Conlang Random Thread

[Xwtek]

How do you change the genitive construction so this construction.

warrior-SG.GEN axe-SG.DIR

Turns into:

warrior-SG.GEN 3SG.SG.GEN axe-SG.DIR

[k1234567890y]

How do you change the genitive construction so this construction.

warrior-SG.GEN axe-SG.DIR

Turns into:

warrior-SG.GEN 3SG.SG.GEN axe-SG.DIR

I guess this:
https://en.wikipedia.org/wiki/His_genitive

Also, the Classical Chinese possessor marker 之 was actually a 3rd person pronoun, and is the precedent of modern Chinese possessive postposition 的, but the meaning of personal pronoun has lost in modern Chinese.

[akam chinjir]

Also, the Classical Chinese possessor marker 之 was actually a 3rd person pronoun, and is the precedent of modern Chinese possessive postposition 的, but the meaning of personal pronoun has lost in modern Chinese.

But there's not really any reason to think of 之 or 的 as possession-specific, is there? They're used with attributive phrases of all sorts.

I don't know what relevance it might have, but 之 is specifically an object pronoun (and 的 was never a pronoun of any sort, was it?).

[k1234567890y]

You are right, it is likely that 的 has never been a pronoun, but I think it might lost the meaning as a pronoun in some intermediate stages, maybe I should clarify my words

Also some other languages do also use pronouns directly for possessions:

https://wals.info/chapter/24

The Tiwi example is like the following:


jərəkəpai ŋara tuwaɹa
crocodile he tail(interlinear)
"crocodile's tail

[Xwtek]

How do you change the genitive construction so this construction.

warrior-SG.GEN axe-SG.DIR

Turns into:

warrior-SG.GEN 3SG.SG.GEN axe-SG.DIR

I guess this:
https://en.wikipedia.org/wiki/His_genitive

Also, the Classical Chinese possessor marker 之 was actually a 3rd person pronoun, and is the precedent of modern Chinese possessive postposition 的, but the meaning of personal pronoun has lost in modern Chinese.

No, the problem is not how to insert a possessive pronoun, but is it still realistic if I retain the case of original noun

[malloc]

Furthermore the phonology in my conlang seems remarkably inefficient in terms of feature usage. There are only twelve or so phonemes, making it comparable to Hawai'ian or Cheyenne in scale. Yet the language needs ten features to distinguish all these phonemes and some of those features only distinguish one phoneme. It seems rather extravagant and contrary to the minimalist principles of the language to include the feature [±STRIDENT] for only one phoneme, after all. There are ways to distinguish the phonemes with fewer features, but they require some heterodox analyses and don't capture important phonetic properties.

[Xwtek]

Furthermore the phonology in my conlang seems remarkably inefficient in terms of feature usage. There are only twelve or so phonemes, making it comparable to Hawai'ian or Cheyenne in scale. Yet the language needs ten features to distinguish all these phonemes and some of those features only distinguish one phoneme. It seems rather extravagant and contrary to the minimalist principles of the language to include the feature [±STRIDENT] for only one phoneme, after all. There are ways to distinguish the phonemes with fewer features, but they require some heterodox analyses and don't capture important phonetic properties.

Can you post the phonology and the alternations here?

[malloc]

Orthodox analysis:

Code: Select all

			LABIAL	CORONAL	DORSAL
					-BACK	+BACK
-SON	-STRID		p	t		k
	+STRID			s
+SON	+NASAL		m	n
	-NASAL	-VOC		l
		+VOC	ʋ		j

	-BACK	+BACK	+ROUND
+HIGH	i		u
-HIGH	e		o
+LOW		a

Heterodox analysis:

Code: Select all

			LABIAL	CORONAL	DORSAL
-SON	-CONT		p	t	k
	+CONT			s
+SON	-CONT		m	n
	+CONT	-VOC		l
		+VOC	ʋ	j

	CORONAL	DORSAL	LABIAL
+HIGH	i		u
-HIGH	e	a	o

The heterodox analysis differs notably from how I generally see these sounds described, especially in treating nasality as something added redundantly to sonorant occlusives and ascribing place to vowels. But it does require fewer features and explains some otherwise puzzling patterns in the language.

[akam chinjir]

The heterodox analysis differs notably from how I generally see these sounds described, especially in treating nasality as something added redundantly to sonorant occlusives and ascribing place to vowels. But it does require fewer features and explains some otherwise puzzling patterns in the language.

I was doing some reading recently about place features on vowels and consonants. I'll copy two tables I found in Jaye Padgett, "Consonant-vowel place feature interactions" (The Blackwell Companion to Phonology, 2011). (I'm going to use square-root signs instead of checkmarks because of font/spacing issues.)

First, from Elizabeth Sagey, "The representation of features and relations in nonlinear phonology" (1986):

Code: Select all

                 p  pʲ pʷ t̪  t̪ʲ t̪ʷ k  kʲ kʷ i  y  a  o  u
[LABIAL]         √  √  √        √        √  √  √     √  √
  [round]        -  -  +        +        +  -  +     +  +

[CORONAL]                 √  √  √ 
  [anterior]              +  +  +
  [distributed]           +  +  +

[DORSAL]            √        √     √  √  √  √  √  √  √  √
  [high]            +        +     +  +  +  +  +  -  -  +
  [low]             -        -     -  -  -  -  -  +  -  -
  [back]            -        -     +  -  +  -  -  +  +  +

(Interesting that u gets the same place features as kʷ and i as pʲ.)

Second, from G.N. Clements, "Place of articulation in consonants and vowels: A unified theory" (1991):

Code: Select all

                      p  pʲ pʷ t̪  t̪ʲ t̪ʷ k  kʲ kʷ ħ  i  y  a  o  u
C-Place [labial]      √  √  √
        [coronal]              √  √  √
        [dorsal]                        √  √  √
        [pharyngeal]                             √

V-Place [labial]            √        √        √        √     √  √
        [coronal]        √        √        √        √  √
        [dorsal]                                          √  √  √
        [pharyngeal]                                      √

[bradrn]

Orthodox analysis:

Code: Select all

			LABIAL	CORONAL	DORSAL
					-BACK	+BACK
-SON	-STRID		p	t		k
	+STRID			s
+SON	+NASAL		m	n
	-NASAL	-VOC		l
		+VOC	ʋ		j

	-BACK	+BACK	+ROUND
+HIGH	i		u
-HIGH	e		o
+LOW		a

Heterodox analysis:

Code: Select all

			LABIAL	CORONAL	DORSAL
-SON	-CONT		p	t	k
	+CONT			s
+SON	-CONT		m	n
	+CONT	-VOC		l
		+VOC	ʋ	j

	CORONAL	DORSAL	LABIAL
+HIGH	i		u
-HIGH	e	a	o

The heterodox analysis differs notably from how I generally see these sounds described, especially in treating nasality as something added redundantly to sonorant occlusives and ascribing place to vowels. But it does require fewer features and explains some otherwise puzzling patterns in the language.

Somewhat unrelated question: why is it important to have a featural analysis here? What can this do that simply listing by POA/MOA cannot?

(Then again, I know almost nothing about featural analysis, so I suspect there's some obvious answer here that I don't know...)

[akam chinjir]

Somewhat unrelated question: why is it important to have a featural analysis here? What can this do that simply listing by POA/MOA cannot?

(Then again, I know almost nothing about featural analysis, so I suspect there's some obvious answer here that I don't know...)

It can be good to have a handle on this sort of stuff if you want phonological interactions between vowels and consonants. (In my case, the thing was epenthetic consonants. In some contexts, I wanted k to show up, which isn't especially common, cross-linguistically, and now my semi-official justification is that this happens before u and a, both of which are [+back], like k.)

[bradrn]

Somewhat unrelated question: why is it important to have a featural analysis here? What can this do that simply listing by POA/MOA cannot?

(Then again, I know almost nothing about featural analysis, so I suspect there's some obvious answer here that I don't know...)

It can be good to have a handle on this sort of stuff if you want phonological interactions between vowels and consonants. (In my case, the thing was epenthetic consonants. In some contexts, I wanted k to show up, which isn't especially common, cross-linguistically, and now my semi-official justification is that this happens before u and a, both of which are [+back], like k.)

Interesting! Do you know any resources where I can learn about this topic?

[akam chinjir]

I'd guess your best bet is a phonology textbook. Maybe Hayes, Introductory Phonology? I've mostly picked things up here and there over the years, so I don't have a strong opinion about textbooks. I did just read through Bale and Reiss, Phonology: A Formal Introduction, which I got a lot out of; it's very opinionated in ways you might find infuriating, though.

With some understanding of the basics, there are a few handbooks and such that include lots of helpful articles. I've recently read a bunch from The Blackwell Companion to Phonology, which is great if you can get access to it.

[bradrn]

I'd guess your best bet is a phonology textbook. Maybe Hayes, Introductory Phonology? I've mostly picked things up here and there over the years, so I don't have a strong opinion about textbooks. I did just read through Bale and Reiss, Phonology: A Formal Introduction, which I got a lot out of; it's very opinionated in ways you might find infuriating, though.

With some understanding of the basics, there are a few handbooks and such that include lots of helpful articles. I've recently read a bunch from The Blackwell Companion to Phonology, which is great if you can get access to it.

Thank you!

[akam chinjir]

If I've got progressive nasal harmony, and among the consonants that this can affect are ʋ and w, and the language also has m, how likely is it that ʋ̃ will be/remain phonologically/phonetically distinct?

[Xwtek]

If I've got progressive nasal harmony, and among the consonants that this can affect are ʋ and w, and the language also has m, how likely is it that ʋ̃ will be/remain phonologically/phonetically distinct?

If you have /ŋ/ or /ŋʷ/, then it's likely that it will be phonologically distinct. For former, you can say /ɰ/ > /w/, except when nasalized, because /ɰ̃/ > /ŋ/. The latter is even easier.

[akam chinjir]

But I've got /ʋ/, not /ɰ/. (I do have ŋʷ though.)

[Curlyjimsam]

No, the problem is not how to insert a possessive pronoun, but is it still realistic if I retain the case of original noun

I can't see it being very likely at all if all genitive marking on nouns remains unchanged. Perhaps if the genitive endings for a sizeable number of nouns undergo a lot of phonological reduction or merge with other endings then the possessive pronoun strategy could come about as a way to save the unambiguous marking of possessive function.

[Xwtek]

But I've got /ʋ/, not /ɰ/. (I do have ŋʷ though.)

I mean /*ɰ/ turns to current /w/, so it appears that /w/ turns into /ŋ/ while diachronically, it's actually /*ɰ/ turns into /ŋ/ when nasalized, but /w/ elsewhere.

[akam chinjir]

I mean /*ɰ/ turns to current /w/, so it appears that /w/ turns into /ŋ/ while diachronically, it's actually /*ɰ/ turns into /ŋ/ when nasalized, but /w/ elsewhere.

Okay, I can see how that would work, but I actually like [w̃] and would prefer to keep it, my worry is about [ʋ̃].