Odd natlang features thread

The best topics from Languages & Linguistics, kept on a permanent basis.
Post Reply
User avatar
Whimemsz
Avisaru
Avisaru
Posts: 690
Joined: Fri Jun 20, 2003 4:56 pm
Location: Gimaamaa onibaaganing

Re: Odd natlang features thread

Post by Whimemsz »

Ah, so you did end up getting the grammar, cool!

I'm not sure how common it is. Igbo is often described as having a small closed class of adjectives, although I found a thing a while back that argued the issue was more complicated than that (sadly, I don't remember the details). Cubeo also has a small closed class of adjective roots (ɨra- "big/large"; kɨhĩ- "small"; bãbã- "new/young"; bɨkɨ- "old/great"; bẽa- "good/beautiful"; and ãbẽ- "bad/ugly"). Other verbs can function as adjectives, though, and the adjective roots themselves can be inflected either as nouns or with both nominal and verbal morphology. [source: http://etnolinguistica.wdfiles.com/loca ... _kubeo.pdf]

User avatar
Chagen
Avisaru
Avisaru
Posts: 707
Joined: Thu Sep 22, 2011 11:54 pm

Re: Odd natlang features thread

Post by Chagen »

Xavánte lacks velars: /p t tʃ ʔ/, /b d dʒ/, /w ɾ h/.
/w/ is a velar consonant.
Nūdhrēmnāva naraśva, dṛk śraṣrāsit nūdhrēmanīṣṣ iźdatīyyīm woḥīm madhēyyaṣṣi.
satisfaction-DEF.SG-LOC live.PERFECTIVE-1P.INCL but work-DEF.SG-PRIV satisfaction-DEF.PL.NOM weakeness-DEF.PL-DAT only lead-FUT-3P

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: Odd natlang features thread

Post by ---- »

It could have been written /w/ for convenience, I know in Yeli Dnye there's a phoneme transcribed /w/ a lot of the time but it's allegedly [ð͡β].

User avatar
Ser
Smeric
Smeric
Posts: 1542
Joined: Sat Jul 19, 2008 1:55 am
Location: Vancouver, British Columbia / Colombie Britannique, Canada

Re: Odd natlang features thread

Post by Ser »

I remember some ZBBer once asking about an example where historically two phonemes alternated. As if /i/ became /e/ and /e/ became /i/.

Here's an example from Latin > Spanish:
  • Latin /eː/ and /ɛ/ become Spanish /i/ in racemum [raˈkeːmũ] > racimo, sella [ˈsɛlːa] > Old Spanish siella > silla
  • On the other hand, Latin /iː/ becomes Spanish /e/ in dixi [ˈdiːksiː] > dije [ˈdixe]
This is most conspicuous in what happened with Latin ueni [ˈweːniː], where both alternated and became Spanish vine [ˈbine, β̞-].
Chagen wrote:/w/ is a velar consonant.
Tieđđá wrote:It could have been written /w/ for convenience, I know in Yeli Dnye there's a phoneme transcribed /w/ a lot of the time but it's allegedly [ð͡β].
Or maybe whoever is claiming that is analyzing all occurrences of /w/ as diphthongs.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Odd natlang features thread

Post by Nortaneous »

Speaking of strange semivowels, here's Tsou:
Wikipedia wrote:The approximants /w/ and /j/ may surface as non-syllabic mid vowels [e̯] and [o̯], even (for /j/) in initial position (/jo~joskɨ/ [e̯oˈe̯oskɨ] "fishes"; /w/ does not occur in initial position), explaining the spelling Tfuea (/tfuja/) for the name of the dialect. However, stress assignment ([ˈtfue̯a]) and restrictions on consonant clusters (see stress and phonotactics below) demonstrate that they behave as consonants.
And Manambu allophonic prenasalization disharmony:
All the voiced stops and the voiced affricate j share one feature: they are prenasalized in the
word-initial, intervocalic, and word-final position; that is, bal ‘pig’ is pronounced as [mbal], ya:b
‘road’ as [ya:mb], ab ‘head’ as [amb], abawapw i ‘headdress, hat’ as [ambawa"pwi], Juli ‘Julie’ as
[ndZ"uli] and so on. Word-initially, the voiced affricate j is not prenasalized if the next syllable
contains a nasal or another prenasalized stop, e.g. jág@r ‘garfish’ is pronounced as ["dZaNg@r ]
and not as ∗ ["ndZaNg@r].
The velar stop g is not prenasalized word-initially if either the coda or the next syllable
contain a nasal consonant. That is, ga:m ‘song, shout’ is pronounced as [ga:m] and not
as ∗ [Nga:m], and ga:n ‘night’ is pronounced as [ga:n] rather than ∗ [Nga:n]. The initial g is
prenasalized very little if there is a nasal consonant in the next but one syllable within one
phonological word, e.g. [Ngamb@"ma:dZ] (old-lk-story) ‘story’ and [(N)ga"ndZi-n] ‘rubbing’. If
a nasal consonant is further away from the first syllable containing g, it does not affect the
degree of its prenasalization, e.g. [Ngan dZi-mba"na] ‘we rub (something)’. In a word-medial and
word-final position, g is always prenasalized, e.g. [ku"Ngamb] ‘owl’, [war"Ngamb] ‘a Manambu
clan’, [waraN ga"ndu] ‘ancestor’. The presence of another prenasalized stop in a word does not
affect the prenasalization of the word-initial g in [NgandZi-"tua] ‘I rub (something)’, ["Nga:ndZ]
‘small pelican-like white bird’, and [Nga"mbi] ‘creeper plant similar to vanilla’. This shows that
prenasalized stops are not underlyingly nasal.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Odd natlang features thread

Post by Nortaneous »

Gikuyu has ʃ (maybe from earlier *tʃ?) and ð but no s or z, and no other unvoiced fricatives but h. Also, it's a Bantu language with /ð/.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Odd natlang features thread

Post by Nortaneous »

Some dialects of Abkhaz have an ejective fricative (the only one in the inventory), fʼ, from earlier pʼ.
‮suoenatroN wrote:Gikuyu has ʃ (maybe from earlier *tʃ?) and ð but no s or z, and no other unvoiced fricatives but h. Also, it's a Bantu language with /ð/.
As it turns out, /ð/ is not uncommon elsewhere in the Bantu languages; some Southern Bantu languages have it and its unvoiced equivalent. And speaking of Southern Bantu languages, Venda contrasts all of /ɸ β f v pfʰ bv bʷ/, and then adds /sʷ zʷ/.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Odd natlang features thread

Post by linguoboy »

Vuvuzela wrote:Ju|'hoan (a grammar of which I received recently) has adjectives as a closed class. I'm not sure how unusual it is, but I find it interesting.
Also found in Korean. There's a handful of true adjectives (e.g. 새 /say/ "new", 참 /cham/ "true") and everything else is either a noun or "pre-noun"[*] modifier on the one hand or a descriptive verb in a relative construction on the other.

[*] Distinguished from other nouns by being bound morphemes and from adjectives by the fact that they can take the linking consonant written ㅅ but which surfaces only as another coronal (/n/ or /t/) or as reinforcement of the following consonant. Come to think of it, that's an odd natlang feature in itself.

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: Odd natlang features thread

Post by ---- »

‮suoenatroN wrote:Some dialects of Abkhaz have an ejective fricative (the only one in the inventory), fʼ, from earlier pʼ.
Abkhaz is also interesting because its consonant /tʷ/ (and presumably its voiced and ejective brethren) has a couple allophones that are cross-linguistically extremely rare. According to Wikipedia they include [tʙ̥] and [t͡p]. The first only appears in like 3 languages in the Amazon or something and I think the second only appears in like one language in the whole world.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Odd natlang features thread

Post by Nortaneous »

Contrastive preaspiration on every consonant, including the voiced ones. Thus ANADEWing one of my conlangs, unless I rearrange the phonology to not lose the implosive before preaspiration develops.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
Whimemsz
Avisaru
Avisaru
Posts: 690
Joined: Fri Jun 20, 2003 4:56 pm
Location: Gimaamaa onibaaganing

Re: Odd natlang features thread

Post by Whimemsz »

Kuikuro has a pretty neat phoneme inventory: /p t k dʲ ts s h m n ɲ ŋ l w/ (note that the only voiced stop--and only palatalized consonant--is /dʲ/) plus a uvular tap /ʀ̆/.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Odd natlang features thread

Post by Nortaneous »

Whimemsz wrote:a uvular tap /ʀ̆/.
...alternately described as a pharyngeal tap, and coming from /r/ through an intermediate step /l/.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Odd natlang features thread

Post by Nortaneous »

I forget if there's a separate thread for diachronics, but here's Burmese:
Tone *1 verbs with originally voiced initials developed creaky tone in the sentence-final, preparticle slot. Of course not all verb roots occurred exclusively or even primarily in this sentence-final slot; a number of verbal roots served not just in the sentence-final pre-particle slot as main verbs but also they served as adjectives when modifying a head noun and as nouns when they were nominalized. Thus what was originally a single tone *1 etymon developed two separate reflexes: the expected level toned reflex in non-final position and a creaky toned reflex in sentence-final position. ...

Notice that by and large the creaky toned forms have meanings consistent with their status as main verbs while their level toned counterparts have meanings consistent with their role as adjectives (stativity) or as nouns.
Examples + source here: http://www.csuchico.edu/~gthurgood/Pape ... hology.pdf
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

CatDoom
Avisaru
Avisaru
Posts: 739
Joined: Fri Sep 20, 2013 1:12 am

Re: Odd natlang features thread

Post by CatDoom »

Seri has fifty-four basic kinship terms. Many of these were probably compounds in the past, but irregular sound changes have resulted in them being synchronically unanalyzable. An example would be hicóome, "my younger sister (where the speaker is a man)," which contrasts with both hipáac, "my older sister (where the speaker is a man)" and hitcz, "my younger sister (where the speaker is a woman)."

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: Odd natlang features thread

Post by ---- »

In Navajo, "cough" and "think" share the same verb root.

Hadiskees
"I started coughing"

Ntséskees
"I'm thinking"

jmcd
Smeric
Smeric
Posts: 1034
Joined: Fri Mar 12, 2004 11:46 am
Location: Réunion
Contact:

Re: Odd natlang features thread

Post by jmcd »

Could they not just be homophones?

User avatar
Drydic
Smeric
Smeric
Posts: 1652
Joined: Tue Oct 08, 2002 12:23 pm
Location: I am a prisoner in my own mind.
Contact:

Re: Odd natlang features thread

Post by Drydic »

Eventually, isn't the distinction between homophones and two ideas sharing the same root or form down to speaker perception? Two sides of the same coin unless the speakers clearly demarcate between the two (which I rather suspect they do, in this case).
Image Image
Common Zein Scratchpad & other Stuffs! OMG AN ACTUAL CONPOST WTFBBQ

Formerly known as Drydic.

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: Odd natlang features thread

Post by ---- »

I guess they *could*, but the idea of them being 'actually' the same thing is more likely for theoretical reasons. Most verb roots are very general and abstract, and thematic prefixes are appended to the verb to specify it into different things. If you think of the verb root itself as applying to not either concrete meaning in particular but to some kind of action of the head that sends out a particular class of entity, then both 'think' and 'cough' could apply (a thought vs. a gust of air). Every author I've seen write on the subject seems to also insist that they are the same thing while other identical verb roots with apparently divergent meanings are 'different', so this might just be something I don't have enough knowledge to explain fully, or it could also be some kind of folk tradition about language like "English has the most difficult writing system" or "Polish has 500 genders".

I dunno, I just think it's cooler to think of them as extensions of the same idea. This sort of thing happening in Navajo has other examples but this one is the most striking.

User avatar
Aurora Rossa
Smeric
Smeric
Posts: 1138
Joined: Mon Aug 11, 2003 11:46 am
Location: The vendée of America
Contact:

Re: Odd natlang features thread

Post by Aurora Rossa »

"Polish has 500 genders"
Who the hell says that and what do they mean?
Image
"There was a particular car I soon came to think of as distinctly St. Louis-ish: a gigantic white S.U.V. with a W. bumper sticker on it for George W. Bush."

User avatar
Drydic
Smeric
Smeric
Posts: 1652
Joined: Tue Oct 08, 2002 12:23 pm
Location: I am a prisoner in my own mind.
Contact:

Re: Odd natlang features thread

Post by Drydic »

Poles, presumably.
And it refers to this:
Wikipedia/Polish grammar wrote:There are three main genders: masculine, feminine and neuter. Masculine nouns are also divided into animate and inanimate (this distinction being relevant in the singular), and personal and non-personal (this distinction being relevant in the plural). All personal nouns are also animate; some nouns with inanimate meaning (such as papieros "cigarette"; especially many loanwords relating to information technologies, e.g. komputer, walkman, blog) are treated grammatically as animate.
Image Image
Common Zein Scratchpad & other Stuffs! OMG AN ACTUAL CONPOST WTFBBQ

Formerly known as Drydic.

User avatar
Pole, the
Smeric
Smeric
Posts: 1606
Joined: Sat Feb 11, 2012 9:50 am

Re: Odd natlang features thread

Post by Pole, the »

Aurora Rossa wrote:
"Polish has 500 genders"
Who the hell says that and what do they mean?
It means that in each of 500 Polish cities and towns there is a gender monster indoctrinating our children into wearing skirts.
The conlanger formerly known as “the conlanger formerly known as Pole, the”.

If we don't study the mistakes of the future we're doomed to repeat them for the first time.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Odd natlang features thread

Post by Nortaneous »

One of the most distinctive traits of the phonological typology of NMC languages is the pervasive manner in which the features of vowels have become interdependent with those of adjacent consonants. Historically, vowel features have ‘smeared’ onto adjacent consonants, as in POC *Rumaq > PMC *imwa/umwa ‘house’, where the rounding of the vowel has been transferred to the labial consonant in most attested languages. While transfers of rounding from vowels to consonants occur in other Oceanic languages (as in Vanuatu), this generally is an isolated phenomenon. In Micronesia the interpenetration of vowel and consonant features is far more pervasive, and has undergone reanalysis so that synchronically it is sometimes simpler to argue that a transfer of phonological features is occurring not from vowels to consonants, but rather from consonants to vowels.
Also happened in NWC, but, you know.

NMC = Nuclear Micronesian, which includes Marshallese.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Odd natlang features thread

Post by Nortaneous »

13 consonants, ɸ~f contrast.

I'm guessing what happened there is that *b *g > ɸ ɣ.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Seirios
Sanci
Sanci
Posts: 24
Joined: Thu Feb 13, 2014 6:26 am
Location: Shenzhen, Guangdong, China

Re: Odd natlang features thread

Post by Seirios »

I'll keep updating

Ah I mistook the thread I was writing for...and gave some expectable features for you. Well, I'm terribly sorry for that. Perhaps providing the background would be better but it would be too lengthy.

A syllable in Chinese languages are composed of an initial (initial consonant) and a rime (everything else). The Fuzhou dialect of Mindong, in the Min language group of Chinese, with 46 rimes and 7 tones but only 2 consonantal codas /-ŋ/ and /-ʔ/, shifts all its rimes into a secondary pronunciation in three f its tones, in which only 2 pairs of original rimes merge repectively. This is regardless of all the tone sandhis, only relevant with the original tones.

A Chinese dialect in the Guangxi province has a initials group /ʔ/ /ʔʰ/ /h/.

I've met a Chinese language that has four contrasting tones 545 434 323 212.

A dialect of Min dissimilates "ling-gui" into "gui-gui".

Early Middle Chinese, with 161 rimes and 6 phonemic vowels, only has 9 rimes with /o/ and all of them are */uo/ (except a *[i̯ɯ] that was normally treated as */io/ since that's the only occurence of *[ɯ] in it), only standing with no ending or before /ŋ/ /k/, and 4 /u/ ,only before /ŋ/ /k/, where /u/ contrasts with /uo/, which contrast soon disappeared. */uŋ(/k)/ and */uoŋ(/k)/ was formerly respectively */oŋ(/k)/ and */uŋ(/k)/, while /iuŋ(/k)/ and /iuoŋ(/k)/ kept still. More than a half of its syllables have a glide /j/, and EMC contrasts */i/ /i̯˞i/ /i̯ɨ/ /i̯ɨi̯/ /i̯e/ /i̯˞e/. There is a heavy contrast between /a/ and /e/, as 49 rimes are with /a/ and 46 with /e/, many are minimal pairs.

Standard Mandarin has done a */ɲi/ /ɲi̯e/ /ɲi̯ɨ/ → /ɐɻ/

And I'm not that familiar with fields outside historical linguistics and phonology/phonetics, so I don't know if these count as odd.
11+ different words for a conjunction "and" in its parallel sense that function the same as "and", each with its nuance or special function within that meaning;
Adjectival "male" and "female", for animals other than human, share the same sound (in this language adjectives don't change their forms and virtually no lexical inherent distinction between sex in nouns exists like ox vs cow);
"To build" (lit. "to heighten") also means "to destroy", both apply to buildings perfectly and are in use;
The only personal pronoun is a third-person masculine singular, based on which the plural can be formed;
While this language has a normal active voice carried by active forms that functions roughly the same as in other normal languages, its intransitive verbs, like "to be silent" "to be happy" "to die", also have passive forms that are productive, carry passive voice and many of which are in use;

lol.
Last edited by Seirios on Fri Feb 21, 2014 5:11 am, edited 3 times in total.
Always an adventurer, I guess.
-
Tone: Chao's notation.
Apical vowels: [ɿ]≈[z̞̩], [ʅ]≈[ɻ̞̩], [ʮ]≈[z̞̩ʷ], [ʯ]≈[ɻ̞̩ʷ].
Vowels: [ᴇ]=Mid front unrounded, [ᴀ]=Open central unrounded, [ⱺ]=Mid back rounded, [ⱻ]=Mid back unrounded.

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: Odd natlang features thread

Post by Salmoneus »

Seirios wrote: Shanghainese has a four way contrast of /ɔ/ /o/ /ʊ/ /u/, has five tones but only two contrast only in unchecked syllables beginning with voiceless consonants.

Virtually all the Wu languages distinguish voiced sonorants from preglottalized/voiceless ones, and have complex large-scale tone sandhi with basic units as (compound) words/phrases.
Are those things odd?
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

Post Reply