Post by Pabappa »

Are there languages in which the equivalent of "my today's breakfast" would be grammatical? It seems to be a common mistake among English learners, perhaps including young native speakers. The only way to properly express this in English is to take one of the modifiers out of the clause and either say "my breakfast (for) today" or, less commonly, "today's breakfast for me". Or else reword the sentence entirely and have "the breakfast I ate today".

But it seems like such a simple construction .... I'd expect there must be at least some languages that allow it. Ideally from an inflecting language that uses the genitive inflection on both the 1st person pronoun and the word "today", such that the two words fulfill identical roles in the sentence.
Kavunupupis, šiŋuputata.
When I see you pointing at me, I know I'm in trouble. (Play)
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

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

Post by Creyeditor »

German allows the construction but I don't know if it's completely parallel. You can either say (1) or (2) without much difference in meaning:

(1) Mein Frühstück heute
my breakfast today
`my breakfast today'

(2) Mein heut-iges Frühstück
my today-ADJ breakfast
`my breakfast today'

Indonesian allows something similar:

(3) Sarapan=ku hari ini
breakfast=1SG.POSS day this
`my breakfast today'
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
cedh
MVP
MVP
Posts: 386
Joined: 07 Sep 2011 22:25
Location: Tübingen, Germany
Contact:

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

Post by cedh »

Creyeditor wrote: 12 Jun 2020 00:10 German allows the construction but I don't know if it's completely parallel. You can either say (1) or (2) without much difference in meaning:

(1) Mein Frühstück heute
my breakfast today
`my breakfast today'

(2) Mein heut-iges Frühstück
my today-ADJ breakfast
`my breakfast today'
I don't think this is completely parallel, because the suffix -ig is a derivational adjectivalizer and not a genitive. It's more or less the equivalent of a hypothetical English "my today-y breakfast".

Google Translate gives kyō no watashi no chōshoku as the Japanese equivalent. I'm not an expert for Japanese, but AFAICT this is indeed a double genitive construction that would be glossed as "today=GEN 1SG=GEN breakfast" or similar.
Post Reply